``` Joint Technical Support Document: Final Rulemaking for 2017-2025 Light-Duty Vehicle Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards &EPA United States Environmental Protection Agency ***** NHTSA www.nhtsa.gov ``````------- Joint Technical Support Document: Final Rulemaking for 2017-2025 Light-Duty Vehicle Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards Assessment and Standards Division Office of Transportation and Air Quality U.S. Environmental Protection Agency and National Highway Traffic Safety Administration U.S. Department of Transportation &EPA United States Environmental Protection Agency ***** NHTSA www.nhtsa.gov EPA-420-R-12-901 August 2012 ``````------- Contents Executive Summary i Chapter 1: The Baseline and Reference Vehicle Fleets 1-1 1.1 Why do the agencies establish baseline and reference vehicle fleets? 1-1 1.2 The 2008 and 2010 based vehicle fleet projections 1-2 1.2.1 Why did the agencies develop two fleet projections for the final rule? 1-2 1.3 The 2008 Based Fleet Projection 1-4 1.3.1 On what data is the MY2008 baseline vehicle fleet based? 1-4 1.3.2 The MY 2008 Based MY 2017-2025 Reference Fleet 1-13 1.3.3 What are the sales volumes and characteristics of the MY 2008 based reference fleet? 1-25 1 4 The 2010 MY Based Fleet 1-31 1.4.1 On what data is the MY 2010 baseline vehicle fleet based? 1-31 1.4.2 The MY 2010 Based MY 2017-2025 Reference Fleet 1-40 1.4.3 What are the sales volumes and characteristics of the MY 2010 based reference fleet? 1-48 1.5 What are the differences in the sales volumes and characteristics of the MY 2008 based and the MY 2010 based reference fleets? 1-54 Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting, and How Were They Developed? 2-1 2 1 Why are standards attribute-based and defined by a mathematical function? 2-1 22 What attribute are the agencies adopting, and why? 2-2 2.3 What mathematical functions have the agencies previously used, and why? 2-7 2.3.1 NHTSA in MY 2008 and MY 2011 CAFE (constrained logistic) 2-7 232 MYs 2012-2016 Light Duty GHG/CAFE (constrained/piecewise linear) 2-8 2.3.3 How have the agencies defined the mathematical functions for the MYs 2017-2025 standards, and why? 2-8 2.4 What did the agencies propose for the MYs 2017-2025 curves? 2-9 2.4.1 What concerns were the agencies looking to address that led them to change from the approach used for the MYs 2012-2016 curves? 2-10 242 What methodologies and data did the agencies consider in developing the 2017-2025 curves presented in the proposal? 2-12 2.5 Once the agencies determined the appropriate slope for the sloped part, how did the agencies determine the rest of the mathematical function? 2-49 ``````------- Contents 2.5.1 Cutpoints for Passenger Car curve 2-49 2.5.2 Cutpoints for Light Truck curve 2-51 253 Once the agencies determined the complete mathematical function shape, how did the agencies adjust the curves to develop the proposed standards and regulatory alternatives? 2-54 2.6 What does the agencies' updated analysis indicate? 2-56 Chapter 3: Technologies Considered in the Agencies' Analysis 3-2 3.1 What Technologies did the agencies consider for the final 2017-2025 standards?3-3 3.2 How did the agencies determine the costs of each of these technologies? 3-8 3.2.1 Direct Costs 3-8 322 Indirect Costs 3-13 3.2.3 Cost reduction through manufacturer learning 3-23 324 Costs Updated to 2010 Dollars 3-28 3.3 How did the agencies determine effectiveness of each of these technologies? 3-29 3.3.1 Vehicle simulation modeling 3-29 3.3.2 Lumped parameter Modeling 3-69 3.4 What cost and effectiveness estimates have the agencies used for each technology? 3-76 3.4.1 Engine technologies 3-77 3.4.2 Transmission Technologies 3-101 3.4.3 Vehicle electrification and hybrid electric vehicle technologies 3-112 3.4.4 Hardware costs for charging grid-connected vehicles 3-203 3.4.5 Other Technologies Assessed that Reduce CO2 and Improve Fuel Economy 3-208 3.5 How did the agencies consider real-world limits when defining the rate at which technologies can be deployed? 3-249 3.5.1 Refresh and redesign schedules 3-249 3.6 How are the technologies applied in the agencies' respective models? 3-258 3.7 Maintenance and Repair Costs Associated with New Technologies 3-259 3.7.1 Maintenance Costs 3-260 372 Repair Costs 3-264 Chapter 4: Economic and Other Assumptions Used in the Agencies' Analysis 4-1 ``````------- Contents 4.1 How the Agencies use the economic and other assumptions in their analyses 4-1 4.2 What assumptions do the agencies use in the impact analyses? 4-2 4.2.1 The on-road fuel economy "gap" 4-2 422 Fuel prices and the value of saving fuel 4-7 4.2.3 Vehicle Lifetimes and Survival Rates 4-9 424 VMT 4-12 4.2.5 Accounting for the fuel economy rebound effect 4-18 4.2.6 Benefits from additional driving 4-26 4.2.7 Added costs from increased vehicle use 4-26 4.2.8 Petroleum and energy security impacts 4-28 4.2.9 Air pollutant emissions 4-39 4.2.10 Reductions in emissions of greenhouse gases 4-48 4.2.11 Benefits due to reduced refueling time 4-53 4.2.12 Discounting future benefits and costs 4-54 4.2.13 Additional Costs of Vehicle Ownership 4-54 Chapter 5: Air Conditioning, Off-Cycle Credits, and Other Flexibilities 5-1 5.1 Air conditioning technologies and credits 5-1 5.1.1 Overview 5-1 5.1.2 Air Conditioner Leakage 5-3 5.1.3 COi Emissions and Fuel Consumption due to Air Conditioners 5-22 5.1.4 Air Conditioner System Costs 5-58 52 Off-Cycle Technologies and Credits 5-62 5.2.1 Reducing or Offsetting Electrical Loads 5-64 522 Waste Heat Recovery 5-65 523 High Efficiency Exterior Lights 5-69 524 Solar Panels 5-73 5.2.5 Definitions for Electrical Load Offsetting and Reduction Technologies....5-81 5.2.6 Active Aerodynamic Improvements 5-81 5.2.7 Definition for Active Aerodynamic Improvements 5-84 5.2.8 Advanced Load Reductions 5-84 5.2.9 Thermal (and Solar) Control Technologies 5-101 iii ``````------- Contents 5.2.10 Glazing 5-101 5.2.11 Active Seat Ventilation 5-106 5.2.12 Solar Reflective Paint 5-107 5213 Passive and Active Cabin Ventilation 5-108 5.2.14 Summary of Thermal (and Solar) Control Credits 5-109 5.2.15 Definitions for Solar Control Credit Technologies 5-110 5216 Summary of Credits 5-110 5.3 Full-Size Pickup Truck Credits 5-111 5.3.1 Full-Size Pick-up Truck Definition 5-112 5.3.2 Hybrid Pickup Truck Technology 5-113 533 Mild and Strong Hybrid Pickup Truck Definitions 5-113 5.3.4 Pickup Truck Performance Thresholds for Advanced Technology Credits .5- 118 IV ``````------- Executive Summary Executive Summary The Environmental Protection Agency (EPA) and the National Highway Traffic Safety Administration (NHTSA) are issuing a joint final rule to establish new standards for light-duty highway vehicles that will reduce greenhouse gas emissions and improve fuel economy. This joint final rulemaking is consistent with the Presidential Memorandum issued by President Obama on May 21, 2010, requesting that NHTSA and EPA develop through notice and comment rulemaking a coordinated National Program to reduce greenhouse gas emissions and improve the fuel economy of light-duty vehicles for model years 2017-2025. This final rule, consistent with the President's request, responds to the country's critical need to address global climate change and to reduce oil consumption. EPA is regulating greenhouse gas emissions standards under the Clean Air Act, and NHTSA is regulating Corporate Average Fuel Economy standards under the Energy Policy and Conservation Act, as amended. These standards apply to passenger cars, light-duty trucks, and medium-duty passenger vehicles, covering model years 2017 through 2025. They require these vehicles to meet an estimated combined average emissions level of 163 grams of CO2 per mile in MY 2025 under EPA's GHG program, and 49.6 mpg in MY 2025 under NHTSA's CAFE program and represent a harmonized and consistent national program (National Program). These standards are designed such that compliance can be achieved with a single national vehicle fleet whose emissions and fuel economy performance improves each year from MY2017 to 2025. This document describes the supporting technical analysis for areas of these joint rules which are consistent between the two agencies. NHTSA and EPA have coordinated closely to create a nationwide joint fuel economy and GHG program based on consistent compliance structures and technical assumptions. To the extent permitted under each Agency's statutes, NHTSA and EPA have incorporated the same compliance flexibilities, such as averaging, banking, and trading of credits, off-cycle credits, and the same testing protocol for determining the agencies' respective fleet-wide average final standards. In addition, the agencies have worked together to create a common baseline fleet and to harmonize most of the costs and benefit inputs used in the agencies' respective modeling processes for this joint final rule. Chapter 1 of this joint TSD provides an explanation of the agencies' methodology used to develop the baseline and reference case vehicle fleets, including the technology composition of these fleets, and how the agencies projected vehicle sales into the future. One of the fundamental features of this technical analysis is the development of these fleets, which are used by both agencies in their respective models. In order to determine technology costs associated with this joint rulemaking, it is necessary to consider the vehicle fleet absent a rulemaking as a "business as usual" comparison. In past CAFE rulemakings, NHTSA has used confidential product plans submitted by vehicle manufacturers to develop the reference case fleet. In responding to comments from these previous rulemakings that the agencies make these fleets available for public review, the agencies created a new methodology for creating baseline and reference fleets using data, the vast majority of which is publicly available. ``````------- Executive Summary Chapter 2 of this document discusses how NHTSA and EPA developed the mathematical functions which provide the bases for the final car and truck standards. NHTSA and EPA worked together closely to develop regulatory approaches that are fundamentally the same, and have chosen to use an attribute-based program structure based on the footprint attribute, similar to the mathematical functions used in the MYs 2012-2016 rule. The agencies revisited other attributes as candidates for the standard functions, but concluded that footprint remains the best option for balancing the numerous technical and social factors. However, the agencies did adjust the shape of the truck footprint curve, in comparison to the MYs 2012-2016 rule. The agencies also modified the way the car and truck curves change from year to year compared to the MYs 2012-2016 rule. In determining the shape of the footprint curve, the agencies considered factors such as the magnitudes of CO2 reduction and fuel savings, how much that shape may incentivize manufacturers to comply in a manner which circumvents the overall goals of the joint program, whether the standards' stringencies are technically attainable, the utility of vehicles, and the mathematical flexibilities inherent to the statistical fitting of such a function. Chapter 3 contains a detailed analysis of NHTSA and EPA's technology assumptions on which the final regulations were based. Because the majority of technologies that reduce GHG emissions and improve fuel economy are identical, it was crucial that NHTSA and EPA use common assumptions for values pertaining to technology availability, cost, and effectiveness. The agencies collaborated closely in determining which technologies would be considered in the rulemaking, how much these technologies would cost the manufacturers (directly) in the time frame of the final rule, how these costs will be adjusted for learning as well as for indirect cost multipliers, and how effective the technologies are at accomplishing the goals of improving fuel efficiency and GHG emissions. Chapter 4 of this document provides a full description and analysis of the economic factors considered in this joint final rule. EPA and NHTSA harmonized many inputs capturing economic and social factors, such as the discount rates, fuel prices, social costs of carbon, the magnitude of the rebound effect, the value of refueling time, and the social cost of importing oil and fuel. Chapter 5 of this joint TSD discusses adjustments and credits to reflect technologies that improve air conditioner efficiency, that improve efficiency under other off-cycle driving conditions, and that reduce leakage of air conditioner refrigerants that contribute to global warming. The air conditioner credits are similar to the MYs 2012-2016 rule, with two notable exceptions: NHTSA is allowing A/C efficiency improvements to help come into compliance with fuel economy standards, and a new air conditioner test procedure is introduced to help capture efficiency credits. NHTSA is now also allowing off-cycle improvements to help manufacturers come into compliance with fuel economy standards. A list of some technologies and their credits and a streamlined methodology is provided by the agencies to help simplify the credit generating process. Chapter 5 also discusses adjustments to encourage "game changing" technologies (such as hybridized powertrains) for full-size pickup trucks. ``````------- The Baseline and Reference Vehicle Fleets Chapter 1: The Baseline and Reference Vehicle Fleets The passenger cars and light trucks sold currently in the United States, and those that are anticipated to be sold in the MYs 2017-2025 timeframe, are highly varied and satisfy a wide range of consumer needs. From two-seater miniature cars to 11-seater passenger vans to large extended cab pickup trucks, American consumers have a great number of vehicle options to accommodate their needs and preferences. Recent volatility in oil prices and the state of the economy have demonstrated that consumer demand and choice of vehicles within this wide range can be sensitive to these factors. Although it is impossible to precisely predict the future, the agencies need to characterize and quantify the future fleet in order to assess the impacts of rules that would affect that future fleet. The agencies have examined various publicly-available sources, and then used inputs from those sources in a series of models to project the composition of baseline and reference fleets for purposes of this analysis. This chapter describes this process, and the characteristics of each of the two baseline and reference fleets. The agencies have made every effort to make this analysis transparent and duplicable. Because both the input and output sheets from our modeling are public,1 stakeholders can verify and checkNHTSA's and EPA's modeling results, and perform their own analyses with these datasets. 1.1 Why do the agencies establish baseline and reference vehicle fleets? In order to calculate the impacts of the final GHG and CAFE standards, it is necessary to estimate the composition of the future vehicle fleet absent the new standards. EPA and NHTSA have developed a baseline/reference fleet in two parts. The first step was to develop a "baseline" fleet. The agencies create a baseline fleet in order to track the volumes and types of fuel economy-improving and CCVreducing technologies that are already present in the existing vehicle fleet. Creating a baseline fleet helps to keep, to some extent, the agencies' models from adding technologies to vehicles that already have these technologies, which would result in "double counting" of technologies' costs and benefits. The second step was to project the baseline fleet sales into MYs 2017-2025. This is called the "reference" fleet, and it represents the fleet volumes (but, until later steps, not additional levels of technology) that the agencies believe would exist in MYs 2017-2025 absent any change due to regulation in 2017-2025. After determining the reference fleet, a third step is needed to account for technologies (and corresponding increases in cost and reductions in fuel consumption and CC>2 emissions) that could be added to the baseline technology vehicles in the future, taking into account previously-promulgated standards, and assuming MY 2016 standards apply at the same levels through MY 2025. This step uses the OMEGA and CAFE models to add technologies to vehicles in each of the baseline market forecasts such that each manufacturer's car and truck CAFE and average CO2 levels reflect MY 2016 standards. The models' output, the 1-1 ``````------- The Baseline and Reference Vehicle Fleets "reference case", is the light-duty fleet estimated to exist in MYs 2017-2025 without new GHG/CAFE standards. All of the agencies' estimates of emission reductions/fuel economy improvements, costs, and societal impacts for purposes of this final rulemaking (FRM) are developed in relation to the agencies' reference cases. This chapter describes the first two steps of the development of the baseline and reference fleets. The third step of technology addition is developed separately by each agency as the outputs of the OMEGA and CAFE models (see Chapter 3 of the TSD for an explanation of how the models apply technologies to vehicles in order to evaluate potential paths to compliance). 1.2 The 2008 and 2010 based vehicle fleet projections 1.2.1 Why did the agencies develop two fleet projections for the final rule? Although much of the discussion in this and following sections describes the methodology for creating a single baseline and reference fleet, for this final rule the agencies actually developed two baseline and reference fleets. In the NPRM, the agencies used 2008 MY CAFE certification data to establish the "2008-based fleet projection."a The agencies noted that MY 2009 CAFE certification data was not likely to be representative since it was so dramatically influenced by the economic recession (Joint Draft TSD section 1.2.1). The agencies further noted that MY 2010 CAFE certification data might be available for use in the final rulemaking for purposes of developing a baseline fleet (id.). The agencies also stated that a copy of the MY 2010 CAFE certification data would be put in the public docket if it became available during the comment period. The MY 2010 data was reported by the manufacturers throughout calendar year 2011 as the final sales figures were compiled and submitted to the EPA database. Due to the lateness of the CAFE data submissions'3, it was not possible to submit the new 2010 data into the docket during the public comment period. As explained below, however, consistent with the agencies' expectations at proposal, and with the agencies' standard practice of updating relevant information as practicable between proposals and final rules, the agencies are using these data in one of the two fleet-based projections we are using to estimate the impacts of the final rules. For analysis supporting the NPRM, the agencies developed a forecast of the light vehicle market through MY 2025 based on (a) the vehicle models in the MY 2008 CAFE certification data, (b) the AEO2011 interim projection of future fleet sales volumes, and (c) the future fleet forecast conducted by CSM in 2009. In the proposal, the agencies stated we planned to use MY 2010 CAFE certification data, if available, for analysis supporting the final rule (Joint Draft TSD, p. 1-2). The agencies also indicated our intention to, for analysis a 2008 based fleet projection is a new term that is the same as the reference fleet. The term is added to clarify when we are using the 2008 baseline and reference fleet vs. the 2010 baseline and reference fleet. b Partly due to the earthquake and tsunami in Japan and the significant impact this had on their facilities, some manufacturers requested and were granted an extension on the deadline to submit their CAFE data. 1-2 ``````------- The Baseline and Reference Vehicle Fleets supporting the final rule, use the most recent version of EIA's AEO, and a market forecast updated relative to that purchased from CSM (Joint Draft TSD section 1.3.5). For this final rulemaking, the agencies have analyzed the costs and benefits of the standards using two different forecasts of the light vehicle fleet through MY 2025. The agencies have concluded that the significant uncertainty associated with forecasting sales volumes, vehicle technologies, fuel prices, consumer demand, and so forth out to MY 2025, makes it reasonable and appropriate to evaluate the impacts of the final CAFE and GHG standards using two baselines. One market forecast, similar to the one used for the NPRM, uses corrected data regarding the MY2008 fleet, information from AEO 2011, and information purchased from CSM. The agencies received comments regarding the market forecast used in the NPRM suggesting that updates in several respects could be helpful to the agencies' analysis of final standards; given those comments and since the agencies were already planning to produce an updated market forecast, the final rule also contains another market forecast using MY 2010 CAFE certification data, information from AEO 2012, and information purchased from LMC Automotive (formerly JD Power Forecasting). The two market forecasts contain certain differences, although as will be discussed below, the differences are not significant enough to change the agencies' decision as to the structure and stringency of the final standards. For example, MY 2008 certification data represents the most recent model year for which the industry's offerings were not strongly affected by the subsequent economic recession, which may make it reasonable to use if we believe that the future vehicle model offerings are more likely to be reflective of pre-recession offerings than models produced after MY 2008 (e.g., in MY 2010). Also, the MY 2010-based fleet projection employs a future fleet forecast provided by LMC Automotive, which is more current than the projection provided by CSM in 2009. However, the CSM forecast, utilized for the MY2008-based fleet projection, was influenced by the recession, particularly in predicting major declines in market share for some manufacturers (e.g., Chrysler) which the agencies do not believe are reasonably reflective of future trends. The MY 2010 based fleet projection, which is used in EPA's alternative analysis and in NHTSA's co-analysis, employs a future fleet forecast provided by LMC Automotive, which is more current than the projection provided by CSM in 2009, and which reflects the post-proposal MY 2010 CAFE certification data. However, this MY 2010 CAFE data also shows strong effects of the economic recession. For example, industry-wide sales were down by 20% compared to pre-recession MY 2008 levels. For some companies like Chrysler, Mitsubishi, and Subaru, sales were down by 30-40% from MY 2008 levels.0 For BMW, General Motors, Jaguar/Land Rover, Porsche, and Suzuki, sales were down more than 40% from MY 2008 levels. Employing the MY 2008 vehicle data avoids using these baseline ! These figure are arrived at using Table 1-17 and Table 1-39. 1-3 ``````------- The Baseline and Reference Vehicle Fleets market shifts when projecting the future fleet. On the other hand, it also perpetuates vehicle brands and models (and thus, their outdated fuel economy levels and engineering characteristics) that have since been discontinued. The MY 2010 CAFE certification data accounts for the phase-out of some brands (e.g., Saab, Pontiac, Hummer)6 and the introduction of some technologies (e.g., Ford's Ecoboost engine), which may be more reflective of the future fleet in this respect. Thus, given the volume of information that goes into creating a baseline forecast and given the significant uncertainty in any projection out to MY 2025, the agencies think that a reasonable way to illustrate the possible impacts of that uncertainty for purposes of this rulemaking is the approach taken here of analyzing the effects of the final standards under both the MY 2008-based baseline and the MY 2010-based baseline. The agencies' analyses are presented in our respective RIAs and preamble sections. 1.3 The 2008 Based Fleet Projection Differences between the 2008 MY based fleet used in the final rule compared to that used in the NPRM include minor corrections to some of the vehicle footprint data, and minor corrections to technology "overrides" and technology class assignments used in DOT's modeling system. A discussion of the changes is in the section below along with a thorough description of how the projection was created. 1.3.1 On what data is the MY2008 baseline vehicle fleet based? As part of the CAFE program, EPA measures vehicle CC>2 emissions and converts them to mpg, and generates and maintains the federal fuel economy database. See 49 U.S.C 32904 and 40 CFR Part 600. Most of the information about the vehicles that make up the 2008 fleet was gathered from EPA's emission certification and fuel economy database, most of which is available to the public. These data (by individual vehicle model produced in MY 2008) include: vehicle production volume, fuel economy rating for CAFE certification (i.e., on the 2-cycle city-highway test), carbon dioxide emissions (equivalent to fuel economy rating for CAFE certification), fuel type (gasoline, diesel, and/or alternative fuel), number of engine cylinders, displacement, valves per cylinder, engine cycle, transmission type, drive (rear-wheel, all-wheel, etc.), hybrid type (if applicable), and engine aspiration (naturally- aspirated, turbocharged, etc.). In addition to this information about each vehicle model produced in MY 2008, the agencies need additional information about the fuel economy- improving/CO2-reducing technologies already on those vehicle models in order to assess how much and which technologies to apply to determine a path toward future compliance. However, EPA's certification database does not include a detailed description of the types of technologies considered in this FRM because this level of information was not reported in e Based on our review of the CAFE certification data, the MY 2010-based fleet contains no Saabs, and compared to the MY 2008-based fleet, about 90% fewer Hummers and about 75% fewer Pontiacs. 1-4 ``````------- The Baseline and Reference Vehicle Fleets MY 2008 for emission certification or fuel economy testing. Thus, the agencies augmented this description with publicly-available data which includes more complete technology descriptions from Ward's Automotive Group.f'g The agencies also required information about the footprints of MY 2008 vehicles in order to generate potential target footprint curves (as discussed in Chapter 2 of the TSD). In a few instances when relevant vehicle information (such as vehicle track width for footprint) was not available from these two sources, the agencies obtained this information principally from publicly-accessible internet sites such as Motortrend.com or Edmunds.com, and occasionally from other sources (such as articles about specific vehicles revealed from internet search engine research). '' Between the NPRM and the final rule, the agencies found discrepancies in footprint values for a number of vehicles in the MY 2008 CAFE certification data. Specifically, contractors to DOT employed to develop a market share model for incorporation into the CAFE model noted that out of 1,302 vehicles in the MY 2008-based input file used in the agencies' NPRM analysis, in 554 cases, the wheelbase value in the CAFE certification data did not match wheelbase data from Ward's Automotive that the contractor had obtained separately. While wheelbase is not a direct input to the models used in developing the standards, it is a component of footprint, which is a key input in the modeling process. Of the reported differences, 287 (51.8%) were less than or equal to 0.1 inch, and 115 (20.8%) were greater than 0.1 inch but less than or equal to 0.5 inch. The former set of differences is most likely attributable to differences in the number of significant digits in the reported raw data. The latter set of differences may also be due to reporting differences or actual measurement differences, but would not have a significant impact on the computed footprint value, all other things being equal. These differences were not considered further. Of the remaining differences, 14 (2.5%) were greater than 0.5 inch but less than 1 inch. Most significantly, 138 (24.9%) of the differences were greater than 1 inch, ranging in value from 1.1 inch to 23.8 inches. To verify these findings, the Ward's data used by the contractor on wheelbase for the 152 vehicles with a discrepancy greater than 0.5 inches were compared to wheelbase data from Edmunds, cars.com, Motor Trend, and product plans where available, and values reflecting the agencies' best judgment about actual average values was selected. Footprint for the 152 vehicles was thus recalculated based on corrected wheelbase. In the process of validating the wheelbase data, the agencies noted that there were many f WardsAuto.com: Used as a source for engine specifications shown in Table 1-1. 8 Note that WardsAuto.com, where this information was obtained, is a fee-based service, but all information is public to subscribers. Motortrend.com and Edmunds.com: Used as a source for footprint and vehicle weight data. 1 Motortrend.com and Edmunds.com are free, no-fee internet sites. 1-5 ``````------- The Baseline and Reference Vehicle Fleets discrepancies in the track width values, which the agencies also corrected in the calculation of the corrected footprints. The affected vehicles included those of the following manufacturers: Chrysler -4(2 large SUV, 2 small SUV) Daimler -19(1 compact auto, 15 large auto, 1 midsize auto, 2 subcompact auto) Ford - 4 (2 large pickup, 2 small pickup) General Motors - 29 (18 compact auto, 7 midsize auto, 4 subcompact auto) Honda - 17 (3 compact auto, 2 large SUV, 8 midsize auto, 1 small pickup, 3 subcompact auto) Hyundai - 2 (2 subcompact auto) Kia - 8 (2 compact auto, 4 midsize auto, 2 subcompact auto) Mazda - 7 (4 midsize SUV, 2 small pickup, 1 subcompact auto) Nissan - 11 (4 compact auto, 6 large auto, 1 minivan) Subaru - 15 (6 midsize auto, 9 midsize SUV) Tata - 2 (2 midsize auto) Toyota - 29 (3 compact auto, 6 large pickup, 16 large auto, 4 midsize auto) Volkswagen - 5 (4 large auto, 1 midsize auto) Table 1-1 shows the change from the NPRM to the FRM in the average footprint for all vehicles, cars, and trucks. The average change in footprint was very small, although quite a few vehicles' footprints were updated. Table 1-1 2008 MY Footprint changes (Final Rule Values - NPRM Values) Average Footprint of all Vehicles -0.1 Average Footprint Cars -0.2 Average Footprint Trucks 0 The baseline vehicle fleet for the analysis informing these final rules is the same except for the footprint changes to the baseline vehicle fleet used in the MYs 2012-2016 rulemaking, and like that baseline, is comprised of publicly-available data to the largest extent possible. Some of the technology data included in the MYs 2012-2016 analysis' baseline fleet was based on confidential product plan information about MY 2008 vehicles, specifically, data about which vehicles already have low friction lubricants, electric power steering, improved accessories, and low rolling resistance tires applied, the agencies no longer consider that information as needing to be withheld, because by now all MY 2008 vehicle models are already in the on-road fleet. As a result, the agencies are able to make public the exact baseline used in this rulemaking analysis. As explained in the MYs 2012-2016 TSD, creating the 2008 baseline fleet Excel file was an extremely labor-intensive process. EPA in consultation with NHTSA first considered using EPA's CAFE certification data, which contains most of the required information. However, since the deadline for manufacturers to report this data did not allow enough time, 1-6 ``````------- The Baseline and Reference Vehicle Fleets in the MYs 2012-2016 rulemaking, for early modeling review, the agencies began to create the baseline fleet file using an alternative data source. The agencies ultimately relied on a combination of EPA's vehicle emissions certification data, data from a paid subscription to Ward's Automotive Group, and CAFE certification data. EPA's vehicle emissions certification data contains much of the information required for creating a baseline fleet file, but it lacked the production volumes that are necessary for the OMEGA and Volpe models, and also contains some vehicle models that manufacturers certified but did not produce in MY 2008. The data from Ward's contained production volumes (which were not ultimately used, because they did not have volumes for individual vehicles down to the resolution of the specific engine and transmission level) and vehicle specifications, and eliminated extraneous vehicles. The EPA vehicle emissions certification dataset came in two parts, an engine file and a vehicle file, which the agencies combined into one spreadsheet using their common index. The more-specific Ward's data also came in two parts, an engine file and a vehicle file, and also required mapping, which was more difficult than combining the EPA vehicle emissions certification dataset files because there was no common index between the Ward's files. A new index was implanted in the engine file and a search equation in the vehicle file, which identified most of the vehicle and engine combinations. Each vehicle and engine combination was reviewed and corrections were made manually when the search routine failed to give the correct engine and vehicle combination. The combined Ward's data was then mapped to the EPA vehicle emissions certification data by creating a new index in the combined Ward's data and using the same process that was used to combine the Ward's engine and vehicle files. In the next step, CAFE certification data had to be merged in order to fill out the needed production volumes. NHTSA and EPA reviewed the CAFE certification data for MY 2008 as it became available in the MYs 2012-2016 rulemaking. The CAFE certification set could have been used with the Ward's data without the EPA vehicle emission certification data set, but was instead appended to the combined Ward's and EPA vehicle emission certification dataset. That combined dataset was then mapped into the CAFE dataset using the same Excel mapping technique described above. Finally EPA and NHTSA obtained the remaining attribute and technology data, such as footprint, curb weight, and others (for a complete list of data with sources see Table 1-2 below) from other sources, thus completing the baseline dataset. Another step that was done for the first time in the NPRM (and used in this FRM baseline as well) was to disaggregate the footprints of pickup trucks. In the MYs 2012-2016 rulemaking the agencies aggregated full-size pickup data in the baseline by using average values to represent all variants of a given pickup line. While full-size pickups might be offered with various combinations of cab style (e.g., regular, extended, crew) and box length (e.g., 5 l/2\ 6 l/2 , 8'), and therefore multiple footprint sizes, CAFE compliance data for MY 2008 did not contain footprint information, and therefore could not reliably be used to identify which pickup entries correspond to footprint values estimable from public or commercial sources. Therefore, the agencies used the known production levels of average values to 1-7 ``````------- The Baseline and Reference Vehicle Fleets represent all variants of a given pickup line (e.g., all variants of the F-150, or all variants of the Sierra/Silverado) in order to calculate the sales-weighted average footprint/fuel economy value for each pickup family. In retrospect, this may have affected how we fit the light truck target curve, among other things, so the agencies have since created an expanded version of the fleet to account for the variation in footprint/wheelbase for the large pickups of Chrysler, Ford, GM, Nissan and Toyota. In MY 2008, large pickups were available from Nissan with 2, Chrysler and Toyota with 3, and Ford and GM with 5 wheelbase/footprint combinations. The agencies got this footprint data from MY 2008 product plans submitted by the various manufacturers, which can be made public at this time because by now all MY 2008 vehicle models are already in production, which makes footprint data about them essentially public information. The agencies created the expanded fleet by replicating original records from a single pickup footprint model into multiple pickup models with distinct footprint values, in order to reflect the additional pickup model footprints just noted. For example, an F-150 in the MY 2008 baseline used in the MYs 2012-2016 rulemaking analysis with a footprint value of 67 square feet, is disaggregated by replicating 2 times in all respects, except with footprint values of 58, 67, and 73 square feet. Sales volumes of these pickups from the original record were distributed to each of the "58 square feet" and "73 square feet" duplicates based on the distribution of MY 2008 sales by these pickups' wheelbase/footprint, which the agencies took from product plan data submitted by the manufacturers in 2008/2009 in response to requests to support the MYs 2012-2016 rulemaking analysis. The agencies were able to distribute the sales for each of the original pickups by wheelbase/footprint by matching each of the pickups in the baseline fleet with pickups in the product plans on the basis of drive type, transmission type, and engine displacement, cylinders/configuration and HP, and then sorting and summing the sales of the matched pickups in the product plans by wheelbase/footprint. Both agencies used this fleet forecast to populate input files for the agencies' respective modeling systems. The structure of the market forecast input file used for the Volpe model is described the model documentation.2 To help readers who wish to directly examine the baseline fleet file for EPA's OMEGA model, and to provide some idea of its contents for those readers who do not, Table 1-2 shows the columns of the complete fleet file, which includes the MY 2008 baseline data that was compiled. Each column has its name, definition (description) and source. Most elements shown in Table 1-2 also appear in the market forecast input file for DOT's modeling system, which also accommodates some additional data elements discussed in the model documentation. Table 1-2 2008 MY Data, Definitions, and Sources Data Item Index Manufacturer CERT Manufacturer Name Name Plate Definition Index Used to link EPA and NHTSA baselines Common name of company that manufactured vehicle. May include more name plates than Cert Manufacturer Name. Certification name of company that manufactured vehicle Name of Division Where The Data is From Created Certification data Certification data Certification data 1-8 ``````------- The Baseline and Reference Vehicle Fleets Model Reg Class Our Class CSM Class Vehicle Type Number Vehicle Index From Sum Page Traditional Car/Truck NHTSA Defined New Car/Truck Total Production Volume Fuel Econ. (mpg) CO2 Area (sf) Fuel Fuel Type Disp (lit) Effective Cyl Actual Cylinders Valves Per Cylinder Valve Type Valve Actuation VVT VVLT Deac Fuel injection system Boost Engine Cycle Name of Vehicle EPA Fuel Economy Class Name If a car's Footprint<43 then "SubCmpctAuto" If a car's 43<=Footprint<46 then "CompactAuto" If a car's 46<=Footprint<53 then "MidSizeAuto" If a car's Footprint >=53 then "LargeAuto" If a S.U.V.'s Footprint < 43 then "SmallSuv" If a S.U.V.'s 43<=Footprint<46 then "MidSizeSuv" If a S.U. V's Footprint >=46 then "LargeSuv" If a Truck's Footprint < 50 then "SmallPickup" If a Truck's Footprint>=50 then "LargPickup" If a Van's Structure is Ladder then "CargoVan" If a Van's Structure is Unibody then "Minivan" CSM Worldwide' s class for the vehicle. Used to weight vehicles based on CSM data. Vehicle Type Number assigned to a vehicle based on its number of cylinders, valves per cylinder, and valve actuation technology Number to be used as a cross reference with the Sum Pages. Traditional Car Truck value for reference. New NHTSA Car Truck value as defined in 201 1 Fuel economy regulations. Used in calculations. Total number of vehicles produced for that model. EPA Unadjusted Fuel Economy CO2 calculated from MPG. CO2 weighted 1.15 times higher for diesel vehicles. Average Track x Wheelbase Gas or Diesel Gas or Diesel or Electric Engine Cylinder Displacement Size in Liters Number of Cylinder + 2 if the engine has a turbo or super charger. Actual Number of Engine Cylinders Number of Valves Per Actual Cylinder Type of valve actuation. Type of valve actuation with values compatible with the package file. Type of valve timing with values compatible with the package file. Type of valve lift with values compatible with the package file. Cylinder Deactivation with a value that is compatible with the package file. Type of fuel injection. Type of Boost if any. As Defined by EPA Cert. Definition Certification data Certification data Derived From Certification data and Footprint CSM Worldwide Defined by EPA staff NA Certification data NHTSA Certification data Certification data Certification data Calculated from track width and wheel base Wards Certification data Wards/Certification data Derived From Certification data. Certification data Certification data Wards (Note: Type E is from Cert Data) Wards Wards Wards Wards Wards Wards Wards 1-9 ``````------- The Baseline and Reference Vehicle Fleets Horsepower Torque Trans Type Trans Num of Gears Transmission Structure Drive Drive with AWD Wheelbase Track Width (front) Track Width (rear) Footprint: PU Average Threshold Footprint Curb Weight GVWR Stop- Start/Hybrid/Full EV Import Car Towing Capacity (Maximum) Engine Oil Viscosity Volume 2009 Volume 2010 Volume 20 11 Volume 20 12 Volume 20 13 Max. Horsepower of the Engine Max. Torque of the Engine A=Auto AMT= Automated Manual M=Manual CVT= Continuously Variable Transmission Type Code with number of Gears Number of Gears Transmission definition. Matches the cost definition. Ladder or Unibody Fwd, Rwd, 4wd Fwd, Rwd, Awd, 4wd Length of Wheelbase Length of Track Width in inches Length of Track Width in inches Car and Large Truck Footprints are normal (Average Track x Wheelbase). Medium and Small Truck footprints are the production weighted average for each vehicle. Footprint valve that will be set to 41 for values less than 41 , Will be set to 56 for car values > 56, and will be set to 74 for truck values >74 Curb Weight of the Vehicle Gross Vehicle Weight Rating of the Vehicle Type of Electrification if any. Blank = None Cars Imported Weight a vehicle is rated to tow. Ratio between the applied shear stress and the rate of shear, which measures the resistance of flow of the engine oil (as per SAE Glossary of Automotive Terms) Projected Production Volume for 2009 Projected Production Volume for 2010 Projected Production Volume for 201 1 Projected Production Volume for 2012 Projected Production Volume for 2013 Wards Wards Certification data Certification data Certification data Certification data General Internet Searches Certification data Certification data Some from Edmunds.com or Motortrend.com, Others from product plans with a subset verified with Edmunds.com or Motortrend.com for accuracy. Some from Edmunds.com or Motortrend.com, Others from product plans with a subset verified with Edmunds.com or Motortrend.com for accuracy. Some from Edmunds.com or Motortrend.com, Others from product plans with a subset verified with Edmunds.com or Motortrend.com for accuracy. Derived from data from Edmunds.com or Motortrend.com. Production volumes or specific footprints from product plans. Derived from data from Edmunds.com or Motortrend.com. Production volumes or specific footprints from product plans. Some from Edmunds.com or Motortrend.com, Others from product plans with a subset verified with Edmunds.com or Motortrend.com for accuracy. Some from Edmunds.com or Motortrend.com, Others from product plans with a subset verified with Edmunds.com or Motortrend.com for accuracy. Certification data Certification data Volpe Input File Volpe Input File Calculated based on 2008 volume and Annual Energy Outlook and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. 1-10 ``````------- The Baseline and Reference Vehicle Fleets Volume 20 14 Volume 20 15 Volume 20 16 Volume 20 17 Volume 20 18 Volume 20 19 Volume 2020 Volume 2021 Volume 2022 Volume 2023 Volume 2024 Volume 2025 Low drag brakes Electric Power steering Volpe Index Projected Production Volume for 2014 Projected Production Volume for 2015 Projected Production Volume for 2016 Projected Production Volume for 2017 Projected Production Volume for 2018 Projected Production Volume for 2019 Projected Production Volume for 2020 Projected Production Volume for 2021 Projected Production Volume for 2022 Projected Production Volume for 2023 Projected Production Volume for 2024 Projected Production Volume for 2025 See Volpe Documentation See Volpe Documentation Number used to reorder the vehicles in the EPA baseline in the same order as the Volpe input file. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Calculated based on 2008 volume and AEO and CSM adjustment factors. Volpe Input File Volpe Input File Volpe Input File Notes: 1. For engines not available in the WardsAuto.com engine file, an internet search was done to find this information. 2. These data were obtained from manufacturer's product plans. They were used to block (where possible) the model from adding technology that was already on a vehicle. 3. Ward's Automotive Group data obtained from "2008 Light Vehicle Engines." DOT's CAFE model also uses a series of inputs—referred to as "overrides"—to specify baseline technology content of specific vehicle models (and specific engines and transmissions) and to indicate cases where specific technologies are not applicable to specific vehicle models. In the MY 2008-based market forecast, DOT has corrected some of these settings to indicate that micro-hybrid technology (or more advanced hybrid) is already present on hybrid versions of the Altima, Aura, Civic, Camry, Escape, Highlander, Lexus GS and LS, Lexus RX, Mariner, Malibu, Prius, Tahoe, Tribute, Vue, and Yukon. The CAFE model also uses inputs to assign vehicles to specific "technology classes," where technology-related inputs define the applicability, efficacy, and cost of each technology for vehicles in each technology class. In the MY 2008-based market forecast, DOT has reassigned the Altima 1-11 ``````------- The Baseline and Reference Vehicle Fleets (coupe), Audi A4, Corolla, Impala, Matrix, Passat, and Jetta to technology classes that better represent these vehicles' size and performance characteristics. The sales volumes for the MY 2008 baseline fleet are included in the section below on reference fleet under the MY 2008 columns. Table 1-3 displays the engine technologies present in the baseline fleet. Again, the engine technologies for the vehicles manufactured by these manufacturers in MY 2008 were largely obtained from Ward's Auto online. Table 1-3 2008 Engine Technology Percentages Manufacturer All All All Aston Martin Aston Martin BMW BMW Chrysler/Fiat Chrysler/Fiat Daimler Daimler Ferrari Ferrari Ford Ford Geely/Volvo Geely/Volvo GM GM Honda Honda Hyundai Hyundai Kia Kia Lotus Lotus Mazda Vehicle Type Both Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Trucks Cars Trucks Cars Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars T3 1) I o 1 £ 3% 4% 1% 0% 0% 33% 5% 1% 0% 2% 16% 0% 0% 0% 0% 0% 49% 0% 1% 0% 4% 0% 0% 0% 0% 0% 0% 11% Super Charged 0% 0% 0% 0% 0% 1% 0% 0% 0% 0% 1% 0% 0% 1% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 77% 0% 0% Single Overhead Cam 20% 17% 24% 0% 0% 14% 0% 21% 39% 55% 36% 0% 0% 15% 65% 0% 0% 0% 0% 57% 64% 0% 0% 0% 0% 0% 0% 0% Dual Overhead Cam 63% 73% 48% 100% 0% 86% 100% 72% 4% 45% 64% 100% 0% 85% 32% 100% 100% 31% 56% 43% 36% 100% 100% 100% 100% 100% 0% 99% Overhead Cam 17% 9% 29% 0% 0% 0% 0% 8% 57% 0% 0% 0% 0% 0% 3% 0% 0% 69% 44% 0% 0% 0% 0% 0% 0% 0% 0% 0% Variable Valve Timing Continuous 8% 9% 6% 0% 0% 14% 0% 0% 0% 72% 35% 0% 0% 4% 28% 0% 0% 5% 29% 0% 0% 0% 0% 0% 0% 0% 0% 0% Variable Valve Timing Discrete 22% 24% 19% 100% 0% 86% 100% 42% 4% 4% 17% 100% 0% 0% 1% 100% 100% 17% 31% 27% 4% 0% 0% 0% 0% 100% 0% 7% Variable Valve Timing Intake Only 30% 35% 23% 0% 0% 0% 0% 0% 0% 13% 47% 0% 0% 47% 9% 0% 0% 14% 1% 20% 28% 100% 100% 10% 17% 0% 0% 92% Variable Valve Lift and Timing Continuous 0% 0% 0% 24% 0% 0% 0% 0% 0% 0% 0% 29% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% Variable Valve Lift and Timing Discrete 12% 13% 10% 0% 0% 13% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 100% 100% 0% 0% 0% 0% 0% 0% 0% Cylinder Deactivation 6% 3% 11% 0% 0% 0% 0% 5% 4% 0% 0% 0% 0% 0% 0% 0% 0% 40% 4% 11% 0% 0% 0% 0% 0% 0% 0% 0% a .0 o .° Q 5% 7% 3% 0% 0% 33% 6% 0% 0% 2% 16% 0% 0% 0% 0% 0% 0% 0% 6% 0% 4% 0% 0% 0% 0% 0% 0% 11% 1-12 ``````------- The Baseline and Reference Vehicle Fleets Mazda Mitsubishi Mitsubishi Nissan Nissan Porsche Porsche Spyker/Saab Spyker/Saab Subaru Subaru Suzuki Suzuki Tata/JLR Tata/JLR Tesla Tesla Toyota Toyota Volkswagen Volkswagen Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 24% 6% 0% 0% 0% 17% 12% 100% 0% 15% 3% 0% 0% 0% 0% 0% 0% 0% 0% 43% 1% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 20% 0% 0% 0% 0% 0% 0% 1% 100% 100% 0% 0% 0% 0% 0% 0% 69% 70% 0% 0% 0% 0% 0% 0% 0% 0% 85% 0% 99% 0% 0% 100% 100% 100% 100% 100% 62% 31% 30% 100% 100% 100% 100% 0% 0% 100% 100% 15% 100% 0% 0% 0% 0% 0% 0% 0% 0% 38% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 100% 38% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 13% 0% 0% 4% 0% 100% 100% 17% 0% 0% 23% 0% 0% 76% 0% 0% 0% 29% 61% 48% 99% 87% 0% 0% 96% 100% 0% 0% 0% 62% 31% 7% 0% 0% 24% 100% 0% 0% 71% 39% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 100% 0% 0% 1% 27% 0% 0% 0% 0% 0% 0% 0% 0% 1% 79% 0% 0% 0% 0% 0% 0% 0% 0% 28% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 24% 0% 0% 0% 0% 17% 100% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 8% 6% 89% 100% The data in Table 1-3 indicates that manufacturers had already begun implementing a number of fuel economy/GHG reduction technologies in the baseline (2008) fleet. For example, VW stands out as having a significant number of turbocharged direct injection engines, though it is uncertain whether their engines are also downsized. Some of the valve and cam technologies are quite common in the baseline fleet: for example, nearly half the baseline fleet already has dual cam phasing, while Honda and GM have considerable levels of engines with cylinder deactivation. Honda also has already implemented continuously variable valve lift on a majority of their engines. Part of the implication of these technologies already being present in the baseline is that if manufacturers have already implemented them, they are therefore not available in the rulemaking analysis for improving fuel economy and reducing CC>2 emissions further, requiring the agencies to look toward increasing penetration of these and other technologies and increasingly advanced technologies to project continued improvements in stringency over time. The section below provides further detail on the conversion of the MY 2008 baseline into the MYs 2017-2025 reference fleet. It also describes more of the data contained in the baseline spreadsheet. 1.3.2 The MY 2008 Based MY 2017-2025 Reference Fleet The reference fleet aims to reflect the current market conditions and expectations about conditions of the vehicle fleet during the model years to which the agencies' rules 1-13 ``````------- The Baseline and Reference Vehicle Fleets apply. Fundamentally, constructing this fleet involved projecting the MY 2008 baseline fleet into the MYs 2017-2025 model years. It also included the assumption that none of the vehicle models had changes during this period. Projecting this future fleet is a process that is necessarily uncertain. NHTSA and EPA therefore relied on many sources of reputable information to make these projections. 1.3.2.1 On what data is the reference vehicle fleet based (using the 2008 baseline)? For the MY 2008-based reference fleet, EPA and NHTSA have based the projection of total car and light truck sales on the 2011 projections made by the Energy Information Administration (EIA). EIA publishes a projection of national energy use annually called the Annual Energy Outlook (AEO).3 EIA issued an early release version of AEO2012 January 2012. The agencies are continuing to use this AEO data for the MY 2008 baseline consistent with the NPRM. EPA and NHTSA are employing the newer version of AEO in projecting the reference fleet for the 2010 MY based baseline and reference fleet projection as discussed in section 1.4.2.1. As in the NPRM, the agencies used the Energy Information Administration's (EIA's) National Energy Modeling System (NEMS) to estimate the future relative market shares of passenger cars and light trucks. However, NEMS shifts the market toward passenger cars in order to ensure compliance with EISA's requirement that CAFE standards cause the fleet to achieve 35 mpg by 2020. Because we use our market projection as a baseline relative to which we measure the effects of new standards, and we attempt to estimate the industry's ability to comply with new standards without changing product mix (i.e., we analyze the effects of the final rules assuming manufacturers will not change fleet composition as a compliance strategy), using AEO 2011-projected shift in passenger car market share as provided by EIA would cause the agencies to understate the cost of achieving compliance through additional technology alone. Therefore, for analyses supporting today's final rule, the agencies developed a new projection of passenger car and light truck sales shares by using NEMS to run scenarios from the AEO 2011 reference case, after first deactivating the above- mentioned sales-volume shifting methodology and holding post-2017 CAFE standards constant at MY 2016 levels. Incorporating these changes reduced the projected passenger car share of the light vehicle market by an average of about 5% during 2017-2025. This case is referred to as the "Unforced Reference Case," and the values are shown below in Table 1-5. Table 1-4 AEO 2011 Reference Case Volumes Model Year 2017 2018 2019 2020 2021 2022 Cars 8,984,200 8,998,200 9,170,900 9,553,600 9,801,100 10,056,600 Trucks 6,812,000 6,552,200 6,391,300 6,336,200 6,380,000 6,384,600 Total Vehicles 15,796,100 15,550,400 15,562,200 15,889,800 16,181,100 16,441,200 1-14 ``````------- The Baseline and Reference Vehicle Fleets 2023 2024 2025 10,244,500 10,483,400 10,739,600 6,396,500 6,407,700 6,470,200 16,641,000 16,891,100 17,209,800 Table 1-5 AEO 2011 Interim Unforced Reference Case Volumes Model Year 2017 2018 2019 2020 2021 2022 2023 2024 2025 Cars 8,440,703 8,376,192 8,464,457 8,725,709 8,911,173 9,123,436 9,344,051 9,580,693 9,836,330 Trucks 7,365,619 7,200,218 7,114,201 7,170,230 7,277,894 7,316,337 7,311,438 7,353,394 7,414,129 Total Vehicles 15,806,322 15,576,410 15,578,658 15,895,939 16,189,066 16,439,772 16,655,489 16,934,087 17,250,459 In this 2017 projection, car and light truck sales are expected to get up to 8.4 and 7.4 million units, respectively. While the total level of sales of 15.8 million units is similar to pre-2008 levels, the fraction of car sales in 2017 and beyond is projected to be higher than in the 2000-2007 time frame. Note that EIA's definition of cars and trucks follows that used by NHTSA prior to the MY 2011 CAFE final rule. The MY 2011 CAFE final rule reclassified approximately 1 million 2-wheel drive sport utility vehicles from the truck fleet to the car fleet. EIA's sales projections of cars and trucks for the 2017-2025 model years under the old NHTSA truck definition are shown above in Table 1-4 and Table 1-5. In addition to a shift towards more car sales, sales of segments within both the car and truck markets have also been changing and are expected to continue to change in the future. Manufacturers are continuing to introduce more crossover models which offer much of the utility of SUVs but use more car-like designs and unibody structures. In order to reflect these changes in fleet makeup, EPA and NHTSA used a custom long range forecast purchased from CSM Worldwide (CSM). CSMjk is a well-known industry analyst that provided the forecast used by the agencies for the 2012-2016 final rule. NHTSA and EPA decided to use the forecast from CSM in the MY 2008 baseline reference fleet for several reasons. One, CSM J CSM World Wide is a paid service provider. k As with any long range forecast, CSM World Wide's forecast out to 2025 has uncertainties since many manufacturers do not have full future product plans out that far. 1-15 ``````------- The Baseline and Reference Vehicle Fleets uses a ground up approach (e.g., looking at the number of plants and capacity for specific engines, transmissions, and vehicles) for their forecast, which the agencies believe is a robust forecasting approach1. Two, CSM agreed to allow us to publish their high level data, on which the forecast is based, in the public domain. Three, the CSM forecast covered all the timeframe of greatest relevance to this analysis (2017-2025 model years). Four, it provided projections of vehicle sales both by manufacturer and by market segment. And five, it utilized market segments similar to those used in the EPA emission certification program and fuel economy guide, such that the agencies could include only the vehicle types covered by the final standards. The agencies note that CSM developed the forecast during a period when the United States economy was undergoing significant stress and some automobile manufacturers were experiencing a high degree of financial uncertainty. In the time since CSM developed its forecast,industry sales and in particular the sales for some individual manufacturers have turned out differently than in the CSM forecast. Because forecasting the market out to MY 2025 has uncertainties, the agencies believe there are benefits from using the CSM forecast for one of the two analyses cases to reflect some level of uncertainty in the final rule analysis. It is feasible that the CSM forecast could represent what might happen in the future. CSM created a forecast that covered model years 2017-2025. Since the agencies used this forecast to generate the reference fleet (i.e., the fleet expected to be sold absent any increases in the stringency regulations after the 2016 model year), it is important for the forecast not to reflect changes in fleet composition during 2017-2025 attributed to CAFE/ GHG standards. However, CSM assumed that CAFE and GHG standards would continue to increase in stringency after 2016, although CSM did not use specific future standards as quantitative inputs to its model. In its quantitative analysis, CSM used fuel price, industry demand, consumer demand and other economic factors to project the composition of the future fleet. In response to question by the agencies, CSM indicated that their assumption of future standards had a negligible (non-discernible) impact on their forecast since it was not a direct quantitative input to the model such that CSM's forecast would have been essentially the same had CSM assumed no stringency increases after 2016. The agencies combined the CSM forecast with data from other sources to create the reference fleet projections. This process is discussed in sections that follow. 1 There are other forecasting groups that do similar projections and meet all these criteria. LMC Automotive (formerly JD Power Forecasting) is another, and this was used for the alternate reference case projection as described below. 1-16 ``````------- The Baseline and Reference Vehicle Fleets 1.3.2.2 How do the agencies develop the reference vehicle fleet? The process of producing the 2017-2025 reference fleet involved combining the baseline fleet with the projection data described above. This was a complex multistep procedure, which is described in this section. 1.3.2.3 How was the 2008 baseline data merged with the CSM data? For the NPRM, EPA and NHTSA employed the same methodology as in the 2012-16 rule for mapping certification vehicles to CSM vehicles; the results were used again for analysis supporting today's final rule. Merging the 2008 baseline data with the 2017-2025 CSM data required a thorough mapping of certification vehicles to CSM vehicles by individual make and model. One challenge that the agencies faced when determining a reference case fleet was that the sales data projected by CSM had different market segmentation than the data contained in EPA's database. In order to create a common segmentation between the two databases, the agencies performed a side-by-side comparison of each vehicle model in both datasets, and created an additional "CSM segment" modifier in the spreadsheet to map the two datasets. The reference fleet sales based on the "CSM segmentation" was then projected. The baseline data and reference fleet volumes are available to the public. The baseline Excel spreadsheet in the docket is the result of the merged files.4 The spreadsheet provides specific details on the sources and definitions for the data. The Excel file contains several tabs. They are: "Data", "Data Tech Definitions", "SUM", "SUM Tech Definitions", "Truck Vehicle Type Map", and "Car Vehicle Type Map". "Data" is the tab with the raw data. "Data Tech Definitions" is the tab where each column is defined and its data source named. "SUM" is the tab where the raw data is processed to be used in the OMEGA and Volpe models. The "SUM" tab minus columns A-F and minus the Generic vehicles is the input file for the models. The "Generic" manufacturer (shown in the "SUM" tab) is the sum of all manufacturers and is calculated as a reference, and for data verification purposes. It is used to validate the manufacturers' totals. It also gives an overview of the fleet. Table 1-6 shows the sum of the models chosen. The number of models is determined by the number of unique segment and vehicle type combinations. These combinations of segment and vehicle type (the vehicle type number is the same as the technology package number) are determined by the technology packages discussed in the EPA RIA. "SUM Tech Definitions" is the tab where the columns of the "SUM" tab are defined. The "Truck Vehicle Type Map" and "Car Vehicle Type Map" map the number of cylinder and valve actuation technology to the "tech package" vehicle type number. Table 1-6 Models from the SUM Tab Model Model Car Like LargeSuv >=V8 Vehicle Type: 13 Car Like LargeSuv V6 Vehicle Type: 16 Car Like LargeSuv V6 Vehicle Type: 12 1-17 ``````------- The Baseline and Reference Vehicle Fleets Car Like LargeSuv V6 Vehicle Type: 9 Car Like LargeSuv 14 and 15 Vehicle Type: 7 Car Like MidSizeSuv V6 Vehicle Type: 8 Car Like MidSizeSuv V6 Vehicle Type: 5 Car Like MidSizeSuv 14 Vehicle Type: 7 Car Like SmallSuv V6 Vehicle Type: 12 Car Like SmallSuv V6 Vehicle Type: 4 Car Like SmallSuv 14 Vehicle Type: 3 LargeAuto >=V8 Vehicle Type: 13 LargeAuto >=V8 Vehicle Type: 10 LargeAuto >=V8 Vehicle Type: 6 LargeAuto V6 Vehicle Type: 12 LargeAuto V6 Vehicle Type: 5 MidSizeAuto >=V8 Vehicle Type: 13 MidSizeAuto >=V8 Vehicle Type: 10 MidSizeAuto >=V8 (7 or >) Vehicle Type: 6 MidSizeAuto V6 Vehicle Type: 12 MidSizeAuto V6 Vehicle Type: 8 MidSizeAuto V6 Vehicle Type: 5 MidSizeAuto 14 Vehicle Type: 3 In the combined EPA certification and CSM database, all 2008 vehicle models were assumed to continue out to 2025, though their volumes changed in proportion to CSM projections. Also, any new models expected to be introduced within the 2009-2025 timeframe are not included in the data. These volumes are reassigned to the existing models to keep the overall fleet volume the same. All MYs 2017-2025 vehicles are mapped to the existing vehicles by a process of mapping to manufacturer market share and overall segment distribution. The mappings are discussed in the next section. Further discussion of this limitation is discussed below in section 1.3.2.4. The statistics of this fleet will be presented below since further modifications were required to the volumes as the next section describes. 1.3.2.4 How were the CSM forecasts normalized to the AEO forecasts for the 2008- based fleet? The next step in the agencies' generation of the reference fleet is one of the more complicated steps to explain. Here, the projected CSM forecasts for relative sales of cars and trucks by manufacturer and by market segment was normalized (set equal) to the total sales estimates of the Early Release of the 2011 Annual Energy Outlook (AEO). NHTSA and EPA used projected car and truck volumes for this period from Early AEO 2011. However, the AEO projects sales only at the car and truck level, not at the manufacturer and model-specific level, and the agencies' analysis requires this further level of detail. The CSM data provided year-by-year percentages of cars and trucks sold by each manufacturer as well as the percentages of each vehicle segment. Using these percentages normalized to the AEO- projected volumes then provided the manufacturer-specific market share and model-specific sales for model years 2017-2025 (it is worth clarifying that the agencies are not using the 1-18 ``````------- The Baseline and Reference Vehicle Fleets model-specific sales volumes from CSM, only the higher-level volumes by manufacturer and segment). This process is described in greater detail in the following paragraphs. In order to determine future production volumes, the agencies developed multipliers by manufacturer and vehicle segment that could be applied to MY 2008 volumes. The process for developing the multipliers is complicated, but is easiest to explain as a three-step process, though the first step is combined with both the second and third step, so only one multiplier per manufacturer and vehicle segment is developed. The three steps are: 1. Adjust total car and truck sales to match AEO projections. 2. Adjust car sales to match CSM market share projections for each manufacturer and car segment. 3. Adjust truck sales to match CSM market share projections for each manufacturer and truck segment. The first step is the adjustment of total car and truck sales in 2008 to match AEO projections of total car and truck sales in 2017-2025. The volumes for all of the trucks in 2008 were added up (TruckSum2008), and so were the volumes of all the cars (CarSum2008). A multiplier was developed to scale the volumes in 2008 to the AEO projections. The example equation below shows the general form of how to calculate a car or truck multiplier. The AEO projections are shown above in Table 1-4. Example Equation : TruckMultiplier(Year X) = AEOProjectionforTrucks(Year X) / TruckSum2008 CarMultiplier(Year X) = AEOProjectionforCars(Year X) / CarSum2008 Where: Year X is the model year of the multiplier. The AEO projection is different for each model year. Therefore, the multipliers are different for each model year. The multipliers can be applied to each 2008 vehicle as a first adjustment, but multipliers based solely on AEO have limited value since those multipliers can only give an adjustment that will give the correct total numbers of cars and trucks without the correct market share or vehicle mix. A correction factor based on the CSM data, which does contain market share and vehicle segment mix, is therefore necessary, so combining the AEO multiplier with CSM multipliers (one per manufacturer, segment, and model year) will give the best multipliers. There were several steps in developing an adjustment for Cars based on the CSM data. CSM provided data on the market share and vehicle segment distribution. The first step in determining the adjustment for Cars was to total the number of Cars in each vehicle segment by manufacturer in MY 2008. A total for all manufacturers in each segment was also calculated. The next step was to multiply the volume of each segment for each manufacturer 1-19 ``````------- The Baseline and Reference Vehicle Fleets by the CSM market share. The AEO multiplier was also applied at this time. This gave projected volumes with AEO total volumes and market share correction for Cars. This is shown in the "Adjusted for 2017AEO and Manufacturer Market Share" column of Table 1-7. The next step is to adjust the sales volumes for CSM vehicle segment distribution. The process for adjusting for vehicle segment is more complicated than a simple one step multiplication. In order to keep manufacturers' volumes constant and still have the correct vehicle segment distribution, vehicles need to move from segment to segment while maintaining constant manufacturers' totals. Six rules and one assumption were applied to accomplish the shift. The assumption (based on the shift in vehicle sales in 2008 and 2009) is that people are moving to smaller vehicles in the rulemaking time frame independently of regulatory requirements. A higher-level (less detailed) example of this procedure is provided in Section II of the preamble. Vehicles from CSM's "Luxury Car," "Specialty Car," and "Other Car" segments, if reduced, will be equally distributed to the remaining four categories ("Full-Size Car," "Mid- Size Car," "Small Car," "Mini Car"). If these sales increased, they were taken from the remaining four categories so that the relative sales in these four categories remained constant. Vehicles from CSM's "Luxury Car," "Specialty Car," and "Other Car" segments, if increased will take equally from the remaining categories ("Full-Size Car," "Mid-Size Car," "Small Car," "Mini Car"). All manufacturers have the same multiplier for a given segment shift based on moving all vehicles in that segment to achieve the CSM distribution. Table 1-7 shows how the 2017 vehicles moved and the multipliers that were created for each adjustment. This does not mean that new vehicle segments will be added (except for Generic Mini Car described in the next step) to manufacturers that do not produce them. Vehicles within each manufacturer will be shifted as close to the distribution as possible given the other rules. Table 1-8 has the percentages of Cars per CSM segment. These percentages are multiplied by the total number of vehicles in a given year to get the total sales in the segment. Table 1-7 shows the totals for 2017 in the "2017 AEO-CSM Sales Goal" column. When "Full-Size Car," "Mid-Size Car," "Small Car" are processed, if vehicles need to move in or out of the segment, they will move into or out of the next smaller segment. So, if Mid-Size Cars are being processed they can only move to or be taken from Small Cars. Note: In order to accomplish this, a "Generic Mini Car" segment was added to manufacturers who did not have a Mini (type) Car in production in 2008, but needed to shift down vehicles from the Small Car segment. The data must be processed in the following order: "Luxury Car," "Specialty Car," "Other Car," "Full-Size Car," "Mid-Size Car," "Small Car." The "Mini Car" does not need to be processed separately. By using this order, it works out that vehicles will always move toward the correct distribution. There are two exceptions, BMW and Porsche only have 1-20 ``````------- The Baseline and Reference Vehicle Fleets "Luxury Car," "Specialty Car," and "Other Car" vehicles, so their volumes were not changed or shifted since these rules did not apply to them. When an individual manufacturer multiplier is applied for a segment, the vehicles move to or from the appropriate segments as specified in the previous rules and as shown in Table 1-7. Table 1-7 2017 Model Year Volume Shift* CSM Segment All Full-Size Car All Luxury Car All Mid-Size Car All Mini Car All Small Car All Specialty Car All Others 2008 MY Sales 829,896 1,048,341 2,103,108 617,902 1,912,736 469,324 0 Adjusted for 2017AEOand Manufacturer Market Share 830,832 1,408,104 2,500,723 868,339 2,548,393 627,425 0 Luxury, Specialty, Other Adjustment 818,226 1,423,691 2,475,267 851,234 2,513,350 702,048 0 Full Size Adjustment 347,034 1,423,691 2,946,459 851,234 2,513,350 702,048 0 Midsize Adjustment 347,034 1,423,691 2,431,715 851,234 3,028,094 702,048 0 Small Car Adjustment 347,034 1,423,691 2,431,715 1,439,985 2,439,343 702,048 0 2017 AEO- CSM Sales Goal 347,034 1,423,691 2,431,715 1,439,985 2,439,343 702,048 0 Number Vehicles that shift and Where All Full-Size Car All Luxury Car All Mid-Size Car All Mini Car All Small Car All Specialty Car All Others (12,606) 15,587 (25,456) (17,105) (35,043) 74,623 0 (471,192) 0 471,192 0 0 0 0 0 0 (514,744) 0 514,744 0 0 0 0 0 588,751 (588,751) 0 0 Individual Manufacturer Multiplier All Full-Size Car All Luxury Car All Mid-Size Car All Mini Car All Small Car All Specialty Car All Others 0.973 0.963 1 0.42 0.97 1.55 0.96 1-21 ``````------- The Baseline and Reference Vehicle Fleets Table 1-8 CSM - Percent of Cars per Segment* CSM Segment Compact Car Full-Size Car Luxury Car Mid-Size Car Mini Car Small Car Specialty Car Others 2017 0.00% 3.95% 16.70% 27.68% 15.33% 27.77% 8.56% 0.00% 2018 0.00% 3.56% 16.87% 27.77% 15.46% 27.57% 8.76% 0.00% 2019 0.00% 3.35% 17.14% 27.47% 15.45% 27.74% 8.84% 0.00% 2020 0.00% 4.10% 17.23% 26.94% 15.46% 27.99% 8.27% 0.00% 2021 0.00% 3.59% 17.05% 27.18% 15.59% 28.29% 8.29% 0.00% 2022 0.00% 3.03% 17.02% 27.82% 15.67% 28.43% 8.03% 0.00% 2023 0.00% 2.97% 17.10% 28.51% 15.47% 28.18% 7.77% 0.00% 2024 0.00% 2.46% 17.40% 28.11% 15.23% 28.49% 8.31% 0.00% 2025 0.00% 2.46% 17.40% 28.11% 15.23% 28.49% 8.31% 0.00% Mathematically, an individual manufacturer multiplier is calculated by making the segment the goal and dividing by the previous total for the segment (shown in Table 1-8). If the number is greater than 1, the vehicles are entering the segment, and if the number is less than 1, the vehicles are leaving the segment. So, for example, if Luxury Cars have an adjustment of 1.5, then for a specific manufacturer who has Luxury Cars, a multiplier of 1.5 is applied to its luxury car volume, and the total number of vehicles that shifted into the Luxury segment is subtracted from the remaining segments to maintain that company's market share. On the other hand, if Large Cars have an adjustment of 0.7, then for a specific manufacturer who has Large Cars, a multiplier of 0.7 is applied to its Large Cars, and the total number of vehicles leaving that segment is transferred into that manufacturer's Mid-Size Cars. After the vehicle volumes are shifted using the above rules, a total for each manufacturer and vehicle segment is maintained. The total for each manufacturer segment for a specific model year (e.g., 2017 General Motors Luxury Cars) divided by the MY 2008 total for that manufacturer segment (e.g., 2008 General Motors Luxury Cars) is the new multiplier used to determine the future vehicle volume for each vehicle model. This is done by taking the multiplier (which is for a specific manufacturer and segment) times the MY 2008 volume for the specific vehicle model (e.g., 2008 General Motors Luxury Car Cadillac CTS). This process is repeated for each model year (2017-2025). The method used to adjust CSM Trucks to the AEO market share was different than the method used for Cars. The process for Cars is different than Trucks because it is not possible to predict how vehicles would shift between segments based on current market trends. This is because of the added utility of some trucks that makes their sales more insensitive to factors like fuel price. Again, CSM provided data on the market share and vehicle segment distribution. The process for having the fleet match CSM's market share and vehicle segment distribution was iterative. The following totals were determined: 1-22 ``````------- The Baseline and Reference Vehicle Fleets The total number of trucks for each manufacturer in 2008 model year. The total number of trucks in each truck segment in 2008 model year. The total number of truck in each segment for each manufacturer in 2008 model year. The total number of trucks for each manufacturer in a specific future model year based on the AEO and CSM data. This is the goal for market share. The total number of trucks in each truck segment in a specific future model year based on the AEO and CSM data. This is the goal for vehicle segment distribution. Table 1-9 has the percentages of Trucks per CSM segment. Table 1-9 CSM - Percent of Trucks per Segment CSM Segment Full-Size CUV Full-Size Pickup Full-Size SUV Full-Size Van Mid-Size CUV Mid-Size MAV Mid-Size Pickup Mid-Size SUV Mid-Size Van Small CUV Small MAV Small SUV 2017 5.9% 16.8% 1.9% 1.2% 18.0% 4.5% 6.1% 4.1% 11.6% 26.0% 2.5% 1.3% 2018 6.3% 16.5% 1.5% 1.2% 17.4% 4.6% 6.1% 4.8% 11.9% 25.9% 2.6% 1.2% 2019 6.8% 15.9% 1.3% 1.1% 17.6% 4.9% 6.1% 4.8% 11.9% 25.7% 2.8% 1.1% 2020 7.5% 16.1% 1.0% 1.4% 17.2% 5.4% 5.6% 4.5% 11.7% 25.6% 2.9% 1.2% 2021 8.3% 15.4% 0.9% 1.3% 16.9% 5.9% 5.7% 4.7% 11.6% 25.1% 3.0% 1.1% 2022 8.8% 15.1% 0.8% 1.3% 16.8% 6.2% 5.7% 4.8% 11.6% 24.9% 3.1% 1.1% 2023 9.5% 14.3% 0.5% 1.3% 16.8% 6.5% 5.8% 4.8% 11.6% 24.7% 3.1% 1.1% 2024 9.2% 13.8% 0.5% 1.2% 17.0% 7.1% 5.9% 4.6% 11.3% 25.3% 3.2% 1.0% 2025 9.1% 13.5% 0.6% 1.2% 17.0% 7.4% 5.8% 4.6% 11.3% 25.3% 3.2% 1.0% To start, the agencies created two different types of tables. One table had each manufacturer with its total sales for 2008 (similar to Table 1-11). This table will have the goal for each manufacturer, and a column added for each iteration with the current total. The second table has a truck segment total by manufacturer. The second table starts out with a "Generic" manufacturer (Table 1-11) which is the table where the goal resides. Each manufacturer (BMW for example is shown in Table 1-12) is then listed below the "Generic" manufacturer. With each iteration, a new total is added for each segment that is calculated and added to the table. This is not shown in the tables below. The agencies then engaged in a process of first adjusting the numbers in the tables to the goal for market share distribution. This was followed by adjusting to the goal for vehicle segment distribution. Each time an adjustment was done a new column was added. An adjustment was done by creating a multiplier (either segment distribution-based or manufacturer distribution-based) and applying it to each vehicle segment total in the current iteration. A manufacturer-based multiplier is calculated by taking the goal total for a manufacturer and dividing by the current total (starting with 2008 model year volumes) for a manufacturer. A segment distribution-based 1-23 ``````------- The Baseline and Reference Vehicle Fleets multiplier is calculated by taking the goal distribution volumes in the Generic manufacturer set and dividing them by the current volume. Table 1-10, Table 1-11, and Table 1-12 below illustrates two iterations using BMW as an example. Table 1-10 Manufacturer Truck Totals BMW 2008 Model Year Sales 61,324 Manufacturer Distribution 2017 Volume Goal 138.053 Multiplier for Iteration 1 138,053/61324=2.25 Table 1-11 Segment Specific Truck Totals for All Manufacturers Manufacturer Generic** Generic Generic Generic Generic Generic Generic Generic Generic Generic Generic Generic CSM Segment Full-Size Pickup Mid-Size Pickup Full-Size Van Mid-Size Van Mid-Size MAV Small MAV Full-Size SUV Mid-Size SUV Small SUV Full-Size CUV Mid-Size CUV Small CUV 2008 Model Year Sales 1,332,335 452,013 33,384 719,529 110,353 231,265 559,160 436,080 196,424 264,717 923,165 1,612,029 Segment Distribution 2017 Volume Goal 1,240,844 452,017 85,381 855,022 331,829 186,637 138,821 305,382 94,657 433,683 1,327,905 1,913,439 Multipliers 0.931 1.000 2.558 1.188 3.007 0.807 0.248 0.700 0.482 1.638 1.438 1.187 ** Generic means all manufacturers. 1-24 ``````------- The Baseline and Reference Vehicle Fleets Table 1-12 Segment Specific Truck Totals for BMW Manufacturer BMW BMW BMW BMW BMW BMW BMW BMW BMW BMW BMW BMW CSM Segment Full-Size Pickup Mid-Size Pickup Full-Size Van Mid-Size Van Mid-Size MAV Small MAV Full-Size SUV Mid-Size SUV Small SUV Full-Size CUV Mid-Size CUV Small CUV Total BMW Vehicles 2008 Model Year Sales 3,882 36,409 21,033 61,324 Iteration 1 Adjust for Market Share 2.25*3,882=8,739 2.25*36,409=81,964 2.25*21,033=47,350 138,053 Iteration 2 Adjust for Segment Distribution 2.85*8,739=24,907 1.1*81,964=90,134 1.. 02*47,350=48,306 163,347 Using this process, the numbers will get closer to the goal of matching CSM's market share for each manufacturer and distribution for each vehicle segment after each of the iterations. The iterative process is carried out until the totals nearly match the goals. After 19 iterations, all numbers were within 0.01% of CSM's distributions. The calculation iterations could have been stopped sooner, but they were continued to observe how the numbers would converge. After the market share and segment distribution were complete, the totals need to be used to create multipliers that could be applied to the original individual 2008 model year vehicle volumes (each unique manufacture models volume). The total for each manufacturer segment divided by the 2008 model year total for each manufacturer segment gives a multiplier that can be applied to each vehicle based on its manufacturer and segment. The above process is done for each model year needed (2017-2025). The multipliers are then applied to each vehicle in 2008 model year, which gives a volume for each vehicle in 2017 through 2025 model year. 1.3.3 What are the sales volumes and characteristics of the MY 2008 based reference fleet? Table 1-13 and Table 1-15 below contain the sales volumes that result from the process above for MY 2008 and 2017-2020. Table 1-14 and Table 1-16 below contain the sales volumes that result from the process above for MY 2021-2025. Table 1-13 Vehicle Segment Volumes" Reference Class Segment Actual and Projected Sales Volume 2008 2017 2018 2019 2020 1-25 ``````------- The Baseline and Reference Vehicle Fleets LargeAuto MidSizeAuto CompactAuto SubCmpctAuto LargePickup SmallPickup LargeSUV MidSizeSUV Small SUV MiniVan Cargo Van 562,240 3,098,927 1,979,461 1,365,833 1,582,226 177,497 2,783,949 1,263,360 285,355 642,055 110,858 376,107 3,311,268 2,347,980 2,458,222 1,514,619 156,227 3,194,489 1,358,755 148,251 754,562 185,841 356,768 3,290,408 2,325,393 2,454,112 1,443,766 157,932 3,150,101 1,309,212 149,933 739,551 199,234 353,609 3,303,621 2,369,301 2,489,208 1,383,190 160,752 3,177,868 1,267,394 154,675 717,065 201,974 394,864 3,381,785 2,448,021 2,553,350 1,386,195 146,029 3,203,244 1,285,822 162,677 714,323 219,628 a Volumes in this table are based on the pre-2011 NHTSA definition of Cars and Trucks. Table 1-14 Vehicle Segment Volumes" Reference Class Segment LargeAuto MidSizeAuto CompactAuto SubCmpctAuto LargePickup SmallPickup LargeSUV MidSizeSUV Small SUV MiniVan Cargo Van Projected Sales Volume 2021 380,192 3,442,116 2,520,977 2,626,364 1,368,301 150,123 3,312,914 1,281,240 167,223 729,078 210,539 2022 358,295 3,548,263 2,592,199 2,687,167 1,349,421 147,138 3,362,608 1,283,244 169,643 738,982 202,812 2023 362,672 3,692,533 2,632,926 2,721,102 1,301,293 151,315 3,412,753 1,268,288 170,239 740,785 201,585 2024 356,173 3,751,496 2,744,634 2,796,061 1,271,751 154,627 3,475,873 1,292,662 173,191 720,720 196,900 2025 368,843 3,814,941 2,843,069 2,878,288 1,260,389 154,838 3,520,992 1,305,362 175,713 726,256 201,768 a Volumes in this table are based on the pre-2011 NHTSA definition of Cars and Trucks. Table 1-15 2011+ NHTSA Car and Truck Definition Based Volumes Vehicle Type Trucks Cars Cars and Trucks Actual and Projected Sales Volume 2008 5,621,193 8,230,568 13,851,761 2017 5,818,655 9,987,667 15,806,322 2018 5,671,046 9,905,364 15,576,410 2019 5,582,962 9,995,696 15,578,658 2020 5,604,377 10,291,562 15,895,939 Table 1-16 2011+ NHTSA Car and Truck Definition Based Volumes Vehicle Type Projected Sales Volume 2021 2022 2023 2024 2025 1-26 ``````------- The Baseline and Reference Vehicle Fleets Trucks Cars Cars and Trucks 5,683,902 10,505,165 16,189,066 5,703,996 10,735,777 16,439,772 5,687,486 10,968,003 16,655,489 5,675,949 11,258,138 16,934,087 5,708,899 11,541,560 17,250,459 Table 1-17 and Table 1-18 below contain the sales volumes by manufacturer and vehicle type for MY 2008 and 2017-2025. Table 1-17 NHTSA Car and Truck Definition Manufacturer Volumes Manufacturers All All All Aston Martin Aston Martin BMW BMW Chrysler/Fiat Chrysler/Fiat Daimler Daimler Ferrari Ferrari Ford Ford Geely/Volvo Geely/Volvo GM GM HONDA HONDA HYUNDAI HYUNDAI Kia Kia Lotus Lotus Mazda Vehicle Type Both Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars 2008 Baseline Sales 13,851,761 8,230,568 5,621,193 1,370 - 291,796 61,324 703,158 956,792 208,195 79,135 1,450 - 956,699 814,194 32,748 65,649 1,507,797 1,587,391 1,006,639 505,140 337,869 53,158 221,980 59,472 252 - 246,661 2017 Projected Volume 15,806,322 9,987,667 5,818,655 1,035 - 313,022 138,053 418,763 409,702 284,847 86,913 6,676 - 1,299,899 763,549 41,887 88,234 1,362,761 1,462,204 1,154,600 596,481 592,027 152,885 322,044 98,702 240 - 253,540 2018 Projected Volume 15,576,410 9,905,364 5,671,046 1,051 - 322,939 131,942 397,538 387,858 276,409 83,651 6,700 - 1,311,467 748,829 42,187 89,394 1,438,355 1,474,076 1,138,087 544,619 578,373 151,461 312,370 98,280 243 - 262,512 2019 Projected Volume 15,578,658 9,995,696 5,582,962 1,072 - 346,075 131,373 391,689 366,447 281,425 88,188 6,794 - 1,332,039 717,773 43,125 91,575 1,505,025 1,493,511 1,144,639 527,535 582,971 155,642 314,879 100,679 250 - 266,951 2020 Projected Volume 15,895,939 10,291,562 5,604,377 1,034 - 357,942 128,339 415,319 360,677 290,989 92,919 6,916 - 1,378,789 717,037 42,615 93,003 1,530,755 1,544,983 1,163,666 525,089 598,283 154,173 323,676 96,535 266 - 270,078 1-27 ``````------- The Baseline and Reference Vehicle Fleets Mazda Mitsubishi Mitsubishi Nissan Nissan PORSCHE PORSCHE Spyker/Saab Spyker/Saab Subaru Subaru Suzuki Suzuki Tata/JLR Tata/JLR Tesla Tesla Toyota Toyota Volkswagen Volkswagen Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 55,885 85,358 15,371 717,869 305,546 18,909 18,797 21,706 4,250 116,035 82,546 79,339 35,319 9,596 55,584 800 - 1,260,364 951,136 291,483 26,999 51,788 65,099 37,632 870,797 444,938 35,093 13,233 20,024 2,871 224,112 78,242 90,708 22,109 55,881 57,579 27,986 - 1,849,196 1,330,511 551,638 128,819 57,535 63,671 36,300 849,678 412,383 35,444 12,001 20,007 3,596 216,598 75,152 89,932 21,385 56,222 56,606 28,435 - 1,834,181 1,223,415 540,036 145,491 57,494 63,826 35,454 854,400 398,559 36,116 11,469 20,144 3,826 217,095 72,832 90,568 20,692 57,267 57,854 28,990 - 1,836,306 1,142,104 537,114 146,891 58,154 65,080 35,215 882,791 397,869 35,963 11,141 21,069 3,509 223,466 72,458 93,548 20,675 58,182 56,213 27,965 - 1,883,734 1,154,304 554,822 146,700 Table 1-18 NHTSA Car and Truck Definition Manufacturer Volumes Manufacturers All All All Aston Martin Aston Martin BMW BMW Chrysler/Fiat Chrysler/Fiat Daimler Daimler Ferrari Ferrari Vehicle Type Both Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 2021 Projected Volume 16,189,066 10,505,165 5,683,902 1,058 - 359,098 128,724 421,013 348,613 300,378 99,449 7,059 - 2022 Projected Volume 16,439,772 10,735,777 5,703,996 1,049 - 360,034 128,899 424,173 363,008 304,738 100,935 7,138 - 2023 Projected Volume 16,655,489 10,968,003 5,687,486 1,041 - 360,561 127,521 423,882 361,064 312,507 105,315 7,227 - 2024 Projected Volume 16,934,087 11,258,138 5,675,949 1,141 - 388,193 146,525 426,017 344,962 332,337 107,084 7,441 - 2025 Projected Volume 17,250,459 11,541,560 5,708,899 1,182 - 405,256 145,409 436,479 331,762 340,719 101,067 7,658 - 1-28 ``````------- The Baseline and Reference Vehicle Fleets Ford Ford Geely/Volvo Geely/Volvo GM GM HONDA HONDA HYUNDAI HYUNDAI Kia Kia Lotus Lotus Mazda Mazda Mitsubishi Mitsubishi Nissan Nissan PORSCHE PORSCHE Spyker/Saab Spyker/Saab Subaru Subaru Suzuki Suzuki Tata/JLR Tata/JLR Tesla Tesla Toyota Toyota Volkswagen Volkswagen Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 1,401,617 714,181 41,768 92,726 1,530,020 1,564,277 1,198,880 535,916 613,355 156,466 331,319 95,432 278 - 274,740 59,227 65,851 35,309 912,629 408,029 36,475 11,242 21,294 3,560 230,780 72,773 95,725 20,767 58,677 58,153 28,623 - 1,903,706 1,215,539 585,607 148,734 1,415,221 714,266 41,686 92,512 1,507,653 1,578,556 1,237,504 539,235 627,964 157,493 339,102 94,694 290 - 281,150 60,307 67,261 35,227 937,447 411,883 36,607 11,385 21,709 3,461 238,613 72,736 97,599 20,734 59,349 58,590 28,369 - 1,986,077 1,235,052 593,314 146,750 1,474,797 700,005 42,031 96,840 1,496,819 1,606,495 1,265,564 536,898 634,308 161,189 342,746 95,688 299 - 296,910 61,966 67,680 35,469 954,340 417,121 36,993 11,370 22,410 3,435 241,612 73,022 99,263 20,803 60,639 58,865 28,150 - 2,036,992 1,224,980 596,749 153,927 1,503,670 688,854 42,461 99,181 1,493,597 1,636,805 1,307,851 536,994 657,710 166,092 351,882 96,119 308 - 300,614 61,971 70,728 36,001 982,771 422,217 39,504 11,409 22,800 3,426 248,283 74,142 100,447 21,162 63,728 57,981 30,862 - 2,080,528 1,208,013 605,336 156,939 1,540,109 684,476 42,588 101,107 1,524,008 1,673,936 1,340,321 557,697 677,250 168,136 362,783 97,653 316 - 306,804 61,368 73,305 36,387 1,014,775 426,454 40,696 11,219 23,130 3,475 256,970 74,722 103,154 21,374 65,418 56,805 31,974 - 2,108,053 1,210,016 630,163 154,284 Table 1-19 also shows how the change in fleet make-up may affect the footprint distributions over time. The resulting data indicate that footprint will not change significantly between 2008 and 2025. There will be an increase in the number of cars sold, which will cause the average footprints for cars and trucks combined to be slightly smaller (about 2%). 1-29 ``````------- The Baseline and Reference Vehicle Fleets This is the result of AEO projecting an increased number of cars, and CSM predicting that most of that increase will be in the subcompact segment. Again, we note that in order to ensure that our baseline inputs were not influenced by the final regulations, agencies re-ran AEO to hold standards constant after 2016 (the reader will remember from the text above that CSM had indicated that its projections were not sensitive to assumptions about new standards). Table 1-19 Production Weighted Foot Print Mean Model Year 2008 2017 2018 2019 2020 2021 2022 2023 2024 2025 Average Footprint of all Vehicles 48.8 48.0 47.9 47.8 47.8 47.8 47.7 47.7 47.5 47.5 Average Footprint Cars 45.2 44.6 44.6 44.6 44.6 44.6 44.6 44.6 44.6 44.6 Average Footprint Trucks 53.9 53.8 53.7 53.6 53.7 53.6 53.6 53.5 53.3 53.3 Table 1-20 below shows the changes in engine cylinders over the model years. The current assumptions show that engines will be downsized over the model years to which these final rules apply. This shift is a projected consequence of the expected changes in class and segment mix as predicted by AEO and CSM, and does not represent engine downsizing attributable to the 2012-2016 light-duty CAFE and GHG standards. Table 1-20 Percentages of 4,6,8 Cylinder Engines by Model Year Model Year 2008 2017 2018 2019 2020 2021 2022 Trucks 4 Cylinders 10.3% 10.9% 10.6% 10.4% 10.3% 10.3% 10.3% 6 Cylinders 56.4% 63.7% 64.5% 65.5% 65.6% 66.3% 66.7% 8 Cylinders 33.3% 25.4% 24.8% 24.1% 24.1% 23.4% 23.0% Cars 4 Cylinders 56.9% 60.6% 60.7% 60.7% 60.3% 60.6% 61.1% 6 Cylinders 37.8% 34.5% 34.4% 34.3% 34.7% 34.4% 34.2% 8 Cylinders 5.3% 5.0% 5.0% 5.0% 5.0% 4.9% 4.8% 1-30 ``````------- The Baseline and Reference Vehicle Fleets 2023 2024 2025 10.3% 10.5% 10.5% 67.7% 68.1% 68.2% 22.0% 21.4% 21.3% 60.9% 61.0% 61.1% 34.3% 34.1% 34.0% 4.8% 4.8% 4.8% As discussed above, the agencies also developed a second market forecast using updated data. The following section describes those efforts and their results. 1.4 The 2010 MY Based Fleet The 2010 MY based fleet is similar to the 2008 MY based fleet in that it was created with similar types of information. The 2010 MY based fleet uses interim AEO 2012 total car and truck volumes, a long range forecast from LMC Automotive (formerly J.D. Powers Forecasting) used for manufacturer market share and product mix, and 2010 CAFE certification data for 2010 model volumes and technology. The 2008 MY based fleet, in contrast, uses interim AEO 2011, a long range forecast from CSM World Wide, and 2008 CAFE certification data. The remainder of section 1.4 describes the 2010 based fleet projection and how it was created. 1.4.1 On what data is the MY 2010 baseline vehicle fleet based? Similar to the 2008 baseline, most of the information about the vehicles that make up the 2010 fleet was gathered from EPA's emission certification and fuel economy database, most of which is available to the public. These data included, by individual vehicle model produced in MY 2010, vehicle production volume, fuel economy rating for CAFE certification, carbon dioxide emissions, fuel type, fuel injection type, EGR, number of engine cylinders, displacement, intake valves per cylinder, exhaust valves per cylinder, variable valve timing, variable valve lift, engine cycle, cylinder deactivation, transmission type, drive (rear- wheel, all-wheel, etc.), hybrid type (if applicable), and aspiration (naturally-aspirated, turbocharged, etc.). In addition to this information about each vehicle model produced in MY 2010, the agencies augmented this description with publicly-available data which includes more complete technology descriptions from Ward's Automotive Group.111'11 As with the 2008 baseline, the agencies also used Edmunds.com and Motortrend.com°'p'q Like the MY 2008 baseline fleet and the baseline vehicle fleet used in the MYs 2012-2016 rulemaking, the MY 2010 baseline vehicle fleet is developed using publicly-available data to the largest extent possible. 0 Motortrend.com and Edmunds.com: Used as a source for footprint and vehicle weight data. p Motortrend.com and Edmunds.com are free, no-fee internet sites. q A small amount of footprint data from manufacturers' MY 2008 product plans submitted to the agencies was used in the development of the baseline. 1-31 ``````------- The Baseline and Reference Vehicle Fleets The process for creating the 2010 baseline fleet Excel file was streamlined when compared with the past rulemaking. EPA and NHTSA worked together to create the baseline using 2010 CAFE certification data from EPA's Verify database. EPA contracted LMC Automotive (formerly JD Power Forecasting) to produce an up to date long range forecast of volumes for the future fleet. Using information sources discussed below, NHTSA identified technology and footprint information for every vehicle model in the 2010 CAFE certification data. EPA used the forecast from LMC Automotive to project the future fleet's volume projections (a detailed discussion of the method used to project the future fleet volumes is in 1.4.2.1 of this chapter.) Both agencies used the previously mentioned data to populate input files for the agencies' respective modeling systems. The structure of the market forecast input file used for DOT's CAFE Compliance and Effects Modeling System (a.k.a. "the CAFE model") is described in the model documentation.5 To help readers who wish to directly examine the baseline fleet file for EPA's OMEGA model, and to provide some idea of its contents for those readers who do not, Table 1-21 shows the columns of the complete fleet file, which includes the MY 2008 baseline data that was compiled. Each column has its name, definition (description) and source. Most elements shown in Table 1-21 also appear in the market forecast input file for DOT's modeling system, which accommodates some additional data elements discussed in the model documentation. Table 1-21 Data, Definitions, and Sources Data Item Index Manufacturer CERT Manufacturer Name Name Plate Model Reg Class Our Class Definition Index Used to link EPA and NHTSA baselines Common name of company that manufactured vehicle. May include more name plates than Cert Manufacturer Name. Certification name of company that manufactured vehicle Name of Division Name of Vehicle EPA Fuel Economy Class Name If a car's Footprint<43 then "SubCmpctAuto" If a car's 43<=Footprint<46 then "CompactAuto" If a car's 46<=Footprint<53 then "MidSizeAuto" If a car's Footprint >=53 then "LargeAuto" If a S.U.V.'s Footprint < 43 then "SmallSuv" If a S U V 's 43<=Footprint<46 then "MidSizeSuv" If a S.U.V's Footprint >=46 then "LargeSuv" If a Truck's Footprint < 50 then "SmallPickup" If a Truck's Footprint>=50 then "LargPickup" If a Van's Structure is Ladder then "CargoVan" If a Van's Structure is Unibody then "Minivan" Data Type Number Name (Ex.Chrysler) Name (Ex.Chrysler) Name (Ex. Dodge) Name (Ex. Viper) EPA Class Name (Ex. SUBCOMPACT CARS) Name(Ex SmallSuv) Wards Engine Acronyms NA NA NA NA NA NA NA Where The Data is From Created Certification data Certification data Certification data Certification data Certification data Derived From Certification data and Footprint 1-32 ``````------- The Baseline and Reference Vehicle Fleets NEW SEGMENT Vehicle Type Number Generic Vehicle Index From Sum Page Vehicle Index From Sum Page Pre2011 NHTSA Defined C/T Our Class C/T Traditional Car/Truck NHTSA Defined New NHTSA Car/Truck Total Production Volume Fuel Econ. (mpg) C02 Fuel (G,D,C) Fuel Type Disp (lit.) Effective Cyl Actual Cylinders Valves Per Cylinder Valve Type Valve Actuation WT WLT LMC Automotive (formerly J. D. Powers) new segmentation for the vehicle. Vehicle Type Number assigned to a vehicle based on its number of cylinders, valves per cylinder, and valve actuation technology. See Truck Vehicle Type Map and Car Vehicle Type Map sheets for details. Number to be used as a cross reference with the Sum Pages. Number to be used as a cross reference with the Sum Pages. C= Car, T=Truck. As defined in the certification database. C= Car, T=Truck. As defined in the certification database. Not used in calculations. DP=Domestic Passenger Cars, I=Import Passenger Car, LT= Light duty Truck. As defined in the certification database. Not used in calculations. New NHTSA Car Truck value as determined by NHTSA. Used in calculations. Total number of vehicles produced for that model. EPA Unadjusted Fuel Economy CO2 calculated from MPG. CO2 weighted 1.15 times higher for diesel vehicles. Gas or Diesel or CNG Gas or Diesel or CNG Engine Cylinder Displacement Size in Liters Number of Cylinder + 2 if the engine has a turbo or super charger. Actual Number of Engine Cylinders Number of Valves Per Actual Cylinder Type of valve actuation. Type of valve actuation with values compatible with the package file. Type of valve timing with values compatible with the package file. Type of valve lift with values compatible with the package file. Name (Ex. Compact Sporty, Large Pickup, etc.) Number Number Number Letter(C or T) Letter(C or T) IP,DP,LT Letter(C or T) number(ex.5500) number(ex.25) Number G,D,C Gas or Diesel or CNG number(ex. 4) number(ex. 6) number(ex. 4) number(ex. 4) Acronym(Ex. DOHC, SOHC, OHV, E, R) DOHC, SOHC, OHV VVTC,VVTD, VVTI VVTLC, VVTLD NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA DOHC, SOHC, OHV, E, R DOHC, SOHC, OHV VVTC,VVTD, VVTI VVTLC, VVTLD LMC Automotive Mapped by EPA staff NA NA Certification data Created Certification data Certification data Certification data Certification data Certification data Certification data Certification data Wards/Certificati on data Derived From Certification data. Certification data Certification data Wards (Note: Type E is from Cert Data) Wards Wards Wards 1-33 ``````------- The Baseline and Reference Vehicle Fleets Deac Inline or V Engine Fuel injection system Boost Engine Cycle Horsepower Torque Cooled EGR Trans Type Tran Num of Gears Structure Drive Drive with AWD Wheelbase Track Width (front) Track Width (rear) Footprint Threshold Footprint Curb Weight ITW GVWR Stop-Start/ Hybrid/ Full EV Towing Capacity (Maximum) Engine Oil Viscosity Low drag brakes Cylinder Deactivation with a value that is compatible with the package file. Configuration of the Engine Type of fuel injection. Type of Boost if any. As Defined by EPA Cert. Definition Max. Horsepower of the Engine Max. Torque of the Engine Cooled Exhaust Gas Recirculation A=Auto AMT= Automated Manual M=Manual CVT= Continuously Variable Transmission Type Code with number of Gears Number of Gears Ladder or Unibody Fwd, Rwd, 4wd Fwd, Rwd, Awd, 4wd Length of Wheelbase Length of Track Width in inches Length of Track Width in inches Car and Large Truck Footprints are normal (Average Track x Wheelbase). Medium and Small Truck footprints are the production weighted average for each vehicle. Footprint valve that will be set to 41 for values less than 41, Will be set to 56 for car values > 56, and will be set to 66 for truck values >66 Curb Weight of the Vehicle Inertia Test Weight Gross Vehicle Weight Rating of the Vehicle Type of Electrification if any. Blank = None Weight a vehicle is rated to tow. Ratio between the applied shear stress and the rate of shear, which measures the resistance of flow of the engine oil (as per SAE Glossary of Automotive Terms) See Volpe Documentation Deac lorV DI, MPI Super Charged (Single), Turbo (Single) Letter Ex. G for Gas) number(ex. 125) number(ex. 125) YorN letter(ex. A) letters and possible a number(ex.A5, ex. CVT) number(ex. 4) Unibody or Ladder (Ex. Ladder) Acronym(Ex. Rwd) Acronym(Ex. Awd) number(ex. 125) number(ex. 45) number(ex. 45) Number Number number(ex.4500) number(ex.4500) number(ex.4500) EV75,2-Mode- IMA,Power- Split, Stop-Start Number (in Pounds) Text (Ex. OW20; 5W20) See Volpe CD lorV DI, SFI, EFI, MPI TRB,SPR NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA Wards Wards Wards Wards Wards Wards Wards Certification data Certification data Certification data Certification data Volpe Input File Certification data Certification data From Edmonds.com or Motortrend.com, From Edmonds.com or Motortrend.com From Edmonds.com or Motortrend.com From Edmonds.com or Motortrend.com Derived from data from Edmunds.com or Motortrend.com Certification data Certification data Volpe Input File Certification data Volpe Input File Volpe Input File Volpe Input File 1-34 ``````------- The Baseline and Reference Vehicle Fleets Power steering Technology Class Safety Class Safety Class Number Volume 20 10 Volume 20 11 Volume 20 12 Volume 20 13 Volume 20 14 Volume 20 15 Volume 20 16 See Volpe Documentation For technology application purposes only and should not be confused with vehicle classification for regulatory purposes. Defined by DOT. See Volpe Documentation See Volpe Documentation Projected Production Volume for 2010 Projected Production Volume for 2011 Projected Production Volume for 2012 Projected Production Volume for 2013 Projected Production Volume for 2014 Projected Production Volume for 2015 Projected Production Volume for 2016 Documentation See Volpe Documentation Text (Ex. Subcompact, Subcompact Performance, Compact, Compact Performance, Midsize, Midsize Performance, Large, Large Performance, Minivan, Small LT, Midsize LT, Large LT; (LT = SUV/Pickup/Van)) See Volpe Documentation See Volpe Documentation Number Number Number Number Number Number Number NA NA NA NA NA NA NA NA NA NA NA Volpe Input File Volpe Input File Volpe Input File Volpe Input File Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. 1-35 ``````------- The Baseline and Reference Vehicle Fleets Volume 20 17 Volume 20 18 Volume 20 19 Volume 2020 Volume 2021 Volume 2022 Volume 2023 Volume 2024 Volume 2025 Projected Production Volume for 2017 Projected Production Volume for 2018 Projected Production Volume for 2019 Projected Production Volume for 2020 Projected Production Volume for 2021 Projected Production Volume for 2022 Projected Production Volume for 2023 Projected Production Volume for 2024 Projected Production Volume for 2025 Number Number Number Number Number Number Number Number Number NA NA NA NA NA NA NA NA NA Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Calculated based onMY2010 volume and AEO and LMC adjustment factors. Table 1-22 displays the engine technologies present in the MY 2010 baseline fleet. Again, the engine technologies for the vehicles manufactured by these manufacturers in MY 2010 were largely obtained from data found on Ward's Auto online. 1-36 ``````------- The Baseline and Reference Vehicle Fleets Table 1-22 2010 Engine Technology Percentages Manufacturers All All All Aston Martin Aston Martin BMW BMW Chrysler/Fiat Chrysler/Fiat Daimler Daimler Ferrari Ferrari Ford Ford Geely Geely General Motors General Motors Honda Honda Hyundai Hyundai Kia Kia Lotus Lotus Mazda Mazda Mitsubishi Mitsubishi Nissan Nissan Porsche Porsche Spyker Spyker Subaru Vehicle Type Both Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Trucks Cars Trucks Cars Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Turbo Charged 3% 4% 2% 0% 0% 38% 33% 0% 0% 0% 8% 0% 0% 1% 1% 38% 25% 0% 0% 1% 2% 3% 0% 0% 0% 0% 0% 4% 11% 6% 0% 0% 0% 16% 1% 0% 0% 6% Super Charged 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 1% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 16% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% Single Overhead Cam 22% 18% 29% 0% 0% 0% 0% 42% 30% 50% 24% 0% 0% 12% 70% 0% 0% 0% 0% 58% 63% 0% 0% 0% 0% 0% 0% 0% 0% 100% 100% 0% 0% 0% 0% 0% 0% 92% Dual Over Head Cam 68% 78% 50% 100% 0% 100% 100% 49% 4% 50% 76% 100% 0% 88% 30% 100% 100% 45% 75% 42% 37% 100% 100% 100% 100% 100% 0% 100% 100% 0% 0% 100% 100% 100% 100% 0% 0% 8% Over Head Cam 10% 4% 21% 0% 0% 0% 0% 9% 66% 0% 0% 0% 0% 0% 0% 0% 0% 55% 25% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% Variable Valve Timing Continuous 41% 11% 17% 0% 0% 28% 0% 0% 0% 52% 24% 0% 0% 2% 50% 0% 0% 0% 0% 58% 63% 0% 0% 0% 0% 0% 0% 0% 0% 96% 74% 0% 0% 0% 0% 0% 0% 0% Variable Valve Timing Discrete 26% 26% 27% 0% 0% 70% 82% 41% 4% 46% 69% 100% 0% 0% 0% 100% 100% 42% 73% 42% 37% 0% 0% 0% 0% 100% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 2% Variable Valve Timing Intake Only 39% 48% 22% 38% 0% 0% 0% 0% 0% 0% 0% 0% 0% 69% 26% 0% 0% 0% 1% 0% 0% 100% 100% 100% 73% 0% 0% 100% 100% 0% 0% 100% 100% 100% 100% 0% 0% 6% Variable Valve Lift and Timing Continuous 6% 6% 5% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 42% 37% 0% 0% 0% 0% 23% 0% 0% 0% 0% 0% 9% 0% 100% 100% 0% 0% 2% Variable Valve Lift and Timing Discrete 2% 2% 2% 0% 0% 45% 67% 0% 0% 46% 76% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% Cylinder Deactivation 4% 3% 7% 0% 0% 0% 0% 5% 13% 0% 0% 0% 0% 0% 0% 0% 0% 4% 3% 17% 45% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% Direct Injection 9% 9% 9% 0% 0% 38% 33% 0% 0% 0% 8% 90% 0% 1% 1% 0% 0% 37% 31% 0% 0% 0% 0% 0% 0% 0% 0% 2% 0% 0% 0% 0% 0% 83% 100% 0% 0% 0% 1-37 ``````------- The Baseline and Reference Vehicle Fleets Subaru Suzuki Suzuki Tata Tata Tesla Tesla Toyota Toyota Volkswagen Volkswagen Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 0% 0% 0% 0% 0% 0% 0% 0% 0% 62% 33% 0% 0% 0% 22% 15% 0% 0% 0% 0% 4% 0% 87% 0% 0% 0% 0% 0% 0% 0% 0% 68% 21% 13% 100% 100% 100% 100% 0% 0% 100% 100% 32% 79% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 47% 21% 13% 2% 50% 67% 67% 0% 0% 16% 62% 32% 67% 0% 98% 50% 33% 33% 0% 0% 84% 38% 0% 0% 13% 0% 0% 45% 64% 0% 0% 0% 0% 1% 52% 0% 0% 0% 7% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 67% 67% 0% 0% 4% 0% 68% 100% The data in Table 1-22 indicate that manufacturers had already begun implementing a number of fuel economy/GHG reduction technologies in the baseline (2010) fleet. For example, as in the 2008 baseline fleet, VW stands out as having a significant number of turbocharged direct injection engines. Some of the valve and cam technologies are quite common in the baseline fleet: for example, nearly half the baseline fleet already has dual cam phasing, while Honda and Chrysler have considerable levels of engines with cylinder deactivation. Honda also has already implemented continuously variable valve lift on a majority of their engines. Part of the implication of these technologies already being present in the baseline is that if manufacturers have already implemented them, they are therefore not available in the rulemaking analysis for improving fuel economy and reducing CC>2 emissions further, requiring the agencies to look toward increasing penetration of these and other technologies and increasingly advanced technologies to project continued improvements in stringency over time. The data in Table 1-23 shows the changes between the 2010 engine technology penetrations and the 2008 engine technology penetrations. Perhaps to increase fuel economy, manufacturers applied considerable additional technology between 2008 and 2010. Volkswagen's trucks have direct injection increased to 100 percent (although VW's cars had a 21% decrease). Manufacturers changed variable valve timing, presumably based on engine- specific design considerations. For example, Honda replaced discrete valve timing with continuous valve lift or timing, and Kia added variable valve lift and timing to 90% of its cars and 56% of its trucks. 1-38 ``````------- The Baseline and Reference Vehicle Fleets Table 1-23 The difference (2010-2008) in Engine Technology Percentages Manufacturer All All All Aston Martin Aston Martin BMW BMW Chrysler/Fiat Chrysler/Fiat Daimler Daimler Ferrari Ferrari Ford Ford Geely/Volvo Geely/Volvo GM GM Honda Honda Hyundai Hyundai Kia Kia Lotus Lotus Mazda Mazda Mitsubishi Mitsubishi Nissan Nissan Porsche Porsche Spyker/Saab Spyker/Saab Subaru Subaru Suzuki Vehicle Type Both Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Trucks Cars Trucks Cars Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars 1 o o | H 0% 0% 1% 0% 0% 5% 28% -1% 0% -2% -8% 0% 0% 1% 1% 38% -24% 0% -1% 1% -2% 3% 0% 0% 0% 0% 0% -7% -13% 0% 0% 0% 0% -1% -11% NA NA -9% -3% 0% Super Charged 0% 0% 0% 0% 0% -1% 0% 0% 0% 0% 0% 0% 0% -1% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% -61% 0% 0% 0% 0% 0% 0% 0% 0% 0% NA NA 0% 0% 0% Single Overhead Cam 2% 1% 5% 0% 0% -14% 0% 21% -9% -5% -12% 0% 0% -3% 5% 0% 0% 0% 0% 1% -1% 0% 0% 0% 0% 0% 0% 0% -1% 0% 0% 0% 0% 0% 0% NA NA 23% 17% 0% Dual Overhead Cam 5% 5% 2% 0% 0% 14% 0% -23% 0% 5% 12% 0% 0% 3% -2% 0% 0% 14% 19% -1% 1% 0% 0% 0% 0% 0% 0% 1% 1% 0% 0% 0% 0% 0% 0% NA NA -23% -17% 0% Overhead Cam -7% -5% -8% 0% 0% 0% 0% 1% 9% 0% 0% 0% 0% 0% -3% 0% 0% -14% -19% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% NA NA 0% 0% 0% Variable Valve Timing Continuous 33% 2% 11% 0% 0% 14% 0% 0% 0% -20% -11% 0% 0% -2% 22% 0% 0% -5% -29% 58% 63% 0% 0% 0% 0% 0% 0% 0% 0% -4% 36% 0% 0% 0% 0% NA NA 0% 0% 0% Variable Valve Timing Discrete 4% 2% 8% -100% 0% -16% -18% -1% 0% 42% 52% 0% 0% 0% -1% 0% 0% 25% 42% 15% 33% 0% 0% 0% 0% 0% 0% -7% -13% 0% 0% -4% 0% -100% -100% NA NA 2% -10% 2% Variable Valve Timing Intake Only 9% 13% -1% 38% 0% 0% 0% 0% 0% -13% -47% 0% 0% 22% 17% 0% 0% -14% 0% -20% -28% 0% 0% 90% 56% 0% 0% 8% 13% 0% 0% 4% 0% 100% 100% NA NA -25% -7% 98% Variable Valve Lift and Timing Continuous 6% 6% 5% -24% 0% 0% 0% 0% 0% 0% 0% -29% 0% 0% 0% 0% 0% 0% 0% 42% 37% 0% 0% 0% 0% 23% 0% 0% 0% 0% 0% 9% 0% 100% 100% NA NA 2% 13% 0% Variable Valve Lift and Timing Discrete -10% -11% -8% 0% 0% 32% 67% 0% 0% 46% 76% 0% 0% 0% 0% 0% 0% 0% 0% -100% -100% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% -100% NA NA -1% -27% 0% o ta > o 1 >-> U -2% 0% -4% 0% 0% 0% 0% 0% 9% 0% 0% 0% 0% 0% 0% 0% 0% -36% -1% 6% 45% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% NA NA 0% 0% 0% .0 o o '& Q 4% 2% 6% 0% 0% 5% 27% 0% 0% -2% -8% 90% 0% 1% 1% 0% 0% 37% 25% 0% -4% 0% 0% 0% 0% 0% 0% -9% -24% 0% 0% 0% 0% 66% 0% NA NA 0% 0% 0% 1-39 ``````------- The Baseline and Reference Vehicle Fleets Suzuki Tata/JLR Tata/JLR Tesla Tesla Toyota Toyota Volkswagen Volkswagen Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 0% 0% 0% NA NA 0% 0% 19% 1% 0% 22% -5% NA NA 0% 0% 4% 0% 0% 0% 0% NA NA 0% 0% -17% 0% 0% 0% 0% NA NA 0% 0% 17% 100% 0% 0% 0% NA NA 0% 0% 0% 0% 0% 0% 0% NA NA 0% 0% 47% 0% 50% -9% 67% NA NA -13% 1% -16% 99% 50% 9% -67% NA NA 13% -1% 0% 0% 0% 45% 64% NA NA 0% 0% 1% 0% 0% 7% 0% NA NA 0% 0% -1% 79% 0% 0% 0% NA NA 0% 0% 0% 0% 0% 67% 67% NA NA -4% -6% -21% 100% The section below provides further detail on the conversion of the MY 2010 baseline into the MYs 2017-2025 reference fleet. It also describes more of the data contained in the baseline spreadsheet. 1.4.2 The MY 2010 Based MY 2017-2025 Reference Fleet The reference fleet aims to reflect the current market conditions and expectations about conditions of the vehicle fleet during the model years to which the agencies' rules apply. Fundamentally, constructing this fleet involved projecting the MY 2010 baseline fleet into the MYs 2017-2025 model years. It also included the assumption that none of the vehicle models had changes during this period. Projecting this future fleet is a process that is necessarily uncertain. As with the MY 2008-based MY 2017-2025 reference fleet, NHTSA and EPA relied on many sources of reputable information to make these projections. 1.4.2.1 On what data is the reference vehicle fleet based (using the MY2010 baseline)? EPA and NHTSA have based the projection of total car and light truck sales on the most recent projections available made by the Energy Information Administration (EIA). EIA's Annual Energy Outlook (AEO) projects future energy production, consumption and prices.6 EIA issued an "early release" version of AEO 2012 in January 2012. The complete final version of AEO 2012 was released June 25, 2012, but by that time EPA/NHTSA had already completed analyses supporting the final 2017-2025 standards using the interim data release. Similar to the analyses supporting the MYs 2012-2016 rulemaking and for the 2008 based fleet projection, the agencies have used the Energy Information Administration's (EIA's) National Energy Modeling System (NEMS) to estimate the future relative market shares of passenger cars and light trucks. However, as explained above, NEMS shifts the market toward passenger cars in order to ensure compliance with EISA's requirement that CAFE standards cause the fleet to achieve 35 mpg by 2020. Because we use our market projection as a baseline relative to which we measure the effects of new standards, and we attempt to estimate the industry's ability to comply with new standards without changing product mix (i.e.., we analyze the effects of the final rules assuming manufacturers will not change fleet composition as a compliance strategy), using the Interim AEO 2012-projected shift in passenger car market share as provided by EIA would cause the agencies to understate the cost of achieving compliance through additional technology, alone. Therefore, for the current analysis, the agencies developed a new projection of passenger car and light truck sales shares by using NEMS to run scenarios from the Interim AEO 2012 reference case, after 1-40 ``````------- The Baseline and Reference Vehicle Fleets first deactivating the above-mentioned sales-volume shifting methodology and holding post- 2017 CAFE standards constant at MY 2016 levels. Incorporating these changes reduced the projected passenger car share of the light vehicle market by an average of about 5% during 2017-2025. As with the comparable exercise for the 2008 MY baseline fleet, this case is referred to as the "Unforced Reference Case," and the values are shown below in Table 1-24. Table 1-24 AEO 2012 Interim Unforced Reference Case Values used in the 2010 Market Fleet Projection Model Year 2017 2018 2019 2020 2021 2022 2023 2024 2025 Cars 8,713,800 8,631,900 8,688,600 8,774,500 8,898,400 9,033,900 9,179,600 9,368,800 9,525,700 Trucks 7,098,300 6,973,500 6,973,500 6,855,700 6,831,700 6,853,300 6,827,600 6,878,200 6,929,100 Total Vehicles 15,812,100 15,605,400 15,662,100 15,630,200 15,730,100 15,887,200 16,007,200 16,247,000 16,454,800 In 2017, car and light truck sales are projected to be 8.7 and 7.1 million units, respectively (compared to 8.4 and 7.4 million in the 2010 AEO projection). While the total level of sales of 15.8 million units is similar to pre-2008 levels, the fraction of car sales in 2017 and beyond is projected to be higher than in the previous AEO projections. In addition to a shift towards more car sales, sales of segments within both the car and truck markets have also been changing and are expected to continue to change in the future. The agencies also wanted to use the most updated information on Chrysler projections, as the older NPRM projection conducted by CSM showed Chrysler sales to be very low in 2025. The agencies agree with the Chrysler comments that the NPRM projections are most likely outdated and too low with respect to Chrysler's market share. In order to reflect these changes in fleet makeup, EPA and NHTSA used a custom long range forecast purchased from LMC Automotive (formerly J.D. Powers Forecasting). J.D. Powers is a well-known industry analyst. NHTSA and EPA decided to use the forecast from LMC Automotive (J.D. Powers Forecasting) for MY2010-based market forecast for several reasons. First, Like CSM, LMC Automotive uses a ground up approach (e.g., looking at the number of plants and capacity for specific engines, transmissions, and vehicles) for their forecast, which the agencies believe is a robust forecasting approach. Second, LMC Automotive allows us to publish their entire forecast in the public domain. Third, the LMC Automotive forecast covered all the timeframe of greatest relevance to this analysis (2017-2025 model years). Fourth, it provided projections of vehicle sales both by manufacturer and by market segment. Fifth, it utilized market segments similar to those used in the EPA emission certification program and fuel economy 1-41 ``````------- The Baseline and Reference Vehicle Fleets guide, such that the agencies could include only the vehicle types covered by the final standards. And finally, it had a more updated projection of Chrysler sales. LMC Automotive created a forecast that covered model years 2010-2025. Since the agencies used this forecast to generate the reference fleet (i.e., the fleet expected to be sold absent any increases in the stringency regulations after the 2016 model year), it is important for the forecast to be independent of increases during 2017-2025 in the stringency of CAFE/ GHG standards. LMC Automotive does not use the CAFE or GHG standard as an input to their model, and specifically had no assumption of increase in stringency in the 2017-2025 time frame. The agencies combined the LMC Automotive forecast with data from other sources to create the 2010 baseline reference fleet projections. This process is discussed in sections that follow. 1.4.2.2 How do the agencies develop the 2010 baseline 2017-2025 reference vehicle fleet? The process of producing the MY 2010 baseline 2017-2025 reference fleet involved combining the baseline fleet with the projection data described above. This was a complex multistep procedure, which is described in this section. The procedure is new and some of the steps are different than those used with the MY2008 baseline fleet projection. 1.4.2.3 How was the 2010 baseline data merged with the LMC Automotive data? EPA and NHTSA employed a different method from the method used in the NPRM for mapping certification vehicles to LMC Automotive (LMC) vehicles. Merging the 2010 baseline data with the 2017-2025 LMC data required a thorough mapping of certification vehicles to LMC vehicles by individual make and model. One challenge that the agencies faced when determining a reference case fleet was that the sales data projected by LMC had different market segmentation than the data contained in EPA's internal database. In order to create a common segmentation between the two databases, the agencies performed a side-by- side comparison of each vehicle model in both datasets, and created an additional "NEW SEGMENT" modifier in the spreadsheet to map the two datasets. The reference fleet sales based on the "NEW SEGMENT" was then projected. The baseline data and reference fleet volumes are available to the public. The baseline Excel spreadsheet in the docket is the result of the merged files.7 The spreadsheet provides specific details on the sources and definitions for the data. The Excel file contains several tabs. They are: "Data", "Data Tech Definitions", "SUM", "SUM Tech Definitions", "Truck Vehicle Type Map", and "Car Vehicle Type Map". "Data" is the tab with the raw data. "Data Tech Definitions" is the tab where each column is defined and its data source named. "SUM" is the tab where the raw data is processed to be used in the OMEGA and Volpe models. The "SUM" tab minus columns A-F and minus the Generic vehicles is the input file for the models. The "Generic" manufacturer (shown in the "SUM" tab) is the sum of all 1-42 ``````------- The Baseline and Reference Vehicle Fleets manufacturers and is calculated as a reference, and for data verification purposes. It is used to validate the manufacturers' totals. It also gives an overview of the fleet. Table 1-6 shows some of the unique models chosen from the "SUM" tab. A model is made up of a unique combination of segment and vehicle type. The number of models is determined by the number of unique segment and vehicle type combinations. These combinations of segment and vehicle type (the vehicle type number is the same as the technology package number) are determined by the technology packages discussed in the EPA RIA. "SUM Tech Definitions" is the tab where the columns of the "SUM" tab are defined. Table 1-25 Models from the SUM Tab Model Model Car Like LargeSuv 14 Vehicle Type: 7 Car Like LargeSuv 14, V6 Vehicle Type: 8 Car Like Large Suv V6 Vehicle Type: 9 Car Like MidSizeSuv 14 Vehicle Type: 7 Car Like MidSizeSuv 14, V6 Vehicle Type: 8 Car Like MidSizeSuv V6 Vehicle Type: 9 Car Like SmallSuv V6 Vehicle Type: 10 Large Auto V6 Vehicle Type: 3 Large Auto V6 Vehicle Type: 4 Large Auto >=V6 Vehicle Type: 5 Large Auto >=V8 Vehicle Type: 6 MidSizeAuto 14 Vehicle Type: 2 MidSizeAuto 14, V6 Vehicle Type: 3 MidSize Auto V6 Vehicle Type: 4 MidSize Auto >=V6 Vehicle Type: 5 MidSize Auto V8 Vehicle Type: 6 In the combined EPA certification and LMC data, all 2010 vehicle models were assumed to continue out to 2025, though their volumes changed in proportion to LMC projections. Also, any new models expected to be introduced within the 2011-2025 timeframe are not included in the data. These volumes are reassigned to the existing models to keep the overall fleet volume the same. All MYs 2017-2025 vehicles are mapped to the existing vehicles by a process of mapping to manufacturer's future segment volumes. The mappings are discussed in the next section. Further discussion of this limitation is discussed below in section 1.4.2.4. The statistics of this fleet will be presented below since further modifications were required to the volumes as the next section describes. 1-43 ``````------- The Baseline and Reference Vehicle Fleets 1.4.2.4 How were the LMC forecast and the AEO forecast used to project the future fleet volumes? As with the comparable step in the MY 2008 baseline 2017-2025 reference fleet process, the next step in the agencies' generation of the reference fleet is one of the more complicated steps to explain. First, the 2010 CAFE data was mapped to the LMC segments. Second, the breakdown of segment volumes by manufacturer was compared between the LMC and CAFE data sets. Third, a correction was applied for Class 2B vehicles (Large Pickup Trucks) in the LMC data. Fourth, the individual manufacturer segment multipliers were created by year. And finally, the absolute volumes of cars and trucks were normalized (set equal) to the total sales estimates of the Early Release of the 2012 Annual Energy Outlook (AEO). The process started with mapping the LMC segments to the CAFE data. The process was simple yet time consuming. The mapping required determining the LMC segment by looking at each of the 1171 vehicles in the LMC quarter forecast, and labeling it in the "New Segment" column of the new data spreadsheet. The segments were somewhat different from the ones employed by CSM. LMC has 27 segments and CSM has 18 segments. Table 1-26 has both the LMC Segments and the CSM segments for reference. Table 1-27 shows some of the Chrysler/Fiatr vehicles in the CAFE data with their "New Segment" identified. Table 1-26 List of LMC Segments and CSM Segments LMC Segments Compact Conventional Compact CUV Compact MPV Compact Premium Conventional Compact Premium CUV Compact Premium Sporty Compact Sporty Compact Utility Large Conventional Large Pickup Large Premium Conventional Large Premium Sporty Large Premium Utility Large Utility Large Van Midsize Conventional Midsize CUV Midsize Pickup Midsize Premium Conventional Midsize Premium CUV Midsize Premium Sporty Midsize Premium Utility CSM Class Full-Size Car Full-Size CUV Full-Size Pickup Full-Size SUV Full-Size Van Luxury Car Mid-Size Car Mid-Size CUV Mid-Size MAV Mid-Size Pickup Mid-Size SUV Mid-Size Van Mini Car Small Car Small CUV Small MAV Small SUV Specialty Car r Chrysler/Fiat is being used as an example throughout this section to make the example calculations easier to follow. 1-44 ``````------- The Baseline and Reference Vehicle Fleets Midsize Sporty Midsize Utility Midsize Van Table 1-27 Example of Chrysler/Fiat vehicles being mapped to segments based on the LMC Forecast Manufacturer Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Name Plate Chrysler Chrysler Chrysler Chrysler Dodge Dodge Dodge Dodge Dodge Dodge Model 300 AWD PT Cruiser Sebring Town & Country FWD Caliber Challenger Charger Dakota Pickup 2wd Grand Caravan FWD Journey 2wd NEW SEGMENT Large Conventional Compact MPV Midsize Conventional Midsize Van Compact Conventional Midsize Sporty Large Conventional Midsize Pickup Midsize Van Midsize CUV In this next step, segment volume by manufacturer was compared between the LMC and CAFE data sets. This is necessary to determine if all of the segments a manufacturer will produce in the future are currently represented by the 2010 CAFE data. Almost all the future segments matched the current segments with the exception of some premium vs. standard class vehicles. In cases where there was not a vehicle model in a premium class (such as Compact Premium CUV) in the future, but there was a model in the standard class (Compact CUV), the future premium class volume was added to the standard class volume. The same thing was done if the opposite was true, i.e. if there was not a vehicle in a standard class (such as Compact CUV) in the future, but there was one in the premium class (Compact Premium CUV), the future standard class volume was added to the premium class volume. Table 1-28 shows the New Segments, the LMC 2010 Volumes, and the LMC 2018 Volumes for Chrysler/Fiat. The Compact Premium Conventional, Compact Premium CUV, and Compact Premium Sporty were not available from Chrysler/Fiat in 2010, but are available in 2018. As mentioned, the volumes from all three of those premium segments were added to the standard segments Compact Conventional, Compact CUV, and Compact Sporty in years were the premium segments were produced. Table 1-28 Example Chrysler/Fiat 2010 Volumes by Segment from the LMC Forecast NEW SEGMENT Compact Conventional Compact CUV Compact MPV Compact Premium Conventional Compact Premium CUV Compact Premium Sporty Compact Utility Large Conventional Large Pickup Large Van LMC 2010 Volume 45,082 54,514 9,440 - - - 166,492 112,513 199,652 - LMC 2018 Volume 91,136 78,307 61,461 35,027 12,783 209 210,979 185,553 284,583 1-45 ``````------- The Baseline and Reference Vehicle Fleets Midsize Conventional Midsize CUV Midsize Pickup Midsize Premium Conventional Midsize Premium CUV Midsize Premium Sporty Midsize Sporty Midsize Utility Midsize Van Sub-Compact Conventional 89,508 48,577 13,047 - - 392 36,791 93,352 215,598 - 88,007 91,880 27,141 9,309 12,476 3,014 - 154,401 155,408 97,342 A step that is related to the comparison step is the filtering of Class 2b vehicles from the LMC forecast. LMC includes Class 2b vehicles (vans and large pickup trucks) in its light- duty forecast. Class 2b vans are all appropriately classified as MDPVs (Medium Duty Passenger Vehicles) and must be included in the forecast since they are regulated under the light-duty CAFE and GHG programs. Class 2b large pickup trucks, however, are not regulated under the light-duty CAFE and GHG programs (rather under the medium- and heavy-duty fuel efficiency and GHG programs, see 76 FR at 57120), and must therefore be removed from the forecast. This is accomplished by a creating a multiplier for each manufacturer's large pickup trucks and applying it to each manufacturer's large pickup truck volume every model year in the LMC forecast; specifically, by taking a manufacturer's 2010 model year large pickup CAFE volume and dividing its 2010 model year large pickup LMC volume. Table 1-29 shows the volumes and the resulting multiplier for Chrysler/Fiat, while Table 1-30 shows the 2025 LMC volume, the multiplier and the result of applying the multiplier to the original volume for Chrysler/Fiat. Table 1-29 Example Values Used to Determine the Class 2b Truck Multiplier for Chrysler/Fiat Manufacturer Chrysler/Fiat NEW SEGMENT Large Pickup LMC 2010 Volume 199,652 2010 CAFE Volume 120,645 Truck Multiplier 0.60 Table 1-30 Example Values Used to Determine Chrysler/Fiat's 2025 Truck Volume Manufacturer Chrysler/Fiat NEW SEGMENT Large Pickup Original 2025 Volume 382,492 Truck Multiplier 0.60 2025 Volume after Multiplier 231,131 After correcting for Class 2b vehicles being in the LMC forecast, it was time to create individual manufacturer segment multipliers to be used with the individual 2010 CAFE vehicle volumes to create projections for the future fleet. The individual manufacturer 1-46 ``````------- The Baseline and Reference Vehicle Fleets segment multipliers are created by dividing each year of the LMC forecast's individual manufacturer segment volume by the manufacturer's individual segment volume determined using 2010 CAFE data. Table 1-31 has the 2010 CAFE Volume, the 2025 LMC large pickup volume after Class 2b vehicles were removed, and the individual manufacturer volume for large pickup trucks. The multiplier is the result of dividing the 2025 volume by the 2010 volume. Table 1-31 Example Values Used to Determine Chrysler/Fiat 2025 Individual Large Pickup Multiplier Manufacturer Chrysler/Fiat NEW SEGMENT Large Pickup 2010 Cafe Volume 120,645 2025 Volume after Multiplier 231,131 Fiat/Chrysler Individual Large Pickup Multiplierfor2025 192% Now that the individual manufacturer segment multipliers are calculated, they can be applied to each vehicle in the 2010 CAFE data. The segment multipliers are applied by multiplying the 2010 CAFE volume for a vehicle by the multiplier for its manufacturer and segment. Table 1-32 shows the 2010 CAFE volumes, the individual manufacturer segment multipliers, and the result of multiplying the multiplier and the volume for 2025 project volumes for many of Chrysler/Fiat's large pickup trucks. Table 1-32 Example Applying the Individual Large Pickup Multiplier for Chrysler/Fiat Manufacturer Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Model Ram 1500 Pickup 2wd Ram 1500 Pickup 2wd Ram 1500 Pickup 2wd Ram 1500 Pickup 2wd Ram 1500 Pickup 2wd Ram 1500 Pickup 4wd Ram 1500 Pickup 4wd Ram 1500 Pickup 4wd Ram 1500 Pickup 4wd NEW SEGMENT Large Pickup Large Pickup Large Pickup Large Pickup Large Pickup Large Pickup Large Pickup Large Pickup Large Pickup 2010 CAFE Volume 23,686 938 3,029 16,505 7,698 1,162 51,417 15,498 712 Fiat/Chrysler Individual Large Pickup Multiplier for 2025 192% 192% 192% 192% 192% 192% 192% 192% 192% 2025 Project Volume Before AEO Normalization 45,377 1,797 5,803 31,620 14,748 2,226 98,504 29,691 1,364 Normalizing to AEO forecast for cars and trucks must be done once the individual manufacturer segment multipliers have been applied to all vehicles across every year (2011- 2025) of the LMC forecast. In order to normalize a year, the number of trucks and the number of cars produced must be determined. Then, the truck and car totals from AEO are used to determine a normalizing multiplier. Table 1-33 has the 2025 car and truck totals 1-47 ``````------- The Baseline and Reference Vehicle Fleets before normalization, the 2025 AEO car and truck total, and the multipliers which are the result of dividing the AEO totals by totals before normalization. Table 1-33 Example 2025 AEO Truck and Car Multipliers Vehicle Type Trucks Cars 2025 Total before Normalization 8,242,936 8,954,382 2025 AEO Total 6,929,100 9,525,700 AEO 2025 Normalizing Multiplier 84% 106% The final step in creating the reference volumes is applying the AEO multipliers. The AEO multipliers are applied by vehicle type. Table 1-34 shows the normalized volume, the AEO 2025 truck multiplier, and the final resulting volume for a number of Chrysler/Fiat pickups. Table 1-34 Example Applying the AEO Truck Multiplier to Chrysler/Fiat Pickups Manufacturer Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Chrysler/Fiat Model Ram 1500 Pickup 2wd Ram 1500 Pickup 2wd Ram 1500 Pickup 2wd Ram 1500 Pickup 2wd Ram 1500 Pickup 2wd Ram 1500 Pickup 4wd Ram 1500 Pickup 4wd Ram 1500 Pickup 4wd Ram 1500 Pickup 4wd Vehicle Type Truck Truck Truck Truck Truck Truck Truck Truck Truck 2025 Project Volume Before AEO Normalization 45,377 1,797 5,803 31,620 14,748 2,226 98,504 29,691 1,364 AEO 2025 Truck Multiplier 84% 84% 84% 84% 84% 84% 84% 84% 84% 2025 Project Volume with AEO Normalization 38,145 1,511 4,878 26,580 12,397 1,871 82,804 24,959 1,147 1.4.3 What are the sales volumes and characteristics of the MY 2010 based reference fleet? Table 1-35 and Table 1-37 below contain the sales volumes that result from the process above for MY 2010 and 2017-2020. Table 1-36 and Table 1-38 below contain the sales volumes that result from the process above for MY 2021-2025. Table 1-35 Vehicle Segment Volumes" Reference Class Segment Large Auto Actual and Projected Sales Volume 2010 393,049 2017 567,514 2018 579,808 2019 598,784 2020 617,135 1-48 ``````------- The Baseline and Reference Vehicle Fleets Mid-Size Auto Compact Auto Sub-Compact Auto Large Pickup Small Pickup Large SUV Mid-Size SUV Small SUV Mini Van Cargo Van 2,189,552 1,894,017 1,615,536 1,201,518 74,780 2,066,629 1,058,340 113,716 565,527 17,516 3,446,643 2,561,669 2,258,243 1,747,062 39,095 3,259,969 1,068,111 148,142 686,492 29,160 3,413,476 2,525,760 2,231,633 1,723,045 39,793 3,208,284 1,036,455 143,413 674,803 28,929 3,523,692 2,524,658 2,161,935 1,773,581 49,185 3,157,778 1,058,492 142,957 641,731 29,308 3,577,767 2,537,591 2,169,551 1,757,204 55,481 3,086,726 1,037,464 142,894 618,567 29,821 a Volumes in this table are based on the pre-2011 NHTSA definition of Cars and Trucks. Table 1-36 Vehicle Segment Volumes" Reference Class Segment Large Auto Mid-Size Auto Compact Auto Sub-Compact Auto Large Pickup Small Pickup Large SUV Mid-Size SUV Small SUV Mini Van Cargo Van Projected Sales Volume 2021 627,571 3,644,746 2,571,913 2,188,554 1,759,426 58,848 3,067,335 1,026,207 143,576 612,054 29,868 2022 641,252 3,684,993 2,613,050 2,236,339 1,761,341 62,556 3,064,546 1,040,034 145,165 607,502 30,422 2023 657,367 3,763,193 2,649,239 2,256,403 1,763,299 66,735 3,043,294 1,031,240 146,476 599,255 30,699 2024 665,152 3,819,396 2,709,562 2,334,855 1,770,423 71,587 3,049,618 1,047,527 148,201 600,002 30,678 2025 678,652 3,902,811 2,750,233 2,359,545 1,787,445 75,596 3,064,625 1,052,812 152,103 599,779 31,198 ' Volumes in this table are based on the pre-2011 NHTSA definition of Cars and Trucks. Table 1-37 2011+ NHTSA Car and Truck Definition Based Volumes Vehicle Type Cars Trucks Cars and Trucks Actual and Projected Sales Volume 2010 7,176,330 4,013,850 11,190,180 2017 10,213,312 5,598,788 15,812,100 2018 10,088,966 5,516,434 15,605,400 2019 10,139,761 5,522,339 15,662,100 2020 10,194,353 5,435,847 15,630,200 1-49 ``````------- The Baseline and Reference Vehicle Fleets Table 1-38 2011+ NHTSA Car and Truck Definition Based Volumes Vehicle Type Cars Trucks Cars and Trucks Projected Sales Volume 2021 10,310,594 5,419,506 15,730,100 2022 10,455,061 5,432,139 15,887,200 2023 10,593,727 5,413,473 16,007,200 2024 10,811,530 5,435,470 16,247,000 2025 10,981,082 5,473,718 16,454,800 Table 1-40 and Table 1-40 below contain the sales volumes by manufacturer and vehicle type for MY 2010 and 2017-2025. Tesla did not report any vehicle sales in 2010 so their projected volume is zero. Spyker/Saab sold no vehicles under the Spyker brand in 2010 so their volume is also zero. Table 1-39 NHTSA Car and Truck Definition Manufacturer Volumes Manufacturers All All All Aston Martin Aston Martin BMW BMW Chrysler/Fiat Chrysler/Fiat Daimler Daimler Ferrari Ferrari Ford Ford Geely Geely General Motors General Motors Honda Honda Hyundai Hyundai Kia Vehicl e Type Both Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars 2010 Baseline Sales 11,190,180 7,176,330 4,013,850 601 - 143,638 26,788 496,998 665,806 157,453 72,393 1,780 - 940,241 858,798 28,223 29,719 1,010,524 735,367 845,318 390,028 375,656 35,360 226,157 2017 Projected Volume 11,190,180 10,213,312 5,598,788 634 - 320,634 106,150 728,817 774,065 252,820 99,125 1,878 - 1,348,543 1,035,400 60,422 35,087 1,652,946 1,213,192 1,122,558 536,998 865,069 131,912 345,314 2018 Projected Volume 15,605,400 10,088,966 5,516,434 617 - 318,821 104,625 736,022 743,375 240,222 108,510 1,828 - 1,347,544 1,023,955 57,655 32,438 1,616,449 1,201,479 1,139,856 525,327 849,727 127,289 339,180 2019 Projected Volume 15,662,100 10,139,761 5,522,339 620 - 327,091 105,104 769,256 749,206 245,807 108,294 1,836 - 1,341,628 1,016,328 60,338 33,299 1,611,415 1,217,167 1,147,055 527,814 857,497 122,193 328,872 2020 Projected Volume 15,630,200 10,194,353 5,435,847 620 - 329,304 101,805 786,344 740,640 245,888 108,598 1,837 - 1,347,596 995,702 60,040 32,149 1,612,666 1,211,435 1,167,627 517,268 861,062 118,265 327,694 1-50 ``````------- The Baseline and Reference Vehicle Fleets Kia Lotus Lotus Mazda Mazda Mitsubishi Mitsubishi Nissan Nissan Porsche Porsche Spyker/Saab Spyker/Saab Subaru Subaru Suzuki Suzuki Tata/JLR Tata/JLR Tesla Tesla Toyota Toyota Volkswagen Volkswagen Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 21,721 354 - 249,489 61,451 54,263 9,146 619,918 255,566 11,937 3,978 - - 184,587 73,665 25,002 3,938 11,279 37,475 - - 1,508,866 696,324 284,046 36,327 43,374 374 - 254,270 59,862 61,058 13,701 889,039 305,943 18,430 20,105 - - 209,137 96,938 43,253 3,399 28,012 54,033 - - 1,528,208 966,417 481,894 103,088 43,209 364 - 249,048 59,114 58,152 13,840 867,771 306,537 18,138 19,647 - - 205,550 94,441 42,515 3,347 27,188 53,423 - - 1,501,492 955,281 470,826 100,596 41,648 365 - 247,203 55,108 60,387 14,276 873,076 309,179 17,255 19,573 - - 205,868 92,177 43,399 3,690 28,194 52,682 - - 1,509,270 951,691 463,329 102,910 40,270 365 - 248,350 53,334 60,619 14,262 874,098 304,196 17,065 18,851 - - 205,749 90,751 44,081 3,676 28,430 51,461 - - 1,515,051 932,267 459,868 100,916 Table 1-40 NHTSA Car and Truck Definition Manufacturer Volumes Manufacturers All All All Aston Martin Aston Martin BMW BMW Chrysler/Fiat Chrysler/Fiat Daimler Daimler Ferrari Ferrari Ford Ford Vehicle Type Both Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 2021 Projected Volume 15,730,100 10,310,594 5,419,506 623 - 335,753 101,238 805,113 733,257 249,219 110,235 1,845 - 1,359,990 990,243 2022 Projected Volume 15,887,200 10,455,061 5,432,139 626 - 341,613 100,345 828,656 735,937 251,461 112,133 1,853 - 1,377,947 990,827 2023 Projected Volume 16,007,200 10,593,727 5,413,473 630 - 346,903 99,084 850,402 731,269 253,688 113,550 1,865 - 1,394,907 985,782 2024 Projected Volume 16,247,000 10,811,530 5,435,470 634 - 357,948 101,174 877,751 722,213 258,742 116,867 1,878 - 1,418,568 991,767 2025 Projected Volume 16,454,800 10,981,082 5,473,718 639 - 363,380 101,013 899,843 726,403 261 ,242 119,090 1,894 - 1,441,350 997,694 1-51 ``````------- The Baseline and Reference Vehicle Fleets Geely Geely General Motors General Motors Honda Honda Hyundai Hyundai Kia Kia Lotus Lotus Mazda Mazda Mitsubishi Mitsubishi Nissan Nissan Porsche Porsche Spyker Spyker Subaru Subaru Suzuki Suzuki Tata/JLR Tata/JLR Tesla Tesla Toyota Toyota Volkswagen Volkswagen Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 61,433 31,977 1,624,561 1,218,265 1,187,756 512,800 873,625 117,565 330,416 39,205 367 - 249,288 52,946 61,785 14,307 879,450 303,616 17,289 18,863 - - 206,863 91,673 44,765 3,760 28,977 50,984 - - 1,530,699 927,227 460,777 101,344 62,399 31,598 1,638,066 1,226,184 1,212,900 515,656 887,004 116,208 335,846 38,857 368 - 252,522 52,752 63,390 14,778 884,816 304,381 17,216 18,598 - - 209,828 91,940 45,769 3,879 29,416 50,767 - - 1,548,354 925,277 465,011 102,022 63,076 31,007 1,652,324 1,232,502 1,238,278 509,628 899,936 115,339 338,791 38,203 371 - 254,751 52,158 63,937 14,824 893,622 304,703 17,292 18,562 - - 211,621 92,337 46,590 3,939 29,898 50,280 - - 1,567,676 918,749 467,170 101,558 65,157 31,796 1,676,558 1,244,178 1,267,745 505,534 918,938 116,430 346,828 38,034 374 - 259,488 52,998 67,026 15,229 907,823 308,510 17,517 18,861 - - 215,567 94,300 47,824 4,085 30,546 50,340 - - 1,598,715 918,479 475,903 104,673 65,883 31,528 1,696,474 1,261,546 1,295,234 504,020 935,619 117,662 350,765 37,957 377 - 262,732 53,183 67,925 15,464 919,920 312,005 17,609 19,091 - - 218,870 96,326 48,710 4,173 30,949 50,369 - - 1 ,622,242 921,183 479,423 105,009 Table 1-41 also shows how the change in fleet make-up may affect the footprint distributions over time. The resulting data indicate that footprint will not change significantly between 2010 and 2025. The footprints are somewhat larger than in the 2008 based fleet projection (Table 1-19). Table 1-41 Production Weighted Foot Print Mean Model Average Footprint of all Average Footprint Average Footprint 1-52 ``````------- The Baseline and Reference Vehicle Fleets Year 2010 2017 2018 2019 2020 2021 2022 2023 2024 2025 Vehicles 48.6 48.7 48.8 48.8 48.8 48.8 48.7 48.7 48.6 48.6 Cars 45.2 45.4 45.4 45.5 45.5 45.5 45.5 45.5 45.5 45.5 Trucks 54.5 54.9 54.9 55.0 55.0 55.0 55.0 55.0 54.9 55.0 Table 1-42 below shows the changes in engine cylinders over the model years. The current assumptions show that engines will increase in size between 2010 and 2017 and then remain relatively constant over the model years to which these final rules apply. Table 1-42 Percentages of 4,6,8 Cylinder Engines by Model Year Model Year 2010 2017 2018 2019 2020 2021 2022 2023 2024 2025 Trucks 4 Cylinders 15.7% 13.9% 13.7% 13.6% 13.5% 13.4% 13.5% 13.5% 13.7% 13.6% 6 Cylinders 52.5% 50.2% 50.3% 50.0% 49.9% 49.8% 49.7% 49.6% 49.6% 49.5% 8 Cylinders 31.8% 35.9% 36.0% 36.4% 36.7% 36.8% 36.8% 36.9% 36.8% 36.8% Cars 4 Cylinders 69.2% 66.3% 66.2% 65.7% 65.7% 65.7% 65.8% 65.7% 65.9% 65.9% 6 Cylinders 26.6% 29.0% 29.1% 29.6% 29.6% 29.6% 29.5% 29.5% 29.4% 29.4% 8 Cylinders 4.1% 4.7% 4.7% 4.7% 4.7% 4.7% 4.8% 4.8% 4.7% 4.8% 1-53 ``````------- The Baseline and Reference Vehicle Fleets 1.5 What are the differences in the sales volumes and characteristics of the MY 2008 based and the MY 2010 based reference fleets? This section compares some of the differences between the fleet based on MY 2008 CAFE and the fleet based on MY 2010 CAFE data. As stated before, the 2008 fleet projection is based on MY 2008 CAFE data, a long range forecast provided by CSM, and interim AEO 2011. The 2010 fleet projection is based on MY 2010 CAFE, a long range forecast provided by LMC Automotive, and interim AEO 2012. Table 1-43, Table 1-44, Table 1-45 and Table 1-46 below contain the sales volume differences between the two fleets, from subtracting the 2008 MY based fleet projection from the 2010 MY based fleet projection. The sales in MY 2010 are significantly lower (by 2,661,581 vehicles) than in MY 2008 (reflecting the continued economic recession, as noted earlier). The sales in MY 2010 are depressed but sales are expected to recover to their MY 2008 levels before 2017. There is an increase in the number of large trucks, midsize autos, and large autos by 2025. There is also decreased volume in the remaining segment in 2025. These differences are due to the LMC forecast and the newer AEO projection. Table 1-43 Vehicle Segment Volumes Differences" Reference Class Segment LargeAuto MidSizeAuto CompactAuto SubCmpctAuto LargePickup SmallPickup LargeSUV MidSizeSUV SmallSUV MiniVan CargoVan Actual Sales Volume 2010-2008 -169,191 -909,375 -85,444 249,703 -380,708 -102,717 -717,320 -205,020 -171,639 -76,528 -93,342 Projected Sales Volume 2017 191,407 135,375 213,689 -199,979 232,443 -117,132 65,480 -290,644 -109 -68,070 -156,681 2018 223,040 123,068 200,367 -222,479 279,279 -118,139 58,183 -272,757 -6,520 -64,748 -170,305 2019 245,175 220,071 155,357 -327,273 390,391 -111,567 -20,090 -208,902 -11,718 -75,334 -172,666 2020 222,271 195,982 89,570 -383,799 371,009 -90,548 -116,518 -248,358 -19,783 -95,756 -189,807 a Volumes in this table are based on the pre-2011 NHTSA definition of Cars and Trucks. 1-54 ``````------- The Baseline and Reference Vehicle Fleets Table 1-44 Vehicle Segment Volumes Differences" Reference Class Segment LargeAuto MidSizeAuto CompactAuto SubCmpctAuto LargePickup SmallPickup LargeSUV MidSizeSUV SmallSUV MiniVan CargoVan Projected Sales Volume 2021 247,379 202,630 50,936 -437,810 391,125 -91,275 -245,579 -255,033 -23,647 -117,024 -180,671 2022 282,957 136,730 20,851 -450,828 411,920 -84,582 -298,062 -243,210 -24,478 -131,480 -172,390 2023 294,695 70,660 16,313 -464,699 462,006 -84,580 -369,459 -237,048 -23,763 -141,530 -170,886 2024 308,979 67,900 -35,072 -461,206 498,672 -83,040 -426,255 -245,135 -24,990 -120,718 -166,222 2025 309,809 87,870 -92,836 -518,743 527,056 -79,242 -456,367 -252,550 -23,610 -126,477 -170,570 a Volumes in this table are based on the pre-2011 NHTSA definition of Cars and Trucks. Table 1-45 2011+ NHTSA Car and Truck Definition Based Volumes Differences Vehicle Type Cars Trucks Cars and Trucks Actual Sales Volume 2010-2008 -1,054,238 -1,607,343 -2,661,581 Projected Sales Volume 2017 225,645 -219,867 5,778 2018 183,602 -154,612 28,990 2019 144,065 -60,623 83,442 2020 -97,209 -168,530 -265,739 Table 1-46 2011+ NHTSA Car and Truck Definition Based Volumes Differences Vehicle Type Cars Trucks Cars and Trucks Projected Sales Volume 2021 -194,571 -264,396 -458,966 2022 -280,716 -271,857 -552,572 2023 -374,276 -274,013 -648,289 2024 -446,608 -240,479 -687,087 2025 -560,478 -235,181 -795,659 1-55 ``````------- The Baseline and Reference Vehicle Fleets Table 1-47 and Table 1-48 below contain the differences in sales volumes by manufacturer and vehicle type between the 2008 MY based fleet and the 2010 MY based fleet. Table 1-48 shows that Chrysler/Fiat cars and trucks, Ford trucks, Hyundai cars, and Porsche trucks are projected to have significant increases in volume in MY 2025, though Table 1-48 also shows the market down overall in MY 2025 by 795,659 vehicles. Table 1-47 NHTSA Car and Truck Definition Manufacturer Volumes Differences Manufacturers All All All Aston Martin Aston Martin BMW BMW Chrysler/Fiat Chrysler/Fiat Daimler Daimler Ferrari Ferrari Ford Ford Geely/Volvo Geely/Volvo GM GM HONDA HONDA HYUNDAI HYUNDAI Kia Kia Lotus Lotus Mazda Mazda Mitsubishi Mitsubishi Vehicle Type Both Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 2010-2008 Difference in Sales -2,661,581 -1,054,238 -1,607,343 -769 NA -148,158 -34,536 -206,160 -290,986 -50,742 -6,742 330 NA -16,458 44,604 -4,525 -35,930 -497,273 -852,024 -161,321 -115,112 37,787 -17,798 4,177 -37,751 102 NA 2,828 5,566 -31,095 -6,225 2017 Difference in Volume -4,616,142 225,645 -219,867 -401 NA 7,612 -31,903 310,054 364,363 -32,027 12,212 -4,798 NA 48,644 271,851 18,535 -53,147 290,185 -249,012 -32,042 -59,483 273,042 -20,973 23,270 -55,328 134 NA 730 8,074 -4,041 -23,931 2018 Difference in Volume 28,990 183,602 -154,612 -434 NA -4,118 -27,317 338,484 355,517 -36,187 24,859 -4,872 NA 36,077 275,126 15,468 -56,956 178,094 -272,597 1,769 -19,292 271,354 -24,172 26,810 -55,071 121 NA -13,464 1,579 -5,519 -22,460 2019 Difference in Volume 83,442 144,065 -60,623 -452 NA -18,984 -26,269 377,567 382,759 -35,618 20,106 -4,958 NA 9,589 298,555 17,213 -58,276 106,390 -276,344 2,416 279 274,526 -33,449 13,993 -59,031 115 NA -19,748 -2,386 -3,439 -21,178 2020 Difference in Volume -265,739 -97,209 -168,530 -414 NA -28,638 -26,534 371,025 379,963 -45,101 15,679 -5,079 NA -31,193 278,665 17,425 -60,854 81,911 -333,548 3,961 -7,821 262,779 -35,908 4,018 -56,265 99 NA -21,728 -4,820 -4,461 -20,953 1-56 ``````------- The Baseline and Reference Vehicle Fleets Nissan Nissan PORSCHE PORSCHE Spyker/Saab Spyker/Saab Subaru Subaru Suzuki Suzuki Tata/JLR Tata/JLR Tesla Tesla Toyota Toyota Volkswagen Volkswagen Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks -97,951 -49,980 -6,972 -14,819 NA NA 68,552 -8,881 -54,337 -31,381 1,683 -18,109 NA NA 248,502 -254,812 -7,437 9,328 18,242 -138,995 -16,663 6,872 NA NA -14,975 18,696 -47,455 -18,710 -27,869 -3,546 NA NA -320,988 -364,094 -69,744 -25,731 18,093 -105,846 -17,306 7,646 NA NA -11,048 19,289 -47,417 -18,038 -29,034 -3,183 NA NA -332,689 -268,134 -69,210 -44,895 18,676 -89,380 -18,861 8,104 NA NA -11,227 19,345 -47,169 -17,002 -29,073 -5,172 NA NA -327,036 -190,413 -73,785 -43,981 -8,693 -93,673 -18,898 7,710 NA NA -17,717 18,293 -49,467 -16,999 -29,752 -4,752 NA NA -368,683 -222,037 -94,954 -45,784 Table 1-48 NHTSA Car and Truck Definition Manufacturer Volumes Differences Manufacturers All All All Aston Martin Aston Martin BMW BMW Chrysler/Fiat Chrysler/Fiat Daimler Daimler Ferrari Ferrari Ford Ford Geely/JLR Geely/JLR GM GM Vehicle Type Both Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks 2021 Difference in Volume -458,966 -194,571 -264,396 -435 NA -23,345 -27,486 384,100 384,644 -51,159 10,786 -5,214 NA -41,627 276,062 19,665 -60,749 94,541 -346,012 2022 Difference in Volume -552,572 -280,716 -271,857 -423 NA -18,421 -28,554 404,483 372,929 -53,277 11,198 -5,285 NA -37,274 276,561 20,713 -60,914 130,413 -352,372 2023 Difference in Volume -648,289 -374,276 -274,013 -411 NA -13,658 -28,437 426,520 370,205 -58,819 8,235 -5,362 NA -79,890 285,777 21,045 -65,833 155,505 -373,993 2024 Difference in Volume -687,087 -446,608 -240,479 -507 NA -30,245 -45,351 451,734 377,251 -73,595 9,783 -5,563 NA -85,102 302,913 22,696 -67,385 182,961 -392,627 2025 Difference in Volume -795,659 -560,478 -235,181 -543 NA -41,876 -44,396 463,364 394,641 -79,477 18,023 -5,764 NA -98,759 313,218 23,295 -69,579 172,466 -412,390 1-57 ``````------- The Baseline and Reference Vehicle Fleets HONDA HONDA HYUNDAI HYUNDAI Kia Kia Lotus Lotus Mazda Mazda Mitsubishi Mitsubishi Nissan Nissan PORSCHE PORSCHE Spyker/Saab Spyker/Saab Subaru Subaru Suzuki Suzuki Tata/JLR Tata/JLR Tesla Tesla Toyota Toyota Volkswagen Volkswagen Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks Cars Trucks -11,124 -23,116 260,270 -38,901 -903 -56,227 89 NA -25,452 -6,281 -4,066 -21,002 -33,179 -104,413 -19,186 7,621 NA NA -23,917 18,900 -50,960 -17,007 -29,700 -7,169 NA NA -373,007 -288,312 -124,830 -47,390 -24,604 -23,579 259,040 -41,285 -3,256 -55,837 78 NA -28,628 -7,555 -3,871 -20,449 -52,631 -107,502 -19,391 7,213 NA NA -28,785 19,204 -51,830 -16,855 -29,933 -7,823 NA NA -437,723 -309,775 -128,303 -44,728 -27,286 -27,270 265,628 -45,850 -3,955 -57,485 72 NA -42,159 -9,808 -3,743 -20,645 -60,718 -112,418 -19,701 7,192 NA NA -29,991 19,315 -52,673 -16,864 -30,741 -8,585 NA NA -469,316 -306,231 -129,579 -52,369 -40,106 -31,460 261,228 -49,662 -5,054 -58,085 66 NA -41,126 -8,973 -3,702 -20,772 -74,948 -113,707 -21,987 7,452 NA NA -32,716 20,158 -52,623 -17,077 -33,182 -7,641 NA NA -481,813 -289,534 -129,433 -52,266 -45,087 -53,677 258,369 -50,474 -12,018 -59,696 61 NA -44,072 -8,185 -5,380 -20,923 -94,855 -114,449 -23,087 7,872 NA NA -38,100 21,604 -54,444 -17,201 -34,469 -6,436 NA NA -485,811 -288,833 -150,740 -49,275 Table 1-49 shows the difference in footprint distributions between the 2010 based fleet projection and the 2008 based fleet projection. The differences between MYs 2010 and 2008 are small and are just the result of the manufacturers' product mix in those model years. MY 2025 shows an increase in both the average truck and average car footprints. This is due to the increased number of large cars and large trucks forecast in the 2010 based fleet projection. Also, in several MYs, the change in the average footprint of all vehicles is outside the range between the changes in the corresponding car and truck fleets. This is due to production weighting. Because the total numbers of cars and trucks differs, production weighting can affect the average for the whole fleet as compared to the averages for cars and trucks. This can cause a counterintuitive effect when taking the difference of the averages. 1-58 ``````------- The Baseline and Reference Vehicle Fleets Table 1-49 Production Weighted Foot Print Mean Difference* Model Year 2010-2008 2017 2018 2019 2020 2021 2022 2023 2024 2025 Average Footprint of all Vehicles 48.6 -48. 8 = -0.2 48.7-48.0 = 0.7 48.8-47.9 = 0.9 48.8-47.8 = 1.0 48.8-47.8 = 1.0 48.8-47.7 = 1.0 48.7-47.7 = 1.0 48.7-47.7 = 1.0 48.6-47.5 = 1.1 48.6-47.5 = 1.1 Average Footprint Cars 45.2-45.2 = 0.0 45.4-44.6 = 0.8 45.4.44.6 = 0.8 45.5-44.6 = 0.9 45.5.44.6 = 0.9 45.5-44.6 = 0.9 45.5-44.6 = 0.9 45.5-44.6 = 0.9 45.5-44.6 = 0.9 45.5-44.6 = 0.9 Average Footprint Trucks 54.5-53.9 = 0.6 54.9-53.8 = 1.1 54.9-53.7 = 1.2 55.0-53.6 = 1.4 55.0-53.7 = 1.3 55.0-53.6 = 1.4 55.0-53.6 = 1.4 55.0-53.5 = 1.5 54.9-53.3 = 1.6 55.0-53.3 = 1.7 *Note: This table is the difference calculated from Table 1-19 and Table 1-41. Table 1-50 shows the difference in engine cylinders distribution between the 2010 MY based fleet and the 2008 MY based fleet. MY 2010 has fewer vehicles with 6 cylinder engines. Fewer 6 cylinders in the baseline fleet along with vehicle mix changes results in more 4 and 8 cylinder engines in trucks and more 4 cylinder cars by 2025. Table 1-50 Differences in Percentages of 4,6,8 Cylinder Engines by Model Year Model Year 2010-2008 2017 2018 2019 2020 2021 2022 2023 2024 2025 Trucks 4 Cylinders 5.40% 3.00% 3.10% 3.20% 3.20% 3.10% 3.20% 3.20% 3.20% 3.10% 6 Cylinders -3.90% -13.50% -14.20% -15.50% -15.70% -16.50% -17.00% -18.10% -18.50% -18.70% 8 Cylinders -1.50% 10.50% 11.20% 12.30% 12.60% 13.40% 13.80% 14.90% 15.40% 15.50% Cars 4 Cylinders 12.30% 5.70% 5.50% 5.00% 5.40% 5.10% 4.70% 4.80% 4.90% 4.80% 6 Cylinders -11.20% -5.50% -5.30% -4.70% -5.10% -4.80% -4.70% -4.80% -4.70% -4.60% 8 Cylinders -1.20% -0.30% -0.30% -0.30% -0.30% -0.20% 0.00% 0.00% -0.10% 0.00% 1-59 ``````------- The Baseline and Reference Vehicle Fleets References: 1 EPA's Omega Model and input sheets are available at http://www.epa.gov/oms/climate/models.htm; DOT/NHTSA's CAFE Compliance and Effects Modeling System (commonly known as the "Volpe Model") and input and output sheets are available at http://www.nhtsa.gov/fuel-economy. It is also available in the docket (Docket EPA-HQ-O AR-2010-0799) 9 _ http://www.nhtsa.gov/Laws+&+Regulations/CAFE+- +Fuel+Economy/CAFE+Compliance+and+Effects+Modeling+System:+The+Volpe+Model 3 Department of Energy, Energy Information Administration, Annual Energy Outlook (AEO) 2011, Early Release. Available at http://www.eia.gov/forecasts/aeo/. 4 The baseline Excel file ("2008-2025 Production Summary Data _Definitions Docket 08_27_2009") is available in the docket (Docket EPA-HQ-O AR-2010-0799). 5 http://www.nhtsa.gov/Laws+&+Regulations/CAFE+- +Fuel+Economy/CAFE+Compliance+and+Effects+Modeling+System:+The+Volpe+Model 6 Department of Energy, Energy Information Administration, Annual Energy Outlook (AEO) 20112012, Early Release. Available at http://www.eia.gov/forecasts/aeo/ (last accessed Aug. 15, 2011 April 9, 2012). 7 The baseline Excel file ("2010-2025 Production Summary Data_Defmitions Docket 05.01.2012") is available in the docket (Docket EPA-HQ-O AR-2010-0799). 1-60 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting, and How Were They Developed? 2.1 Why are standards attribute-based and defined by a mathematical function? As in the MYs 2012-2016 CAFE/GHG rules, and as NHTSA did in the MY 2011 CAFE rule, NHTSA and EPA are promulgating attribute-based CAFE and CC>2 standards that are defined by a mathematical function. EPCA, as amended by EISA, expressly requires that CAFE standards for passenger cars and light trucks be based on one or more vehicle attributes related to fuel economy, and be expressed in the form of a mathematical function.1 The CAA has no such requirement, although such an approach is permissible under section 202 (a) and EPA has used the attribute-based approach in issuing standards under both section 202 (a) and under analogous provisions of the CAA (e.g., criteria pollutant standards for non-road diesel engines using engine size as the attribute,2 in the recent GHG standards for heavy duty pickups and vans using a work factor attribute,3 and in the MYs 2012-2016 GHG rule which used vehicle footprint as the attribute). Public comments on the MYs 2012-2016 rulemaking widely supported attribute-based standards for both agencies' standards. Comments received on the MY 2017 and later proposal also generally supported an attribute-based standard, as further discussed in section 2.2. Under an attribute-based standard, every vehicle model has a performance target (fuel economy and CC>2 emissions for CAFE and CC>2 emissions standards, respectively), the level of which depends on the vehicle's attribute (for this rule, footprint, as discussed below). The manufacturers' fleet average performance is determined by the production-weighted51 average (for CAFE, harmonic average) of those targets. The agencies believe that an attribute-based standard is preferable to a single-industry- wide average standard in the context of CAFE and CC>2 standards for several reasons. First, if the shape is chosen properly, every manufacturer is more likely to be required to continue adding more fuel efficient technology each year across their fleet, because the stringency of the compliance obligation will depend on the particular product mix of each manufacturer. Therefore a maximum feasible attribute-based standard will tend to require greater fuel savings and CO2 emissions reductions overall than would a maximum feasible flat standard (that is, a single mpg or CC>2 level applicable to every manufacturer). Second, depending on the attribute, attribute-based standards reduce the incentive for manufacturers to respond to CAFE and CC>2 standards in ways harmful to safety. Because a Production for sale in the United States. b The 2002 NAS Report described at length and quantified the potential safety problem with average fuel economy standards that specify a single numerical requirement for the entire industry. See 2002 NAS Report at 5, finding 12. Ensuing analyses, including by NHTSA, support the fundamental conclusion that standards structured to minimize incentives to downsize all but the largest vehicles will tend to produce better safety outcomes than flat standards. 2-1 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting each vehicle model has its own target (based on the attribute chosen), properly fitted attribute- based standards provide little, if any, incentive to build smaller vehicles simply to meet a fleet-wide average, because the smaller vehicles will be subject to more stringent compliance targets.0 Third, attribute-based standards provide a more equitable regulatory framework for different vehicle manufacturers.d A single industry-wide average standard imposes disproportionate cost burdens and compliance difficulties on the manufacturers that need to change their product plans to meet the standards, and puts no obligation on those manufacturers that have no need to change their plans. As discussed above, attribute-based standards help to spread the regulatory cost burden for fuel economy more broadly across all of the vehicle manufacturers within the industry. Fourth, attribute-based standards better respect economic conditions and consumer choice, as compared to single-value standards. A flat, or single value, standard encourages a certain vehicle size fleet mix by creating incentives for manufacturers to use vehicle downsizing as a compliance strategy. Under a footprint-based standard, manufacturers have greater incentive (compared to under a flat standard) to invest in technologies that improve the fuel economy of the vehicles they sell rather than shifting product mix, because reducing the size of the vehicle is generally a less viable compliance strategy given that smaller vehicles have more stringent regulatory targets. 2.2 What attribute are the agencies adopting, and why? As in the MYs 2012-2016 CAFE/GHG rules, and as NHTSA did in the MY 2011 CAFE rule, NHTSA and EPA are promulgating CAFE and CC>2 standards that are based on vehicle footprint, which has an observable correlation to fuel economy and emissions. There are several policy and technical reasons why NHTSA and EPA believe that footprint is the most appropriate attribute on which to base the standards for the vehicles covered by this rulemaking, even though some other light-duty vehicle attributes (notably curb weight) are better correlated to fuel economy and emissions. First, in the agencies' judgment, from the standpoint of vehicle safety, it is important that the CAFE and CC>2 standards be set in a way that does not encourage manufacturers to respond by selling vehicles that are less safe. NHTSA's research of historical crash data has found that reductions in vehicle size and reductions in the mass of lighter vehicles tend to compromise overall highway safety, while reductions in the mass of heavier vehicles tend to improve overall highway safety. If footprint-based standards are defined in a way that creates relatively uniform burden for compliance for vehicles of all sizes, then footprint-based standards will not incentivize manufacturers to downsize their fleets as a strategy for 0 Assuming that the attribute is related to vehicle size. d Mat 4-5, finding 10. 2-2 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting compliance which could compromise societal safety, or to upsize their fleets which might reduce the program's fuel savings and GHG emission reduction benefits. Footprint-based standards also enable manufacturers to apply weight-efficient materials and designs to their vehicles while maintaining footprint, as an effective means to improve fuel economy and reduce GHG emissions. On the other hand, depending on their design, weight-based standards can create disincentives for manufacturers to apply weight-efficient materials and designs. This is because weight-based standards would become more stringent as vehicle mass is reduced. The agencies discuss mass reduction and its relation to safety in more detail in Preamble section II.G. Further, although we recognize that weight is better correlated with fuel economy and CC>2 emissions than is footprint, we continue to believe that there is less risk of "gaming" (changing the attribute(s) to achieve a more favorable target) by increasing footprint under footprint-based standards than by increasing vehicle mass under weight-based standards—it is relatively easy for a manufacturer to add enough weight to a vehicle to decrease its applicable fuel economy target a significant amount, as compared to increasing vehicle footprint. We also continue to agree with concerns raised in 2008 by some commenters on the MY 2011 CAFE rulemaking that there would be greater potential for gaming under multi-attribute standards, such as those that also depend on weight, torque, power, towing capability, and/or off-road capability. The agencies agree with the assessment first presented in NHTSA's MY 2011 CAFE final rule4 that the possibility of gaming an attribute-based standard is lowest with footprint-based standards, as opposed to weight-based or multi-attribute-based standards. Specifically, standards that incorporate weight, torque, power, towing capability, and/or off- road capability in addition to footprint would not only be more complex, but by providing degrees of freedom with respect to more easily-adjusted attributes, they could make it less certain that the future fleet would actually achieve the average fuel economy and CO2 reduction levels projected by the agencies.6 This is not to say that a footprint-based system will eliminate gaming, or that a footprint-based system will eliminate the possibility that manufacturers will change vehicles in ways that compromise occupant protection. In the agencies' judgment, footprint-based standards achieved the best balance among affected considerations. The agencies recognize that based on economic and consumer demand factors that are external to this rule, the distribution of footprints in the future may be different (either smaller or larger) than what is projected in this rule. However, the agencies continue to believe that there will not be significant shifts in this distribution as a direct consequence of this rule. We note that comments by CBD, ACEEE, and NACAA referenced a 2011 study by Whitefoot and Skerlos, "Design incentives to increase vehicle size created from the U.S. footprint-based fuel economy standards." This study concluded that the proposed MY 2014 standards eHowever, for heavy-duty pickups and vans not covered by today's standards, the agencies determined that use of footprint and work factor as attributes for heavy duty pickup and van GHG and fuel consumption standards could reasonably avoid excessive risk of gaming. See 76 FR 57106, 57161-62 (Sept. 15, 2011) f Available at Docket ID: EPA-HQ-OAR-2010-0799. 2-3 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting "create an incentive to increase vehicle size except when consumer preference for vehicle size is near its lower bound and preference for acceleration is near its upper bound."g The commenters who cited this study generally did so as part of arguments in favor of flatter standards (i.e., curves that are flatter across the range of footprints) for MYs 2017-2025. While the agencies consider the concept of the Whitefoot and Skerlos analysis to have some potential merits, it is also important to note that, among other things, the authors assumed different inputs than the agencies actually used in the MYs 2012-2016 rules regarding the baseline fleet, the cost and efficacy of potential future technologies, and the relationship between vehicle footprint and fuel economy. Were the agencies to use the Whitefoot and Skerlos methodology (e.g., methods to simulate manufacturers' potential decisions to increase vehicle footprint) with the actual inputs to the MYs 2012-2016 rules, the agencies would likely obtain different findings. Underlining the potential uncertainty, considering a range of scenarios, the authors obtained a wide range of results in their analyses. The agencies discuss this study more fully in the Section II of the preamble, the NHTSA RIA, and the EPA response to comments document. The agencies also recognize that some international attribute-based standards use attributes other than footprint and that there could be benefits for a number of manufacturers if there was greater international harmonization of fuel economy and GHG standards for light- duty vehicles, but this is largely a question of how stringent standards are and how they are tested and enforced. It is entirely possible that footprint-based and weight-based systems can coexist internationally and not present an undue burden for manufacturers if they are carefully crafted. Different countries or regions may find different attributes appropriate for basing standards, depending on the particular challenges they face—from fuel prices, to family size and land use, to safety concerns, to fleet composition and consumer preference, to other environmental challenges besides climate change. The agencies anticipate working more closely with other countries and regions in the future to consider how to address these issues in a way that least burdens manufacturers while respecting each country's need to meet its own particular challenges. In the proposal, the agencies found that footprint was the most appropriate attribute upon which to base the proposed standards. Recognizing strong public interest in this issue, the agencies sought comment on whether a different attribute or combination of attributes should be considered in setting standards for the final rule. The agencies specifically requested that the commenters address the concerns raised in the proposal regarding the use of other attributes, and explain how standards should be developed using the other attribute(s) in a way that contributes more to fuel savings and CO2 reductions than the footprint-based standards, without compromising safety. The agencies received several comments regarding the attribute(s) upon which new CAFE and GHG standards should be based. NADAh and the Consumer Federation of 8Ibid, page 410 h NAD A, Docket No. NHTSA-2010-0131-0261, at 11. 2-4 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting America (CFA)1 expressed support for attribute-based standards, generally, indicating that such standards accommodate consumer preferences, level the playing field between manufacturers, and remove the incentive to push consumers into smaller vehicles. Many commenters, including automobile manufacturers, NGOs, trade associations and parts suppliers (e.g.. General Motors,J Ford, American Chemistry Council, Alliance of Automobile Manufacturers,™ International Council on Clean Transportation,11 Insurance Institute for Highway Safety,0 Society of the Plastics Industry,15 Aluminum Association,11 Motor and Equipment Manufacturers Association/ and others) expressed support for the continued use of vehicle footprint as the attribute upon which to base CAFE and CO2 standards, citing advantages similar to those mentioned by NADA and CFA. Conversely, the Institute for Policy Integrity (IPI) at the New York University School of Law questioned whether non-attribute-based (flat) or an alternative attribute basis would be preferable to footprint-based standards as a means to increase benefits, improve safety, reduce "gaming," and/or equitably distribute compliance obligations.8 IPI argued that, even under flat standards, credit trading provisions would serve to level the playing field between manufacturers. IPI acknowledged that NHTSA, unlike EPA, is required to promulgate attribute-based standards, and agreed that a footprint-based system could have much less risk of gaming than a weight- based system. IPI suggested that the agencies consider a range of options, including a fuel- based system, and select the approach that maximizes net benefits. Ferrari and BMW suggested that the agencies consider weight-based standards, citing the closer correlation between fuel economy and footprint, and BMW further suggested that weight-based standards might facilitate international harmonization (i.e., between U.S. standards and related standards in other countries).1 Porsche commented that the footprint attribute is not well suited for manufacturers of high performance vehicles with a small footprint." Regarding the comments from IPI, as IPI appears to acknowledge, EPCA/EISA expressly requires that CAFE standards be attribute-based and defined in terms of mathematical functions. Also, NHTSA has, in fact, considered and reconsidered options other than footprint, over the course of multiple CAFE rulemakings conducted throughout the past decade. When first contemplating attribute-based systems, NHTSA considered attributes such as weight, "shadow" (overall area), footprint, power, torque, and towing capacity. NHTSA also considered approaches that would combine two or potentially more than two 1 CFA, Docket No. EPA-HQ-OAR-2010-0799-9419at 8, 44. J GM, Docket No. NHTSA-2010-0131-0236, at 2. kFord, Docket No. NHTSA-2010-0131-0235, at 8. 1ACC, Docket No. EPA-HQ-OAR-2010-0799-9517at 2. m Alliance, Docket No. NHTSA-2010-0131-0262, at 85. "ICCT, Docket No. NHTSA-2010-0131-0258, at 48. 0IIHS, Docket No. NHTSA-2010-0131-0222, at 1. p SPI, Docket No. EPA-HQ-OAR-2010-0799-9492, at 4. q Aluminum Association, Docket No. NHTSA-2010-0131-0226, at 1. r MEMA, Docket No. EPA-HQ-OAR-2010-0799-9478], at 1. s IPI, Docket No. EPA-HQ-OAR-2010-0799-11485at 13-15. 1 BMW, Docket No. NHTSA-2010-0131-0250, at 3. u Porsche, EPA-HQ-OAR-2010-0799-9264 2-5 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting such attributes. To date, every time NHTSA (more recently, with EPA) has reconsidered options, the agency has concluded that a properly designed footprint-based approach provides the best means of achieving the basic policy goals (i.e., better balancing compliance burdens among full-line and limited-line manufacturers and reducing incentives for manufacturers to respond to standards by reducing vehicle size in ways that could compromise overall highway safety) involved in applying an attribute-based standards, and at the same time structuring footprint-based standards in a way that furthers the energy and environmental policy goals of EPCA and the CAA by controlling incentives to increase vehicle size in ways that could increase fuel consumption and GHG emissions/ In response to IPI's suggestion to use fuel- based standards as a type of attribute, although neither NHTSA nor EPA have presented quantitative analysis of standards that differentiate between fuel type for light-duty vehicles, such standards would effectively use fuel type to identify different subclasses of vehicles, thus requiring mathematical functions—not addressed by IPI's comments—to recombine these fuel types into regulated classes.w Insofar as EPCA/EISA already specifies how different fuel types are to be treated for purposes of calculating fuel economy and CAFE levels, and moreover, insofar as the EISA revisions to EPCA removed NHTSA's previously-clear authority to set separate CAFE standards for different classes of light trucks, using fuel type to further differentiate subclasses of vehicles could conflict with the intent, and possibly the letter, of NHTSA's governing statute. Finally, in the agencies' judgment, while regarding IPI's suggestion that the agencies select the attribute-based approach that maximizes net benefits may have merit, net benefits are but one of many considerations which lead to the setting of the standard. Also, such an undertaking would be impracticable at this time, considering that the mathematical forms applied under each attribute-based approach would also need to be specified, and that the agencies lack methods to reliably quantify the relative potential for induced changes in vehicle attributes. Regarding Ferrari's and BMW's comments, as stated previously, in the agencies' judgment, footprint-based standards (a) discourage vehicle downsizing that might compromise occupant protection, (b) encourage the application of technology, including weight-efficient materials (e.g., high-strength steel, aluminum, magnesium, composites, etc.), and (c) are less susceptible than standards based on other attributes to "gaming" that could lead to less-than-projected energy and environmental benefits. It is also important to note that there are many differences between both the standards and the on-road light-duty vehicle v See 71 FR 17566, at!7595-17596 (April 6, 2006); 74 FR 14196, at!4359 (March 30, 2009); 75 FR 25324 at 25333 (May 7, 2010). w The agencies did adopt separate standards for gasoline and diesel heavy-duty pickups and vans based on technological differences between gasoline and diesel engines. See 76 FR at 57163-65. However, the agencies stated that "standards that do not distinguish between fuel types are generally preferable where technological and market-based reasons do not strongly argue otherwise. These technological differences exist presently between gasoline and diesel engines for GHGs ... The agencies emphasize, however, that they are not committed to perpetuating separate GHG standards for gasoline and diesel heavy-duty vehicles and engines, and expect to reexamine the need for separate gasoline/diesel standards in the next rulemaking." 76 FR at 57165. IPI did not suggest that there were any such technological distinctions justifying separate fuel-based attributes for light duty vehicles, and the agencies note that EPCA/EISA already specifies how different fuels are to be treated for purposes of CAFE 2-6 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting fleets in Europe and the United States. The stringency of standards, independent of the attribute used, is another factor that influences harmonization. While the agencies agree that international harmonization of test procedures, calculation methods, and/or standards could be a laudable goal, again, harmonization is not simply a function of the attribute upon which the standards are based. Given the differences in the on-road fleet (including vehicle classification and use), in fuel composition and availability, in regional consumer preferences for different vehicle characteristics, in other vehicle regulations besides for fuel economy/CO2 emissions, it would not necessarily be expected that the CAFE and GHG emission standards would align with standards of other countries. Thus, the agencies continue to judge vehicle footprint to be a preferable attribute for the same reasons enumerated in the proposal and reiterated above. Finally, as explained in section III.B.6 and documented in section III.D.6 below, EPA agrees with Porsche that the MY 2017 GHG standards, and the GHG standards for the immediately succeeding model years, pose special challenges of feasibility and (especially) lead time for intermediate volume manufacturers, in particular for limited-line manufacturers of smaller footprint, high performance passenger cars. It is for this reason that EPA has provided additional lead time to these manufacturers. NHTSA, however, is providing no such additional lead time. Under EISA/EPCA, manufacturers continue—as since the 1970s—to have the option of paying civil penalties in lieu of achieving compliance with the standards. 2.3 What mathematical functions have the agencies previously used, and why? 2.3.1 NHTSA in MY 2008 and MY 2011 CAFE (constrained logistic) For the MY 2011 CAFE rule, NHTSA estimated fuel economy levels after normalization for differences in technology, but did not make adjustments to reflect other vehicle attributes (e.g., power-to-weight ratios).x Starting with the technology adjusted passenger car and light truck fleets, NHTSA used minimum absolute deviation (MAD) regression without sales weighting to fit a logistic form as a starting point to develop mathematical functions defining the standards. NHTSA then identified footprints at which to apply minimum and maximum values (rather than letting the standards extend without limit) and transposed these functions vertically (i.e., on a gpm basis, uniformly downward) to produce the promulgated standards. In the preceding rule, for MYs 2008-2011 light truck standards, NHTSA examined a range of potential functional forms, and concluded that, compared to other considered forms, the constrained logistic form provided the expected and appropriate trend (decreasing fuel economy as footprint increases), but avoided creating "kinks" the agency was concerned would provide distortionary incentives for vehicles with neighboring footprints/ xSee 74 FR 14196, 14363-14370 (Mar. 30, 2009) for NHTSA discussion of curve fitting in the MY 2011 CAFE final rule. y See 71 FR 17556, 17609-17613 (Apr. 6, 2006) for NHTSA discussion of "kinks" in the MYs 2008-2011 light truck CAFE final rule (there described as "edge effects"). A "kink," as used here, is a portion of the curve where a small change in footprint results in a disproportionally large change in stringency. 2-7 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting 2.3.2 MYs 2012-2016 Light Duty GHG/CAFE (constrained/piecewise linear) For the MYs 2012-2016 rules, NHTSA and EPA re-evaluated potential methods for specifying mathematical functions to define fuel economy and GHG standards. The agencies concluded that the constrained logistic form, if applied to post-MY 2011 standards, would likely contain a steep mid-section that would provide undue incentive to increase the footprint of midsize passenger cars.5 The agencies judged that a range of methods to fit the curves would be reasonable, and used a minimum absolute deviation (MAD) regression without sales weighting on a technology-adjusted car and light truck fleet to fit a linear equation. This equation was used as a starting point to develop mathematical functions defining the standards as discussed above. The agencies then identified footprints at which to apply minimum and maximum values (rather than letting the standards extend without limit) and transposed these constrained/piecewise linear functions vertically (i.e., on a gpm or CC>2 basis, uniformly downward) to produce the fleetwide fuel economy and CO2 emission levels for cars and light trucks described in the final rule.6 2.3.3 How have the agencies defined the mathematical functions for the MYs 2017- 2025 standards, and why? By requiring NHTSA to set CAFE standards that are attribute-based and defined by a mathematical function, NHTSA interprets Congress as intending that the post-EISA standards to be data-driven - a mathematical function defining the standards, in order to be "attribute- based," should reflect the observed relationship in the data between the attribute chosen and fuel economy.2 EPA is also setting attribute-based CO2 standards defined by similar mathematical functions, for the reasonable technical and policy grounds discussed below and in section II of the preamble to the rule, and to harmonize with the CAFE standards. The relationship between fuel economy (and GHG emissions) and footprint, though directionally clear (i.e., fuel economy tends to decrease and CO2 emissions tend to increase with increasing footprint), is theoretically vague and quantitatively uncertain; in other words, not so precise as to a priori yield only a single possible curve.aa There is thus a range of legitimate options open to the agencies in developing curve shapes. The agencies may of course consider statutory objectives in choosing among the many reasonable alternatives since the statutes do not dictate a particular mathematical function for curve shape. For example, curve shapes that might have some theoretical basis could lead to perverse outcomes contrary z A mathematical function can be defined, of course, that has nothing to do with the relationship between fuel economy and the chosen attribute - the most basic example is an industry-wide standard defined as the mathematical function average required fuel economy =X, where X is the single mpg level set by the agency. Yet a standard that is simply defined as a mathematical function that is not tied to the attribute(s) would not meet the requirement of EISA. aa In fact, numerous manufacturers have confidentially shared with the agencies what they describe as "physics based" curves, with each OEM showing significantly different shapes, and footprint relationships. The sheer variety of curves shown to the agencies further confirm the lack of an underlying principle of "fundamental physics" driving the relationship between CO2 emission or fuel consumption and footprint, and the lack of an underlying principle to dictate any outcome of the agencies' establishment of footprint-based standards. 2-8 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting to the intent of the statutes to conserve energy and reduce GHG emissions. .bb Thus, the decision of how to set the target curves cannot always be just about most "clearly" using a mathematical function to define the relationship between fuel economy and the attribute; it often has to have reflect legitimate policy judgments, where the agencies adjust the function that would define the relationship in order to achieve environmental goals, reduce petroleum consumption, encourage application of fuel-saving technologies, not adversely affect highway safety, reduce disparities of manufacturers' compliance burdens (thereby increasing the likelihood of improved fuel economy and reduced GHG emissions across the entire spectrum of footprint targets), preserve consumer choice, etc. This is true both for the decisions that guide the mathematical function defining the sloped portion of the target curves, and for the separate decisions that guide the agencies' choice of "cutpoints" (if any) that define the fuel economy/CO2 levels and footprints at each end of the curves where the curves become flat. Data informs these decisions, but how the agencies define and interpret the relevant data, and then the choice of methodology for fitting a curve to the data, must include a consideration of both technical data and policy goals. Supporting the consideration and selection of mathematical functions upon which to base new CAFE and GHG standards, the agencies conducted a broad-ranging analysis spanning different techniques for adjusting data and fitting linear functions. The next sections examine the policy concerns that the agencies considered in developing the target curves that define the MYs 2017-2025 CAFE and CO2 standards, technical work (expanding on similar analyses performed by NHTSA when the agency proposed MY 2011-2015 standards, and by both agencies during consideration of options for MY 2012-2016 CAFE and GHG standards) that was completed in the process of reexamining potential mathematical functions for this rulemaking, how the agencies have defined the data, and how the agencies explored statistical curve-fitting methodologies in order to arrive at proposed and final curves. Because the agencies are finalizing the target curves for MYs 2017-2025 as proposed, the following discussion largely mirrors the discussion in the version of the TSD that accompanied the proposal; it is repeated here for the reader's convenience. 2.4 What did the agencies propose for the MYs 2017-2025 curves? The mathematical functions for the proposed MYs 2017-2025 standards were somewhat changed from the functions for the MYs 2012-2016 standards, in response to comments received from stakeholders both pre-proposal and during the public comment period and in order to address technical concerns and policy goals that the agencies judged more significant in this nine-model year rulemaking than in the prior one, which only included five model years.cc This section (2.4) discusses the methodology the agencies bb For example, if the agencies set weight-based standards defined by a steep function, the standards might encourage manufacturers to keep adding weight to their vehicles to obtain less stringent targets. cc We note that although, due to statutory constraints, NHTSA is finalizing standards for only MYs 2017-2021 and presenting augural standards for MYs 2022-2025, the joint analysis was conducted by NHTSA and EPA with respect to shapes of target curves for all nine model years - both because EPA is indeed finalizing all nine years of standard curves, and because NHTSA's augural standards for MYs 2022-2025 represent the agency's 2-9 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting selected as best addressing those technical concerns and policy goals for this rulemaking, given the various technical inputs to the agencies' current analyses. Section 2.5 discusses how the agencies determined the cutpoints and the flat portions of the MYs 2017-2025 target curves. We note that both of these sections address only how the target curves were fit to fuel consumption and CC>2 emission values determined using the city and highway test procedures, and that in determining respective regulatory alternatives, the agencies made further adjustments to the resultant curves in order to account for adjustments for improvements to mobile air conditioners. Thus, recognizing that there are many reasonable statistical methods for fitting curves to data points that define vehicles in terms of footprint and fuel economy, the agencies chose for the proposed rule to fit curves using an ordinary least-squares formulation, on sales- weighted data, using a fleet that has had technology applied, and after adjusting the data for the effects of weight-to-footprint, as described below. This represents a departure from the statistical approach for fitting the curves in the MYs 2012-2016 rules, as explained in the next section (2.4.1). The agencies considered a wide variety of reasonable statistical methods in order to better understand the range of uncertainty regarding the relationship between fuel consumption (the inverse of fuel economy), CC>2 emission rates, and footprint, thereby providing a range within which decisions about standards would be potentially supportable. 2.4.1 What concerns were the agencies looking to address that led them to change from the approach used for the MYs 2012-2016 curves? Before the MY 2017 and later proposal was issued, NHTSA and EPA received a number of comments from stakeholders on how curves should be fitted to the passenger car and light truck fleets.dd Some limited-line manufacturers argued that curves should generally be flatter in order to avoid discouraging production of small vehicles, because steeper curves tend to result in more stringent targets for smaller vehicles. Most full-line manufacturers argued that a passenger car curve similar in slope to the MY 2016 passenger car curve would be appropriate for future model years, but that the light truck curve should be revised to be less stringent for manufacturers selling the largest full-size pickup trucks. These manufacturers argued that the MY 2016 light truck curve was not "physics-based," and that in order for future tightening of standards to be feasible for full-line manufacturers, the truck curve for later model years should be steeper and extended further (i.e., made less stringent) into the larger footprints. As stated in the TSD accompanying the proposal, the agencies do not agree that the MY 2016 light truck curve was somehow deficient in lacking a "physics basis," or that it was somehow overly stringent for manufacturers selling large pickups— manufacturers making these arguments presented no "physics-based" model to explain how fuel economy should depend on footprint.66 The same manufacturers indicated that they best estimate, based on the information currently before it, of the standards that the agency would finalize had it the authority to do so. NHTSA will fully revisit all aspects of the MYs 2022-2025 standards as part of the later rulemaking concurrent with the mid-term evaluation. dd See 75 FR at 76341 for a general summary. ee See footnote aa. 2-10 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting believed that the light truck standard should be somewhat steeper after MY 2016, primarily because, after more than ten years of progressive increases in the stringency of applicable CAFE standards, large pickups would be less capable of achieving further improvements without compromising load carrying and towing capacity. In developing the curve shapes for the proposed rule, the agencies were aware of the current and prior technical concerns raised by OEMs concerning the effects of the stringency on individual manufacturers and their ability to meet the standards with available technologies, while producing vehicles at a cost that allowed them to recover the additional costs of the technologies being applied. Although we continue to believe that the methodology for fitting curves for the MYs 2012-2016 standards was technically sound, we recognize manufacturers' technical concerns regarding their abilities to comply with a similarly shallow curve after MY 2016 given the anticipated mix of light trucks in MYs 2017- 2025. As in the MYs 2012-2016 rules, the agencies considered these concerns in the analysis of potential curve shapes for the MYs 2017-2025 proposal. The agencies also considered safety concerns which could be raised by curve shapes creating an incentive for vehicle downsizing, as well as the potential loss to consumer welfare should vehicle upsizing be unduly disincentivized. In addition, the agencies sought to improve the balance of compliance burdens among manufacturers, and thereby increase the likelihood of improved fuel economy and reduced GHG emissions across the entire spectrum of footprint targets. Among the technical concerns and resultant policy trade-offs the agencies considered were the following: Flatter standards (i.e., curves) increase the risk that both the weight and size of vehicles will be reduced, potentially compromising highway safety. Flatter standards potentially impact the utility of vehicles by providing an incentive for vehicle downsizing. Steeper footprint-based standards may create incentives to upsize vehicles, thus increasing the possibility that fuel economy and greenhouse gas reduction benefits will be less than expected. Given the same industry-wide average required fuel economy or CC>2 standard, flatter standards tend to place greater compliance burdens on full-line manufacturers Given the same industry-wide average required fuel economy or CC>2 standard, steeper standards tend to place greater compliance burdens on limited-line manufacturers (depending of course, on which vehicles are being produced). If cutpoints are adopted, given the same industry-wide average required fuel economy, moving small-vehicle cutpoints to the left (i.e., up in terms of fuel economy, down in terms of CC>2 emissions) discourages the introduction of small vehicles, and reduces the incentive to downsize small vehicles in ways that could compromise overall highway safety. If cutpoints are adopted, given the same industry-wide average required fuel economy, moving large-vehicle cutpoints to the right (i.e., down in terms of fuel economy, up in terms of CC>2 emissions) better accommodates the design requirements of larger vehicles—especially large pickups—and extends the size range over which downsizing is discouraged. 2-11 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting All of these were policy goals that required weighing and consideration. Ultimately, the agencies rejected the argument that the MY 2017 target curves for the proposal, on a relative basis, should be made significantly flatter than the MY 2016 curve,ff as we believed that this would undo some of the safety-related incentives and balancing of compliance burdens among manufacturers—effects that attribute-based standards are intended to provide. Nonetheless, the agencies recognized full-line OEM concerns and tentatively concluded that further increases in the stringency of the light truck standards would be more feasible if the light truck curve is made steeper than the MY 2016 truck curve and the right (large footprint) cut-point is extended over time to larger footprints. This conclusion was supported by the agencies' technical analyses of regulatory alternatives defined using the curves developed in the manner described below. 2.4.2 What methodologies and data did the agencies consider in developing the 2017-2025 curves presented in the proposal? In considering how to address the various policy concerns discussed in the previous sections, the agencies revisited the data and performed a number of analyses using different combinations of the various statistical methods, weighting schemes, adjustments to the data and the addition of technologies to make the fleets less technologically heterogeneous. As discussed in 2.3.3, in the agencies' judgment, there is no single "correct" way to estimate the relationship between CO2 or fuel consumption and footprint - rather, each statistical result is based on the underlying assumptions about the particular functional form, weightings and error structures embodied in the representational approach. These assumptions are the subject of the following discussion. This process of performing many analyses using combinations of statistical methods generated many possible outcomes, each embodying different potentially reasonable combinations of assumptions and each thus reflective of the data as viewed through a particular lens. The choice of a standard developed by a given combination of these statistical methods was consequently a decision based upon the agencies' determination of how, given the policy objectives for this rulemaking and the agencies' MY 2008-based forecast of the market through MY 2025, to appropriately reflect the current understanding of the evolution of automotive technology and costs, the future prospects for the vehicle market, and thereby establish curves (i.e., standards) for cars and light trucks. 2.4.2.1 For the MYs 2017-2025 standards, what information did the agencies use to estimate a relationship between fuel economy, CO2 and footprint? For each fleet, the agencies began with the MY 2008-based market forecast developed to support the proposal (i.e., the baseline fleet), with vehicles' fuel economy levels and technological characteristics at MY 2008 levels.gg The development, scope, and content of ff While "significantly" flatter is subjective qualitative description, the year over year change in curve shapes is discussed in greater detail in Section 2.5.3.1. 88 While the agencies jointly conducted this analysis, the coefficients ultimately used in the slope setting analysis are from the CAFE model. 2-12 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting this market forecast is discussed in detail in Chapter 1 of the joint Technical Support Document supporting the proposed rulemaking. Figure 2-1 shows the MY 2008 CC^by car and truck class as it existed in the EPA OMEGA and NHTSA CAFE NPRM model data files (for a gasoline-only fleet, fuel consumption—the inverse of fuel economy—is directly proportional to CC^). This fleet was the starting point for all analysis in the proposal. 700 f 2008 CO2 v. Footprint 500 - O U 200 - 3>\$\$'.:•• -ff :*.. - • «| |U'" : • &-k- 0 50000 O 100000 O 150000 O 200000 O 250000 40 50 BO 70 40 50 60 70 Footprint Figure 2-1 2008 CO2 vs. Footprint by Car and Truck Although the agencies are finalizing the target curves as proposed, the agencies have also revisited and updated their analyses for this final rule, and found that the proposed curves are well within the ranges spanned by the final rule analyses. See section 2.6 below. As discussed in Chapter 1 of this TSD, the agencies have used two different market forecasts to conduct additional analyses supporting this final rule. The first, referred to here as the "MY 2008-Based Fleet Projection," is largely identical to that used for analysis supporting the NPRM, but includes some corrections to the footprint of some vehicle models discussed in Chapter 1, as well as other minor changes. The second, referred to here as the "MY 2010- Based Fleet Projection," is a post-proposal market forecast based on the MY 2010 fleet of vehicles. Using both of these projected fleets, the agencies repeated the analyses described below, and obtained broadly similar results, details of which are presented in a memorandum 2-13 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting available in NHTSA's docket.1* Because the agencies are promulgating target curve standards identical to those proposed in the NPRM, the remainder of this chapter reviews results supporting the development of those proposed standards. This chapter concludes with a summary of results of the agencies' updated analysis, and discussion of the consideration that analysis was given in selecting mathematical functions upon which to base the standards in the final rules. 2.4.2.2 What adjustments did the agencies evaluate? As indicated in the TSD supporting the NPRM, one possible approach is to fit curves to the minimally adjusted data shown above (the approach still includes sales mix adjustments, which influence results of sales-weighted regressions), much as DOT did when it first began evaluating potential attribute-based standards in 2003.7 However, the agencies found, as in prior rulemakings, that the data are so widely spread (i.e., when graphed, they fall in a loose "cloud" rather than tightly around an obvious line) that they indicate a relationship between footprint and CC>2 and fuel consumption that is real but not particularly strong (Figure 2-1). Therefore, as discussed below, the agencies also explored possible adjustments that could help to explain and/or reduce the ambiguity of this relationship, or could help to produce policy outcomes the agencies judged to be more desirable. 2.4.2.3 Adjustment to reflect differences in technology As in prior rulemakings, the agencies considered technology differences between vehicle models to be a significant factor producing uncertainty regarding the relationship between CC>2/fuel consumption and footprint. Noting that attribute-based standards are intended to encourage the application of additional technology to improve fuel efficiency and reduce CC>2 emissions, the agencies, in addition to considering approaches based on the unadjusted engineering characteristics of MY 2008 vehicle models, therefore also considered approaches in which, as for previous rulemakings, technology is added to vehicles for purposes of the curve fitting analysis in order to produce fleets that are less varied in technology content. This approach helps to reduce "noise" (i.e.., dispersion) in the plot of vehicle footprints and fuel consumption levels and to identify a more technology-neutral relationship between footprint and fuel economy / CC>2 emissions. For the analysis supporting the NPRM, the agencies adjusted the NPRM baseline fleet for technology by adding all technologies considered, except for, diesel engines, integrated starter generators, strong FtEVs, PFtEVs, EVs, FCVs, and the most advanced high-BMEP (brake 14 Docket No. NHTSA-2010-0131. As with the NPRM analysis, EPA and NHTSA jointly analyzed the fleet projections used in this final rulemaking. While the proposal and final rulemaking analyses shown in this chapter are from the NHTSA CAFE model, the EPA OMEGA results are generally similar, and support the same conclusions. A memo containing the OMEGA results for the FRM can be found in EPA docket EPA-HQ-OAR- 2010-0799. 2-14 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting mean effective pressure) gasoline engines." The agencies included 15 percent mass reduction on all vehicles. Figure 2-2 shows the same fleet, with technology adjustment and 2021 sales applied, and the baseline diesel fueled vehicles, HEV and EVs removed from the fleet. Of note, the fleet is now more closely clustered" (and lower in emissions), but the same basic pattern emerges; in both figures, the CO2 emission rate (which, as mentioned above, is directly proportional to fuel consumption for a gasoline-only fleet) increases with increasing footprint, although the relationship is less pronounced for larger light trucks. Max ICE Tech - CO2 v. Footprint 8 o |300- (0 E 0 >* °2QQ - 0 1 0 - 1 •1 * i J Vj • a° 0 "* °£ •"t^J1 :S 1 C * a a ?•• F 0* . • • • • •? •1 ^* W T . iT *° f' a 1%* *& | .i- 2021 Projected Sales n o [• I 50000 [•^ 100000 • 150000 • 200000 • 250000 0 300000 40 50 60 70 80 40 50 60 70 80 Footprint Figure 2-2 2008 CO2 vs. Footprint by Car and Truck, after Adjustment Reflecting Technology Differences, and removing diesel fueled vehicles, HEVs and EVs Updating this analysis using the current MY2008- and MY2010-based fleet projection yielded results generally similar to those shown above. Detailed results of the analyses with 11 As described in the preceding paragraph, applying technology in this manner serves to reduce the effect of technology differences across the vehicle fleet. The particular technologies used for the normalization were chosen as a reasonable selection of technologies which could potentially be used by manufacturer over this time period. JJ For cars, the standard deviation of the CO2 data is reduced from 81 to 54 through the technology normalization. For trucks, the standard deviation is reduced from 62 to 36. 2-15 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting the final rulemaking fleet projections are presented in a memorandum available in NHTSA's docket.1* 2.4.2.4 Adjustments reflecting differences in performance and "density" As discussed in Section 2.4.1, during stakeholder meetings the agencies held while developing the NPRM, some manufacturers indicated that they believed that the light truck standard should be somewhat steeper after MY 2016. As a means to produce a steeper light truck curve, the agencies considered adjustments for other differences between vehicle models (i.e., inflating or deflating the fuel economy of each vehicle model based on the extent to which one of the vehicle's attributes, such as power, is higher or lower than average). Previously, NHTSA had rejected such adjustments because they imply that a multi-attribute standard may be necessary, and as explained above, the agencies judged most multi-attribute standards to be more subject to gaming than a footprint-only standard.mm'8 Having considered this issue again for purposes of this rulemaking, NHTSA and EPA concluded the need to accommodate in the target curves the challenges faced by manufacturers of large pickups currently outweighs these prior concerns (comments on this topic are discussed in Section 0 and 2.4.2.11 and in Section II. C of the preamble). Therefore, the agencies also evaluated curve fitting approaches through which fuel consumption and CC>2 levels were adjusted with respect to weight-to-footprint alone, and in combination with power-to-weight. While the agencies examined these adjustments for purposes of fitting curves, the agencies did not propose a multi-attribute standard; the proposed fuel economy and CC>2 targets for each vehicle were still functions of footprint alone. The agencies are not promulgating a multi- attribute standard, and no adjustment will be used in the compliance process. The agencies also examined some differences between the technology-adjusted car and truck fleets in order to better understand the relationship between footprint and CC>2/fuel consumption in the agencies' MY 2008 based forecast. More direct measures (such as coefficients of drag and rolling resistance), while useful for vehicle simulation, were not practical or readily available at the fleet level. Given this issue, and based on analysis published in the MYs 2012-2016 rule,9 the agencies investigated a sales-weighted (i.e., treating every vehicle unit sold as a separate observation) regression equation involving power to weight ratio and vehicle weight (Equation 2-1).nn This equation provides for a ^ Docket No. NHTSA-2010-0131. 11 See Preamble I.A.2 for a discussion of the stakeholder meetings before the NPRM. mm For example, in comments on NHTSA's 2008 NPRM regarding MY 2011-2015 CAFE standards, Porsche recommended that standards be defined in terms of a "Summed Weighted Attribute", wherein the fuel economy target would calculated as follows: target =f(SWA), where target is the fuel economy target applicable to a given vehicle model and SWA =footprint + torque111 s + weight112 \ (NHTSA-2008-0089-0174). While the standards the agencies proposed for MY 2017-2025 are not multi-attribute standards, that is the target is only a function of footprint, we proposed curve shapes that were developed considering more than one attribute. 1111 These parameters directly relate to the amount of energy required to move the vehicle. As compared to a lighter vehicle, more energy is required to move a heavier vehicle the same distance. Similarly, a more powerful engine, when technology adjusted, is less efficient than a less powerful engine. 2-16 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting strong correlation between HP/WT, weight and CC>2 emissions (R2=0.78, Table 2-1) after accounting for technology adjustments.00 Equation 2-1 - Relationship between vehicle attributes and emissions or fuel consumption Where: HP/Weight= the rated horsepower of the vehicle divided by the curb weight Weight = the curb weight of the vehicle in pounds C = a constant. Table 2-1 - Physical Regression Coefficients against Technology Adjusted CO2 R2 F-test p C Cars 0.78 <0.01 1.09*10' 3.29*10'2 -3.29 Light Trucks 0.78 <0.01 1.13*10' 3.45*10'2 2.73 *In this gasoline only fleet, these coefficients can be divided by 8887 (the amount of CO2 produced by the combustion of a gallon of the fuel used to certify the fuel economy and emissions of gasoline vehicles) to yield the corresponding fuel consumption coefficients. Updating this analysis using the MY 2008- and MY 2010-based fleet projections yielded results generally similar to those shown above. Detailed results of the analyses with the final rulemaking fleet projections are presented in a memorandum available in NHTSA's docket.pp The coefficients above show, for the agencies' MY 2008-based market forecast as developed for the NPRM, strong correlation between these vehicle attributes and the fuel consumption and emissions of the vehicle, as well as strong similarity between car and truck coefficients. (As explained in section 2.6 below, our analysis using the corrected version of the MY 2008 based market forecast used for the final rule, as well as the alternative 2010 based market forecast, is consistent with these results.) Given these very similar parameters, 00 As R does not equal 1, there are remaining unaccounted for differences beyond technology, power and weight. These may include gear ratios, axle ratios, aerodynamics, and other vehicle features not captured in this equation. pp Docket No. NHTSA-2010-0131. 2-17 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting similar distributions of power and weight would be expected to produce similarly arrayed plots of CC>2(or equivalently, fuel consumption) by footprint, regardless of car or truck class. Based on the differences seen in the technology-adjusted plot (Figure 2-2), the agencies further investigated these particular attributes and their relationship to footprint in the agencies' MY 2008-based market forecast developed for the NPRM, to examine the differences across the footprint distribution. Figure 2-3 shows vehicle curb weight charted against footprint, with sales weighted ordinary least squares sales fit (blue) and sales-weighted LOESS fit (red) imposed. For cars, the LOESS fit, which weights nearby points more heavily, qq is nearly identical to the linear fit in the data filled region between about 40 and 56 sq ft (with the gray bar showing standard error on the Loess fit). For this market forecast, average car curb weight is linearly proportional to car footprint between 40 and 56 sq ft, or in other words, cars progress in weight in a regular fashion as they get larger (Figure 2-3). Figure 2-3 By contrast, a linear fit does not overlap with the LOESS fit on the truck side, which indicates that for this market forecast, truck curb weight does not linearly increase with footprint, at least not across the entire truck fleet. The LOESS fit shows that larger trucks (those on the right side of the data bend in Figure 2-2) have a different trend than smaller trucks, and after about 55 sq ft, no longer proportionally increases in weight. The same pattern is seen in Figure 2-1 and Figure 2-2 above. qq: In a LOESS regression, "fitting is done locally. That is, for the fit at point x, the fit is made using points in a neighborhood of x, weighted by their distance from x (with differences in 'parametric' variables being ignored when computing the distance). The size of the neighborhood is controlled by a For a < 1, the neighborhood includes proportion a of the points, and these have tricubic weighting (proportional to (1 - (dist/maxdist)^3)^3. For a > 1, all points are used, with the 'maximum distance' assumed to be a^l/p times the actual maximum distance forp explanatory variables." A span of 1 was used in these images, http://cran.r-project.org/doc/manuals/fullrefman.pdf. p. 1406. 2-18 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting WT v. FP - Weighted OLS and Loess Fit 5000 - 3000 - • . I 40 50 60 70 40 50 60 70 Footprint Figure 2-3 Relationship between Weight and Footprint in Agencies' MY2008-Based Market Forecast Updating this analysis using the revised MY 2008- and the MY 2010-based fleet projection yielded results generally similar to those shown above. Detailed results of the analyses with the final rulemaking fleet projections are presented in a memorandum available in NHTSA's docket." To further pursue this topic, weight divided by footprint (WT/FP) can be thought of as a "density" of a vehicle (although dimensionally it has units of pressure). As seen in Figure 2-4, the trend in WT/FP in the agencies' MY2008-based market forecast is different in trucks than in cars. The linear trend on cars is an increase in WT/FP as footprint increases (Figure 2-4). In contrast, light trucks do not consistently increase in WT/FP ratio as the vehicles grow larger, but WT/FP actually decreases (Figure 2-4). Docket No. NHTSA-2010-0131. 2-19 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting WT/FP v, FP - Weighted OLS 70 40 Footprint Figure 2-4 Relationship between Weight/FP and Footprint in Agencies' MY2008-Based Market Forecast Updating this analysis using the current MY 2008- and MY 2010-based fleet projection yielded results generally similar to those shown above. Detailed results of the analyses with the final rulemaking fleet projections are presented in a memorandum available inNHTSA'sdocket.88 The heterogeneity of the truck fleet explains part of the WT/FP trend, where the pickup truck fleet is largest in footprint, but is also relatively light for its size due to the flat bed (Figure 2-5). Note that the two light truck classes with the smallest WT/FP ratios are small and large pickups. Further, as the only vehicle class with a sales-weighted average footprint above 60 square feet, the large pickup trucks have a strong influence on the slope of the truck curve. As the correlation between weight and CC>2 is strong (Table 2-1), having proportionally lighter vehicles at one extreme of the footprint distribution can bias a curve fit to these vehicles. If no adjustment is made to the curve fitted to the truck fleet, and no other compensating flexibilities or adjustments are made available, manufacturers selling significant numbers of vehicles at the large end of the truck distribution will face compliance burdens that are comparatively more challenging that those faced by manufacturers not serving this part of the light truck market. As noted further below, this consideration 1 Docket No. NHTSA-2010-0131. 2-20 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting provided the basis for the agencies' proposal to change the cutpoint for larger light trucks from 66 feet to 74 feet, and to steepen the slope of the light truck curve for larger light trucks. WT/FP by Vehicle Class Truckfleetavg.: 84.3 Carfleetavg.; 72.7 Footprint Figure 2-5 Class and the WT/FP distribution Updating this analysis using the revised MY 2008- and the MY 2010-based market forecasts yielded results generally similar to those shown above. Detailed results of the analyses with the final rulemaking fleet projections are presented in a memorandum available inNHTSA'sdocket* The agencies also investigated the relationship between HP/WT and footprint in the agencies' MY 2008-based market forecast developed for the NPRM (Figure 2-6). On a sales weighted basis, cars tend to become proportionally more powerful as they get larger. In contrast, there is a minimally positive relationship between HP/WT and footprint for light trucks, indicating that light trucks become only slightly more powerful as they get larger, but that the trend is not especially pronounced. 'DocketNo. NHTSA-2010-0131. 2-21 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting HP/WT v. FP - Weighted OLS 020 • 015 - 010 - 005 V • Tt * 40 50 60 70 40 SO 60 70 Footprint 2021 Salts « 0 ^ 50000 • 100000 I* 150000 • 200000 • 250000 • 300000 Figure 2-6 HP/WT v. FP Updating this analysis using the revised MY 2008- and the MY 2010-based fleet projection yielded results generally similar to those shown above. Detailed results of the analyses with the final rulemaking fleet projections are presented in a memorandum available inNHTSA'sdocket.1111 One factor influencing results of this analysis is the non-homogenous nature of the truck fleet; some vehicles at the smaller end of the footprint curve are different in design and utility from others at the larger end (leading to the observed bend in the LOESS fit, Figure 2-6). There are many high volume four-wheel drive vehicles with smaller footprint in the truck fleet (such as the Chevrolet Equinox, Dodge Nitro, Ford Escape, Honda CR-V, Hyundai Santa Fe, Jeep Liberty, Nissan Rogue, Toyota RAV4, and others) exhibit only select truck characteristics.™ By contrast, the largest pickup trucks in the light truck fleet have unique aerodynamic and power characteristics that tend to increase CO2 emissions and fuel consumption. These disparities contribute to the slopes of lines fitted to the light truck fleet. uu Docket No. NHTSA-2010-0131. vv In most cases, these vehicles have four-wheel drive, but no significant towing capability, and no open-bed. Many of these vehicles are also offered without four-wheel drive, and these two-wheel drive versions are classified as passenger cars, not light trucks. 2-22 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Several comments, such as those by CBD and ACEEE, were submitted with regard to the non-homogenous nature of the truck fleet, and the "unique" attributes of pickup trucks. Ford Motor Company described the attributes of these vehicles, noting that "towing capability generally requires increased aerodynamic drag caused by a modified frontal area, increased rolling resistance, and a heavier frame and suspension to support this additional capability."^ Ford further noted that these vehicles further require auxiliary transmission oil coolers, upgraded radiators, trailer hitch connectors and wiring harness equipment, different steering ratios, upgraded rear bumpers and different springs for heavier tongue load (for upgraded towing packages), body-on-frame (vs. unibody) construction (also known as ladder frame construction) to support this capability and an aggressive duty cycle, and lower axle ratios for better pulling power/capability. In the agencies' judgment, the curves and cutpoints defining the light truck standards appropriately account for engineering differences between different types of vehicles. For example, the agencies' estimates of the applicability, cost, and efficacy of different fuel-saving technologies differentiate between small, medium, and large light trucks. Further discussion on this topic is contained in Section II.C. The agencies' technical analyses of regulatory alternatives developed using curves fitted as described below supported OEM comments that there would be significant compliance challenges for the manufacturers of large pickup trucks, and led toward the agencies' policy goal of a steeper slope for the light truck curve relative to MY 2016. Three primary drivers were as follows: (a) the largest trucks have unique equipment and design, as described in the Ford comment referenced above; (b) the agencies agree with those large truck manufacturers who indicated in discussions prior to the proposal that they believed that the light truck standard should be somewhat steeper after MY 2016, primarily because, after more than ten recent years of progressive increases in the stringency of applicable CAFE standards (after nearly ten years during which Congress did not allow NHTSA to increase light truck CAFE standards), manufacturers of large pickups would have limited options to comply with more stringent standards without resorting to compromising large truck load carrying and towing capacity; and (c) given the relatively few platforms which comprise the majority of the sales at the largest truck footprints, the agencies were concerned about requiring levels of average light truck performance that might lead to overly aggressive advanced technology penetration rates in this important segment of the work fleet. Specifically, the agencies were concerned at proposal, and remain concerned, about issues of lead time and cost with regard to manufacturers of these work vehicles. As noted later in this chapter, while the largest trucks are a small segment of the overall truck fleet, and an even smaller segment of the overall fleet, ** these changes to the truck slope have been made in order to provide a clearer path toward compliance for manufacturers of these vehicles, and reduce the potential that new ww Ford comments, Docket No. [fill in], at [page number]. xx The agencies' market forecast used at proposal includes about 24 vehicle configurations above 74 square feet with a total volume of about 50,000 vehicles or less during any MY in the 2017-2025 time frame, In the MY2010 based market forecast, there are 14 vehicle configurations with a total volume of 130,000 vehicles or less during any MY in the 2017-2025 time frame. This is a similarly small portion of the overall number of vehicle models or vehicle sales. 2-23 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting standards would lead these manufacturers to choose to downpower, modify the structure, or otherwise reduce the utility of these work vehicles. Some commenters disagreed with these policy goals concerning the largest light trucks and argued that higher fuel economy for the largest light trucks is fully compatible with maintaining towing and hauling capacity. These comments, which largely deal with stringency, are addressed in each agency's respective preamble section (HID and IV.F), as well as in Section II. C, which addresses the shapes of the target curves. Consequently, the agencies considered options including fitting curves developed using results of the analysis described above. Specifically, the agencies note that the WT/FP ratio of the light duty fleet potentially has a large impact on a sales-weighted regression.37 The increasing trend in WT/FP versus footprint for cars in the 2008 MY baseline would steepen the slope of the car curve, while the decreasing trend in WT/FP would flatten the truck slope, as compared to a WT/FP adjusted fleet. This result was reflected in the MYs 2012-2016 final rulemaking,10 where the agencies noted the steep car curves resulting from a weighted least-squares analysis. Based on the above analysis, the agencies also considered adjustments for other differences between vehicle models. Therefore, utilizing the coefficients derived in Equation 2-1, the agencies also evaluated curve fitting approaches through which fuel consumption and CC>2 levels were adjusted with respect to weight-to-footprint alone, and in combination with power-to-weight. This adjustment procedure inflates or deflates the fuel economy or CO2 emissions of each vehicle model based on the extent to which one of the vehicle's attributes, such as power, is higher or lower than average. As mentioned above, while the agencies considered this technique for purposes of fitting curves, the agencies did not propose a multi- attribute standard, as the proposed fuel economy and CO2 targets for each vehicle were still functions of footprint alone. The agencies are not promulgating a multi-attribute standard, and no adjustment would be used in the compliance process. The basis for the gallon-per-mile (GPM) adjustments is the sales-weighted linear regression discussed in 2.4 (Equation 2-1, Table 2-1). The coefficients to this equation give the impact of the various car attributes on CC>2 emissions and fuel consumption in the agencies' MY 2008-based market forecast used in the NPRM. For example, gives the impact of weight while holding the ratio horsepower to weight constant. Importantly, this means that as weight changes, horsepower must change as well to keep the power/weight ratio constant. Similarly, gives the CC>2 impact of changing the performance of the vehicle while keeping the weight constant. These coefficients were used to perform an adjustment of the gallons per mile measure for each vehicle to the respective car or truck—i.e., in the case of a HP/WT adjustment, to deflate or inflate the fuel consumption of each vehicle model based on the extent to which the vehicle's power-to-weight ratio is above or below the regression-based value at that footprint. yy As mentioned above, the agencies also performed the same analysis without sales-weighting, and found that the WT/FP ratio also had a directionallv similar effect on the fitted car and truck curves. 2-24 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting The agencies performed this normalization to adjust for differences in vehicle weight per square foot observations in the data discussed in Section 2.4. This adjustment process requires two pieces of information: the weight coefficient from Equation 2-1 and the average weight per footprint (i.e., pounds per square foot) for that vehicle's group. Two groups, passenger cars and light trucks, were used. For each group, the average weight per footprint was calculated as a weighted average with the weight being the same as in the above regression (projected sales by vehicle in 2021). The equation below indicates how this adjustment was carried out. 2-25 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Equation 2-2 WT/FP adjustment The term in parentheses represents the vehicle's deviation from an "expected weight." That is, multiplying the average weight per footprint for a group of vehicles (cars or trucks) by a specific vehicle's footprint gives an estimate of the weight of that specific vehicle if its density were "average," based on the analyzed fleet. Put another way, this factor represents what the weight is "expected" to be, given the vehicle's footprint, and based on the analyzed fleet. This "expected weight" is then subtracted from the vehicle's actual weight. Vehicles that are heavier than their "expected weight" will receive a positive value (i.e., a deflated fuel economy value) here, while vehicles that are lighter than their "expected weight" will receive a negative number (i.e., an inflated fuel economy value). This deviation from "expected weight" is then converted to a gallon value by the regression coefficient. The units on this coefficient are gallons per mile per pound, as can be deduced from equation 1. This value is then subtracted from the vehicle's actual gallons per mile measure. Note that the adjusted truck data no longer exhibits the bend seen in Figure 2-1 and Figure 2-2. in o £ 8 CM O 40 50 60 70 80 40 50 60 70 80 Figure 2-7 WT/FP Adjusted Fuel Consumption vs. Footprint 2-26 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Updating this analysis using the revised MY 2008- and the MY 2010-based fleet projection yielded results generally similar to those shown above. Detailed results of the analyses with the final rulemaking fleet projections are presented in a memorandum available inNHTSA'sdocket.22 This adjustment serves to reduce the variation in gallons per mile measures caused by variation in weight in the agencies' MY 2008-based market forecast used in the NPRM. Importantly, this adjustment serves to reduce the fuel consumption (i.e., inflate fuel economy) for those vehicles which are heavier than their footprint would suggest while increasing the gallons per mile measure (i.e., deflating fuel economy) for those vehicles which are lighter. For trucks, a linear trend is more evident in the data cloud.aaa The following table shows the degree of adjustment for several vehicle models: zz Docket No. NHTSA-2010-0131. aaa Using EPA's dataset, R2 for the sales weighted ordinary least squared linear fit between footprint and CO2 improved from 0.38 (technology adjusted CO2) to 0.64 (technology and weight / footprint adjusted CO2) 2-27 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Table 2-2 - Sample Adjustments for Weight to Footprint, Cars Manufacturer HONDA TOYOTA FORD GENERAL MOTORS HONDA NISSAN GENERAL MOTORS FORD TOYOTA VOLKSWAGEN FORD HONDA HYUNDAI HONDA Model HONDA FIT TOYOTA COROLLA FORD FOCUS CHEVROLET MALIBU HONDA ACCORD INFINITIG37 CHEVROLET CORVETTE FORD MUSTANG TOYOTA CAMRY VOLKSWAGEN JETTA FORD FUSION HONDA ACCORD HYUNDAI SONATA HONDA CIVIC Name Plate FIT COROLLA FOCUS FWD MALIBU ACCORD 4DR SEDAN G37 COUPE CORVETTE MUSTANG CAMRY SOLARA CONVERTIBLE JETTA FUSION FWD ACCORD 2DR COUPE SONATA CIVIC Weight / Footprint 64.4 61.3 62.9 73.5 69.6 76.7 69.3 74.7 75.6 78.0 72.2 71.6 70.7 59.9 Footprint 39.5 42.5 41.7 46.9 46.6 47.6 46.3 46.7 46.9 42.4 46.1 46.6 46.0 43.2 GPM 0.01 0.01 0.02 0.02 0.02 0.02 0.02 0.03 0.02 0.02 0.02 0.02 0.02 0.02 MPG 69.40 69.94 61.94 53.70 57.57 47.83 40.84 31.32 50.87 46.77 59.96 56.92 61.72 64.25 Adjusted GPM 0.0157 0.0164 0.0177 0.0185 0.0179 0.0200 0.0251 0.0316 0.0191 0.0211 0.0168 0.0178 0.0166 0.0177 Adjusted MPG 63.73 60.80 56.34 54.08 55.73 50.08 39.83 31.67 52.27 47.47 59.61 56.26 60.34 56.38 GPM% Adjustment 8.9% 15.0% 9.9% -0.7% 3.3% -4.5% 2.5% -1.1% -2.7% -1.5% 0.6% 1.2% 2.3% 14.0% 2-28 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Table 2-3 - Sample Adjustments for Weight to Footprint, Trucks Manufacturer FORD GENERAL MOTORS FIAT HONDA TOYOTA FORD FIAT TOYOTA TATA GENERAL MOTORS GENERAL MOTORS GENERAL MOTORS TOYOTA Model FORD ESCAPE CHEVROLET CIS JEEP GRAND CHEROKEE HONDA PILOT TOYOTA HIGHLANDER FORD F150 DODGE RAM TUNDRA LAND ROVER RANGE ROVER SPORT CHEVROLET UPLANDER HUMMER H3 PONTIAC TORRENT TACOMA Name Plate ESCAPE FWD CIS SILVERADO 2WD 119WB GRAND CHEROKEE 4WD PILOT 4WD HIGHLANDER 4WD F150 FFV 4WD 145 WB RAM 1500 PICKUP 4WD 140 WB TOYOTA TUNDRA 4WD 145 WB RANGE ROVER SPORT UPLANDER FWD H34WD TORRENT FWD TOYOTA TACOMA 4WD Weight / Footprint 80.1 85.9 103.7 85.2 79.6 73.8 78.1 79.3 118.6 114.4 99.9 84.2 74.8 Footprint 65.2 55.9 47.1 51.3 49.0 67.4 66.3 68.7 47.5 49.2 50.7 48.2 53.4 GPM 0.02 0.03 0.02 0.02 0.02 0.03 0.03 0.03 0.03 0.02 0.03 0.02 0.02 MPG 51.00 39.76 41.45 40.95 45.90 32.70 33.75 32.07 33.17 45.46 36.71 46.64 43.01 Adjusted GPM 0.0181 0.0248 0.0222 0.0243 0.0227 0.0334 0.0316 0.0325 0.0239 0.0163 0.0242 0.0215 0.0252 Adjusted MPG 55.11 40.29 44.98 41.22 44.05 29.97 31.65 30.73 41.92 61.34 41.30 46.56 39.63 GPM% Adjustment -7.5% -1.3% -7.9% -0.6% 4.2% 9.1% 6.6% 4.3% -20.9% -25.9% -11.1% 0.2% 8.5% Updating this analysis using the revised MY 2008- and the MY 2010-based fleet projection yielded results generally similar to those shown above. Detailed results of the analyses with the final rulemaking fleet projections are presented in a memorandum available inNHTSA'sdocket'" bbb Based on Equation 2-1, the agencies also evaluated an adjustment of GPM and CO2 based on HP/WT. Equation 2-3 -Adjustment based on HP/WT bbb Docket No. NHTSA-2010-0131. 2-29 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Figure 2-8 shows the adjusted data and the estimated relationship between the adjusted GPM values and footprint. CO o CL O CM O 50 60 70 80 40 Footprint 50 60 70 80 Figure 2-8 HP/WT Adjusted Fuel Consumption v. Footprint Table 2-4 shows the degree of adjustment for several vehicle models. Those vehicles which have more power than average for their actual curb weight are adjusted downward (i.e., fuel economy ratings are inflated), while those that have less power than average are adjusted upward (i.e., fuel economy ratings are deflated). Table 2-4 - Sample Adjustments for Horsepower to Weight, Cars Manufacturer HONDA TOYOTA FORD GENERAL MOTORS Model HONDA FIT TOYOTA COROLLA FORD FOCUS CHEVROLET MALIBU Name Plate FIT COROLLA FOCUS FWD MALIBU Horsepower 109 126 140 169 Footprint 39.5 42.5 41.7 46.9 GPM 0.01 0.01 0.02 0.02 MPG 69.40 69.94 61.94 53.70 Adjusted GPM 0.0157 0.0164 0.0177 0.0185 Adjusted MPG 63.73 60.80 56.34 54.08 GPM% Adjustment 8.9% 15.0% 9.9% -0.7% 2-30 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting HONDA NISSAN GENERAL MOTORS FORD TOYOTA VOLKSWAGEN FORD HONDA HYUNDAI HONDA HONDA ACCORD INFINITIG37 CHEVROLET CORVETTE FORD MUSTANG TOYOTA CAMRY VOLKSWAGEN JETTA FORD FUSION HONDA ACCORD HYUNDAI SONATA HONDA CIVIC ACCORD 4DR SEDAN G37 COUPE CORVETTE MUSTANG CAMRYSOLARA CONVERTIBLE JETTA FUSION FWD ACCORD 2DR COUPE SONATA CIVIC 190 330 400 500 225 170 160 190 162 140 46.6 47.6 46.3 46.7 46.9 42.4 46.1 46.6 46.0 43.2 0.02 0.02 0.02 0.03 0.02 0.02 0.02 0.02 0.02 0.02 57.57 47.83 40.84 31.32 50.87 46.77 59.96 56.92 61.72 64.25 0.0179 0.0200 0.0251 0.0316 0.0191 0.0211 0.0168 0.0178 0.0166 0.0177 55.73 50.08 39.83 31.67 52.27 47.47 59.61 56.26 60.34 56.38 3.3% -4.5% 2.5% -1.1% -2.7% -1.5% 0.6% 1.2% 2.3% 14.0% Table 2-5 - Sample Adjustments for Horsepower to Weight, Trucks Manufacturer FORD GENERAL MOTORS FIAT HONDA TOYOTA FORD FIAT TOYOTA TATA GENERAL MOTORS GENERAL MOTORS GENERAL MOTORS TOYOTA Model FORD ESCAPE CHEVROLET CIS JEEP GRAND CHEROKEE HONDA PILOT TOYOTA HIGHLANDER FORD F150 DODGE RAM TUNDRA LAN DROVER RANGE ROVER SPORT CHEVROLET UPLANDER HUMMER H3 PONTIAC TORRENT TACOMA Name Plate ESCAPE FWD CIS SILVERADO 2WD 119WB GRAND CHEROKEE 4WD PILOT 4WD HIGHLANDER 4WD F150 FFV 4WD 145 WB RAM 1500 PICKUP 4WD 140 WB TOYOTA TUNDRA 4WD 145 WB RANGE ROVER SPORT UPLANDER FWD H34WD TORRENT FWD TOYOTA TACOMA 4WD Horsepower 153 195 210 244 270 300 345 381 300 240 242 185 236 Footprint 65.2 55.9 47.1 51.3 49.0 67.4 66.3 68.7 47.5 49.2 50.7 48.2 53.4 GPM 0.02 0.03 0.02 0.02 0.02 0.03 0.03 0.03 0.03 0.02 0.03 0.02 0.02 MPG 51.00 39.76 41.45 40.95 45.90 32.70 33.75 32.07 33.17 45.46 36.71 46.64 43.01 Adjusted GPM 0.0181 0.0248 0.0222 0.0243 0.0227 0.0334 0.0316 0.0325 0.0239 0.0163 0.0242 0.0215 0.0252 Adjusted MPG 55.11 40.29 44.98 41.22 44.05 29.97 31.65 30.73 41.92 61.34 41.30 46.56 39.63 GPM % Adjustment -7.5% -1.3% -7.9% -0.6% 4.2% 9.1% 6.6% 4.3% -20.9% -25.9% -11.1% 0.2% 8.5% Updating this analysis using the revised MY 2008- and the MY 2010-based fleet projection yielded results generally similar to those shown above. Detailed results of the 2-31 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting analyses are with the final rulemaking fleet projections presented in a memorandum available inNHTSA'sdocket.ccc The above approaches resulted in three data sets each for (a) vehicles without added technology and (b) vehicles with technology added to reduce technology differences, any of which may provide a reasonable basis for fitting mathematical functions upon which to base the slope of the standard curves: (1) vehicles without any further adjustments; (2) vehicles with adjustments reflecting differences in "density" (weight/footprint); and (3) vehicles with adjustments reflecting differences in "density," and adjustments reflecting differences in performance (power/weight). Further, these sets were developed for both the revised MY 2008-based fleet projection and the post-proposal MY 2010-based fleet projection. Detailed results using these market forecasts are presented in a memorandum available in NHTSA's docket.ddd 2.4.2.5 What statistical methods did the agencies evaluate? Using these data sets, the agencies tested a range of regression methodologies, each judged to be possibly reasonable for application to at least some of these data sets. 2.4.2.6 Regression Approach In the MYs 2012-2016 final rules, the agencies employed a robust regression approach (minimum absolute deviation, or MAD), rather than an ordinary least squares (OLS) regression.11 MAD is generally applied to mitigate the effect of outliers in a dataset, and thus was employed in that rulemaking as part of our interest in attempting to best represent the underlying technology. NHTSA had used OLS in early development of attribute-based CAFE standards, but NHTSA (and then NHTSA and EPA) subsequently chose MAD instead of OLS for both the MY 2011 and the MYs 2012-2016 rulemakings. These decisions on regression technique were made both because OLS gives additional emphasis to outliers12 and because the MAD approach helped achieve the agencies' policy goals with regard to curve slope in those rulemakings.13 In the interest of taking a fresh look at appropriate regression methodologies as promised in the 2012-2016 light duty rulemaking, in developing this proposal, the agencies gave full consideration to both OLS and MAD. The OLS representation, as described, uses squared errors, while MAD employs absolute errors and thus weights outliers less. As noted, one of the reasons stated for choosing MAD over least square regression in the MYs 2012-2016 rulemaking was that MAD reduced the weight placed on outliers in the data. As seen in Figure 2-1, there clearly are some outliers in the data, mostly to the high CO2 and fuel consumption side. However, the agencies have further considered whether it is appropriate to classify these vehicles as outliers. Unlike in traditional datasets, these vehicles' performance is not mischaracterized due to errors in their measurement, a common reason for ccc Docket No. NHTSA-2010-0131. ddd Docket No. NHTSA-2010-0131. 2-32 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting outlier classification. Being certification data, the chances of large measurement errors should be near zero, particularly towards high CC>2 or fuel consumption. Thus, they can only be outliers in the sense that the vehicle designs are unlike those of other vehicles. These outlier vehicles may include performance vehicles, vehicles with high ground clearance, 4WD, or boxy designs. Given that these are equally legitimate on-road vehicle designs, the agencies concluded that it would appropriate to reconsider the treatment of these vehicles in the regression techniques. Based on these considerations as well as on the adjustments discussed above, the agencies concluded it was not meaningful to run MAD regressions on gpm data that had already been adjusted in the manner described above. Normalizing already reduced the variation in the data, and brought outliers towards average values. This was the intended effect, so the agencies deemed it unnecessary to apply an additional remedy to resolve an issue that had already been addressed, but we sought comment on the use of robust regression techniques under such circumstances. One commenter, ACEEE, addressed this question in this rulemaking, indicating (consistent with the agencies' views) that MAD and OLS are both technically sound methods for fitting functions. 2.4.2.7 Sales Weighting Likewise, in the proposal, the agencies reconsidered the application of sales-weighting to represent the data. As explained below, the decision to sales weight or not is ultimately based upon a choice about how to represent the data, and not by an underlying statistical concern. Sales weighting is used if the decision is made to treat each (mass produced) unit sold as a unique physical observation. Doing so thereby changes the extent to which different vehicle model types are emphasized as compared to a non-sales weighted regression. For example, while total General Motors Silverado (332,000) and Ford F-150 (322,000) sales differed by less than 10,000 in MY 2021 market forecast used for the NPRM, 62 F-150s models and 38 Silverado models were reported in the agencies baselines. Without sales- weighting, the F-150 models, because there were more of them, were given 63 percent more weight in the regression despite comprising a similar portion of the marketplace and a relatively homogenous set of vehicle technologies. The agencies did not use sales weighting in the MYs 2012-2016 rulemaking analysis of the curve shapes. A decision to not perform sales weighting reflects judgment that each vehicle model provides an equal amount of information concerning the underlying relationship between footprint and fuel economy. Sales-weighted regression gives the highest sales vehicle model types vastly more emphasis than the lowest-sales vehicle model types thus driving the regression toward the sales-weighted fleet norm. For unweighted regression, vehicle sales do not matter. The agencies note that the light truck market forecast shows MY 2025 sales of 218,000 units for Toyota's 2WD Sienna, and shows 66 model configurations with MY 2025 sales of fewer than 100 units. Similarly, the agencies' market forecast shows MY 2025 sales of 267,000 for the Toyota Prius, and shows 40 model configurations with MY2025 sales of fewer than 100 units. Sales-weighted analysis would give the Toyota Sienna and Prius more than a thousand times the consideration of many vehicle model configurations. Sales-weighted analysis would, therefore, cause a large number of vehicle model configurations to be virtually ignored in the regressions.14 2-33 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting However, the agencies did note in the MYs 2012-2016 final rules that, "sales weighted regression would allow the difference between other vehicle attributes to be reflected in the analysis, and also would reflect consumer demand." 15 In reexamining the sales-weighting for this analysis, the agencies note that there are low-volume model types that account for many of the passenger car model types (50 percent of passenger car model types account for 3.3 percent of sales), and it is unclear whether the engineering characteristics of these model types should equally determine the standard for the remainder of the market. In the interest of taking a fresh look at appropriate methodologies as promised in the last final rule, in developing proposed and final standards for MYs 2017-2025, the agencies gave full consideration to both sales-weighted and unweighted regressions. 2.4.2.8 Analyses Performed We performed regressions describing the relationship between a vehicle's CCVfuel consumption and its footprint, in terms of various combinations of factors: initial (raw) fleets with no technology, versus after technology is applied; sales-weighted versus non-sales weighted; and with and without two sets of normalizing factors applied to the observations. The agencies excluded diesels and dedicated AFVs because the agencies anticipate that advanced gasoline-fueled vehicles are likely to be dominant through MY2025. Results supporting development of the proposed and finalized standards are depicted graphically in Figures 2-9 through 2-16, below. Thus, the basic OLS regression on the initial data (with no technology applied) and no sales-weighting represents one perspective on the relation between footprint and fuel economy. Adding sales weighting changes the interpretation to include the influence of sales volumes, and thus steps away from representing vehicle technology alone. Likewise, MAD is an attempt to reduce the impact of outliers, but reducing the impact of outliers might perhaps be less representative of technical relationships between the variables, although that relationship may change over time in reality. Each combination of methods and data reflects a perspective, and the regression results reflect that perspective in a simple quantifiable manner, expressed as the coefficients determining the line through the average (for OLS) or the median (for MAD) of the data. It is left to policy makers to determine an appropriate perspective and to interpret the consequences of the various alternatives. The agencies sought comment on the application of the weights as described above, and the implications for interpreting the relationship between fuel efficiency and footprint. ACEEE questioned adjustment of the light truck data based on differences in weight/footprint, indicating that, in their view, the adjustment produces too steep a slope and potentially implies overstatement of the efficacy of some technologies as applied to pickup trucks. ACEEE also suggested that adjustment based on differences in power/weight would yield flatter curves and be more consistent with how the EU constructed related CO2 targets. The Alliance, in contrast, supported the weightings applied by the agencies, and the resultant relationships between fuel efficiency and footprint. Both ACEEE and the Alliance commented that the agencies should revisit the application of weights—and broader aspects of analysis to develop mathematical functions—in the future. Moreover, although ACEEE 2-34 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting expressed concern regarding the outcomes of the application of the weight/footprint adjustment, the agencies maintain that the adjustments (including no adjustments) considered in the NPRM are all potentially reasonable to apply for purposes of developing fuel economy and GHG target curves. This issue is discussed in greater detail in Section II.C of the preamble, and related issues-the slope and stringency of the light truck standards—are addressed further in Sections III and IV of the preamble. 2.4.2.9 What results did the agencies obtain? Both agencies employed the same statistical approaches. For regressions against data including technology normalization, NHTSA used the CAFE modeling system, and EPA used EPA's OMEGA model. The agencies obtained similar regression results, and based the joint proposal on those obtained by NHTSA. For illustrative purposes, the set of figures below show the range of curves determined by the possible combinations of regression techniques, with and without sales weighting, with and without the application of technology, and with various adjustments to the gpm variable prior to running a regression. Again, from a statistical perspective, each of these regressions simply represents the assumptions employed. Since they are all univariate linear regressions, they describe the line that will result from minimizing the sum of the residuals (for MAD) or sum of squared residuals (for OLS). Figures show the results for passenger cars, then light trucks, for ordinary least squares (OLS) then similar results for MAD regressions for cars and light trucks, respectively. The various equations are represented by the string of attributes used to define the regression. See the table, Regression Descriptors, below, for the legend. Thus, for example, the line representing "ols_LT_wt_ft_adj_init_w" should be read as follows: an OLS regression, for light trucks, using data adjusted according to weight to footprint, no technology added, and weighted by sales. Table 2-6 Regression Descriptors Notation ols or mad PC or LT hp wt adj wt ft adj wt ft hp wt adj init or final u or w Description Ordinary least squares or mean absolute deviation Passenger car or light truck Adjustment for horsepower to weight Adjustment for weight to footprint Adjustment for both horsepower to weight and weight to Vehicles with no technology (initial) or with technology footprint added (final) Unweighted or weighted by sales 2-35 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Thus, the next figures, for example, represent a family of curves (lines) fit using ordinary least squares on data for passenger cars, not modified for technology, and which therefore permits comparisons of results in terms of the factors that change in each regression. These factors are whether the data are sales-weighted (denoted "w") or unweighted (denoted "u"), as well as the adjustments described above. Each of these adjustments has an influence on the regressions results, depicted in the figures below. Updating this analysis using the revised MY 2008- and the MY 2010-based fleet projection yielded results generally similar to those shown above. See section 2.6 below. Detailed results of the analysis with the final rulemaking fleet projections are presented in a memorandum available in NHTSA's docket.666 ' Docket No. NHTSA-2010-0131. 2-36 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting 0.01 •55 45 75 85 55 65 Footprint olsPC init_w olsPC init_u — olsPC_hp_wt_adj init_w ^^— olsPC_hp_wt_adj init_u olsPC_wt_ft_hp_wt_adj init_w olsPC_wtJt_hp_wt_adj_init_u o Is P C_w t_f t_a clj i n i t_w o I s P C_ w t_f t_a dj i n it_u o No Technology Fleet n Technology Fleet Figure 2-9 Best Fit Results for Various Regressions: Cars, No Added Technology, OLS Figure 2-10, below, shows comparable results, this time with data representing the additional technology that has been added to reduce technological heterogeneity. Note that the data now pass through the relevant data "cloud" for the fleet with the technology 2-37 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting adjustment applied. The slopes of the lines are somewhat more clustered (less divergent) in the chart depicting added technology (as discussed in footnote ii) o.oi 25 35 •olsPC final_w •olsPC_hp_wt_adj final_w 55 65 75 R««*PtiR,t»C_final_u olsPC_hp_wt_aclj final_u 85 ^^— o I s P C_w t_f t_hp_wt_a clj f i na l_w o I s P C_ w t_f t_hp_ wt_a clj f i na l_u olsPC_wt_ft_aclj final_w - olsPC_wt_ft_adj final_u o No Technology Fleet o Technology Fleet Figure 2-10 Best Fit Results for Various Regressions: Cars, with Added Technology, OLS Similar to the figures displaying the results for passenger cars, the figures below display regression lines for trucks, first with no technology added, then subsequently, for the case where technology has been added. Slopes appear more similar to each other here than of passenger cars. 2-38 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting 0.01 25 35 ^^— o I s LT i n i t_w olsLT_hp_wt_adj init_w —— olsLT_hp_wt_aclj init_u ol\$LT_wt_ft_hp_wt_adj init_w olsl_T_wt_ft_hp_wt_adj init_u olsLT_wt_ft_adj_init_w olsLT_wt_ft_adj init_u O No Technology Fleet D Technology Fleet Figure 2-11 Best Fit Results for Various Regressions: Trucks, No Added Technology, OLS 2-39 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting 0.01 75 85 25 35 45 55 65 ol\$LT final_w olsLT final_u «^^ olsLT_hp_wt_atlj f inal_w ^^ olsl_T_hp_wt_adj final_u ol\$LT_wt_ft_hp_wt_adj final_w olsl_T_wt_ft_hp_wt_adj final_u ol\$LT_wt_ft_adj finaljw olsLT_vvt_ft_adj final_u o No Technology Fleet n Technology Fleet Figure 2-12 Best Fit Results for Various Regressions: Trucks, With Added Technology, OLS Figure 2-13, below, displays regression results for the passenger car MAD fitted curves. The technology adjustment does not have, however, the same degree of impact in 2-40 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting reducing the difference in the attained slopes (between those with and without the addition of technology) evidenced in the OLS regressions. 25 35 - maclPC_init_w ^— maclPC_hp_wt_adj_init_w — m a d P C_w t_f t_hp_v»' t_a clj_i n i t_ w madPC_wt_ft_adj_init_w o No Technology Fleet madPC_hp_wt_adj_init_u m a d P C_ w t_f t_hp_w t_a dj_i n i t_u iiiadPC_wt_ft_adj_Wt_u n Technology Fleet Figure 2-13 Best Fit Results for Various Regressions: Cars, No Added Technology, MAD 2-41 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting 75 85 maclPC_final_w 45 55 65 Footprint madPC_final_u v ^—madPC_hp_wt_adj_final_u maclPC_wt_ft_hp_wt_adj_final_w madPC_wt_ft_hp_wt_adj_final_u madPC_wt_ft_adj_final_w madPC_wt_ft_aclj_final_u o No Technology Fleet P Technology Fleet Figure 2-14 Best Fit Results for Various Regressions: Cars, Added Technology, MAD The MAD regression results below in Figure 2-15 show a grouping of the fitted lines similar to that displayed in the OLS fits for trucks. As expected, an additional reduction in 2-42 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting divergence is seen in the case where technology has been added, in Figure 2-15, which can be ascribed to the reduction in heterogeneity of the fleet brought about by the addition of the technology. 25 35 45 madLT_init_w madLT_hp_Vv't_aclj_init_w 55 65 75 85 m a d LT_ w t_f t_a clj_i n i t_w O No Technology Fleet maclLT_h|>_wt_aclj_init_u madLT_wt_ft_hp_wt_adj_init_u mad LT_ w t_f t_a cl j_i n i t_u D Technology Fleet Figure 2-15 Best Fit Results for Various Regressions: Trucks, No Added Technology, MAD 2-43 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting 0.01 25 35 •maclLT_final_w •mad LT_h p_ w t_a dj_f i n a l_w madLT_wt_ft_adj_final_w O No Technology Fleet Footprint madLT_flnal_u madLT_hp_wt_adj_final_u w-^—madLT_wt_ft_hp_wt_adj_final_u niadLT_wt_ft_adj_final_n D Technology Fleet Figure 2-16 Best Fit Results for Various Regressions: Trucks, with Added Technology, MAD Updating this analysis using the revised MY 2008- and the MY 2010-based fleet projections yielded results generally similar to those shown above. Detailed results of the 2-44 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting analyses with the final rulemaking fleet projections are presented in a memorandum available inNHTSA'sdocket.^ 2.4.2.10 Which methodology did the agencies choose for the proposal, and why was it reasonable? For the proposal, the choice among the alternatives presented above was to use the OLS formulation, on sales-weighted data, using a fleet that has had technology applied, and after adjusting the data for the effect of weight-to-footprint, as described above. The agencies believe that this represented a technically reasonable approach for purposes of developing target curves to define the proposed standards, and that it represents a reasonable trade-off among various considerations balancing statistical, technical, and policy matters, which include the statistical representativeness of the curves considered and the steepness of the curve chosen. The agencies judged the application of technology prior to curve fitting to provide a reasonable means—one consistent with the rule's objective of encouraging manufacturers to add technology in order to increase fuel economy and reduce GHG emissions—of reducing variation in the data and thereby helping to estimate a relationship between fuel consumption/CO2 and footprint. Similarly, for the agencies' NPRM MY 2008-based market-forecast and the agencies' estimates of future technology effectiveness, the inclusion of the weight-to-footprint data adjustment prior to running the regression also helped to improve the fit of the curves by reducing the variation in the data, and the agencies believed that the benefits of this adjustment for the proposed rule likely outweighed the potential that resultant curves might somehow encourage reduced load carrying capability or vehicle performance (note that we were not suggesting that we believed these adjustments would reduce load carrying capability or vehicle performance). In addition to reducing the variability, the truck curve was also steepened, and the car curve flattened compared to curves fitted to sales weighted data that do not include these normalizations. The agencies agreed with manufacturers of full-size pick-up trucks that in order to maintain towing and hauling utility, the engines on pick-up trucks must be more powerful, than their low "density" nature statistically suggested based on the agencies' NPRM MY 2008-based market forecast and the agencies' estimates of the effectiveness of different fuel-saving technologies. Therefore, the agencies judged that it may be more appropriate (i.e., in terms of relative compliance challenges faced by different light truck manufacturers) to adjust the slope of the curves defining fuel economy and CC>2 targets. The results of the normalized regressions are displayed in Table, below.ggg Table 2-7 Regression Results fff Docket No. NHTSA-2010-0131. 888 As presented in the draft TSD supporting the NPRM, this table erroneously reported coefficients from the regression using normalization based on differences in horsepower to weight rather than differences in weight per footprint. The differences in this Table as presented in this final TSD reflect this correction. 2-45 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Vehicle Passenger cars Light trucks Slope (gallons/mile) 0.00037782 0.00038891 Constant (gallons/mile) 0.00181033 0.00401336 Updating this analysis using the corrected MY 2008- and the MY 2010-based fleet projection yielded results generally similar to those shown above. Detailed results of the analyses with the final rulemaking fleet projections are presented in a memorandum available inNHTSA'sdocket.'" hhh As described above, however, other approaches are also technically reasonable, and also represent a way of expressing the underlying relationships. The agencies revisited the analysis for the final rule, after correcting the underlying MY 2008 based market forecast, developing a MY 2010 based market forecast, updating estimates of technology effectiveness and cost, and after considering relevant public comments. As presented below in section 2.6, results of these updated analyses were generally similar to those supporting the NPRM analysis results, and the agencies' balancing of considerations led the agencies to select final curves unchanged from the NPRM curves. As shown in the figures below, the line represents the sales-weighted OLS regression fit of gallons per mile regressed on footprint, with the proposal data first adjusted by weight to footprint, as described above. This introduces weight as an additional consideration into the slope of the footprint curve, although in a manner that adjusts the data as described above, and thus maintains a simple graphical interpretation of the curve in a two dimensional space (gallons per mile and footprint). hhh Docket No. NHTSA-2010-0131. 2-46 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting GPM vs. Footprint - Cars Figure 2-17 Gallons per Mile versus Footprint, Cars (Data adjusted by weight to footprint). 2-47 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting GPM vs. Footprint - Trucks Figure 2-18 Gallons per Mile versus Footprint, Trucks (data adjusted by weight to footprint). Updating this analysis using the revised MY 2008- and the MY 2010-based fleet projection yielded results generally similar to those shown above. Detailed results of the analyses with the final rulemaking fleet projections are presented in a memorandum available inNHTSA'sdocket.1" In the preceding two figures, passenger car and light truck data is represented for the specification chosen, with the size of the observation scaled to sales. The agencies note with regard to light trucks that for the MYs 2012-2016 analysis NPRM and final rule analyses, some models of pickups are aggregated , when, for example, the same pickup had been available in different cab configurations with different wheelbases.16 For the analysis presented above, these models have been disaggregated and are represented individually, which leads to a slightly different outcome in the regression results than had they remained aggregated. 111 Docket No. NHTSA-2010-0131. 2-48 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting 2.4.2.11 Implications of the adopted slopes compared to the slopes in MYs 2012-2016 Rules The slope first proposed, and now adopted by the agencies has several implications relative to the MY 2016 curves, with the majority of changes affecting the truck curve. The selected car curve has a slope similar to that finalized in the MYs 2012-2016 rulemaking (4.7 g/mile in MY 2016, vs. 4.5 g/mile proposed in MY 2017). By contrast, the truck curve is steeper in MY 2017 than in MY 2016 (4.0 g/mile in MY 2016 vs. 4.9 g/mile in MY 2017). As discussed previously, a steeper slope relaxes the stringency of targets for larger vehicles relative to those for smaller vehicles, thereby shifting relative compliance burdens among manufacturers based on their respective product mix. Comments regarding the slope of the agencies' proposed curves are discussed in Section II.C of the preamble to today's final rule. 2.5 Once the agencies determined the appropriate slope for the sloped part, how did the agencies determine the rest of the mathematical function? The agencies continue to believe that without a limit at the smallest footprints, the function—whether logistic or linear—can reach values that would be unfairly burdensome for a manufacturer that elects to focus on the market for small vehicles; depending on the underlying data, an unconstrained form could result in stringency levels that are technologically infeasible and/or economically impracticable for those manufacturers that may elect to focus on the smallest vehicles. On the other side of the function, without a limit at the largest footprints, the function may provide no floor on required fuel economy. Also, the safety considerations that support the provision of a disincentive for downsizing as a compliance strategy apply weakly, if at all, to the very largest vehicles. Limiting the function's value for the largest vehicles thus leads to a function with an inherent absolute minimum level of performance, while remaining consistent with safety considerations. Just as for slope, in determining the appropriate footprint and fuel economy values for the "cutpoints," the places along the curve where the sloped portion becomes flat, the agencies took a fresh look for purposes of this rulemaking, taking into account the updated market forecasts and new assumptions about the availability of technologies. The next two sections discuss the agencies' approach to cutpoints for the passenger car and light truck curves separately, as the policy considerations for each vary somewhat. 2.5.1 Cutpoints for Passenger Car curve The passenger car fleet upon which the agencies based the proposed target curves for MYs 2017-2025 was derived from MY 2008 data, as discussed above. In MY 2008, passenger car footprints ranged from 36.7 square feet, the Lotus Exige 5, to 69.3 square feet, the Daimler Maybach 62. In that fleet, several manufacturers offer small, sporty coupes below 41 square feet, such as the BMW Z4 and Mini, Honda S2000, Mazda MX-5 Miata, Porsche Carrera and 911, and Volkswagen New Beetle. Because such vehicles represent a 2-49 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting small portion (less than 10 percent) of the passenger car market, yet often have performance, utility, and/or structural characteristics that could make it technologically infeasible and/or economically impracticable for manufacturers focusing on such vehicles to achieve the very challenging average requirements that could apply in the absence of a constraint, EPA and NHTSA again proposed to cut off the sloped portion of the passenger car function at 41 square feet, consistent with the MYs 2012-2016 rulemaking. The agencies recognized that for manufacturers who make small vehicles in this size range, putting the cutpoint at 41 square feet creates some incentive to downsize (i.e., further reduce the size, and/or increase the production of models currently smaller than 41 square feet) to make it easier to meet the target. Putting the cutpoint here may also create the incentive for manufacturers who do not currently offer such models to do so in the future. However, at the same time, the agencies believe that there is a limit to the market for cars smaller than 41 square feet — most consumers likely have some minimum expectation about interior volume, among other things. The agencies thus believe that the number of consumers who will want vehicles smaller than 41 square feet (regardless of how they are priced) is small, and that the incentive to downsize to less than 41 square feet in response to this proposal, if present, will be at best minimal. On the other hand, the agencies note that some manufacturers are introducing mini cars not reflected in the agencies MY 2008-based market forecast, such as the Fiat 500, to the U.S. market, and that the footprint at which the curve is limited may affect the incentive for manufacturers to do so. Above 56 square feet, the only passenger car models present in the MY 2008 fleet were four luxury vehicles with extremely low sales volumes—the Bentley Arnage and three versions of the Rolls Royce Phantom. As in the MYs 2012-2016 rulemaking, NHTSA and EPA therefore proposed again to cut off the sloped portion of the passenger car function at 56 square feet.JJJ While meeting with manufacturers prior to issuing the proposal, the agencies received comments from some manufacturers that, combined with slope and overall stringency, using 41 square feet as the footprint at which to cap the target for small cars would result in unduly challenging targets for small cars. The agencies do not agree. No specific vehicle need meet its target (because standards apply to fleet average performance), and maintaining a sloped function toward the smaller end of the passenger car market is important to discourage unsafe downsizing, the agencies thus proposed to again "cut off the passenger car curve at 41 square feet, notwithstanding these comments. . The agencies discuss the comments that were received for the cutpoints on both passenger car and light truck curves in the next section. JJJ The MY 2010 based market forecast has a similarly small number of cars above a footprint of 56 sq ft. These nine vehicle models include 5 Rolls Royce models, a Maybach 57-S and three BMW vehicles, with fewer than 20,000 total projected sales in any model year during this timeframe. 2-50 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting 2.5.2 Cutpoints for Light Truck curve The light truck fleet upon which the agencies based the proposed target curves for MYs 2017-2025, like the passenger car fleet, was derived from MY 2008 data, as discussed in Section 2.4 above. In MY 2008, light truck footprints ranged from 41.0 square feet, the Jeep Wrangler, to 77.5 square feet, the Toyota Tundra. For consistency with the curve for passenger cars, the agencies proposed to cut off the sloped portion of the light truck function at the same footprint, 41 square feet, although we recognized that no light trucks are currently offered below 41 square feet. With regard to the upper cutpoint, the agencies heard from a number of manufacturers during the discussions leading up to the proposal of the MYs 2017- 2025 standards that the location of the cutpoint in the MYs 2012-2016 rules, 66 square feet, resulted in very challenging targets for the largest light trucks in the later years of that rulemaking (although, because CAFE and GHG standards are based on average performance, manufacturers to not need to ensure that every vehicle model meets its fuel economy and GHG targets). See 76 FR at 74864-65. Those manufacturers requested that the agencies extend the cutpoint to a larger footprint, to reduce targets for the largest light trucks which represent a significant percentage of those manufacturers' light truck sales. At the same time, in re-examining the light truck fleet data, the agencies concluded that aggregating pickup truck models in the MYs 2012-2016 rule had led the agencies to underestimate the impact of the different pickup truck model configurations above 66 square feet on manufacturers' fleet average fuel economy and CO2 levels (as discussed immediately below). In disaggregating the pickup truck model data, the impact of setting the cutpoint at 66 square feet after model year 2016 became clearer to the agencies. In the agencies' view, these comments have a legitimate basis. The agencies' market forecast used at proposal includes about 24 vehicle configurations above 74 square feet with a total volume of about 50,000 vehicles or less during any MY in the 2017-2025 time frame. While a relatively small portion of the overall truck fleet, for some manufacturers, these vehicles are a non-trivial portion of their sales. As noted above, the very largest light trucks have significant load-carrying and towing capabilities that make it particularly challenging for manufacturers to add fuel economy-improving/C(^-reducing technologies in a way that maintains the full functionality of those capabilities.111 Considering manufacturer CBI and our estimates of the impact of the 66 square foot cutpoint for future model years, the agencies determined to adopt curves that transition to a different cut point. While noting that no specific vehicle need meet its target (because standards apply to fleet average performance), we believe that the information provided to us by manufacturers (i.e.., information provided regarding the accumulated impacts, especially on manufacturers' credit balances, of CAFE standards since MY2005 and GHG standards since MY2012) and our own analysis supported the gradual extension of the cutpoint for large light trucks in the proposal from 66 square feet kkk In the MY2010 based market forecast, there are 14 vehicle configurations with a total volume of 130,000 vehicles or less during any MY in the 2017-2025 time frame. This is a similarly small portion of the overall number of vehicle models or vehicle sales. Comments on this issue are discussed in section 0. 2-51 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting in MY 2016 out to a larger footprint square feet before MY 2025. The agencies' analyses with regard to this topic, and how it relates to the stringency of the standards, are presented in preamble sections HID and IV.F and summarized in preamble section II.C. Footprint Distribution by Car and Truck 2000000 - 1500000 - 1000000 - a 500000 - 2000000 - 1500000 - 1000000 - 500000 - 0 - 40 50 60 Footprint 70 Figure 2-19 Footprint Distribution by Car and Truck* *Proposed truck outpoints for MY 2025 shown in red, car outpoints shown in green Updating this analysis using the revised MY 2008- and the MY 2010-based market forecasts yielded results generally similar to those shown above. Detailed results of the analyses with the final rulemaking fleet projections are presented in a memorandum available inNHTSA'sdocketmmm The agencies proposed to phase in the higher cutpoint for the truck curve in order to avoid any backsliding from the MY 2016 standard. A target that is feasible in one model year should never become less feasible in a subsequent model year since manufacturers should have no reason to remove fuel economy-improving/CO2-reducing technology from a vehicle 1 Docket No. NHTSA-2010-0131. 2-52 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting once it has been applied. Put another way, the agencies proposed to not allow "curve crossing" from one model year to the next. In proposing MYs 2011-2015 CAFE standards and promulgating MY 2011 standards, NHTSA proposed and requested comment on avoiding curve crossing, as an "anti-backsliding measure."17 The MY 2016 2-cycle test curves are therefore a floor for the MYs 2017-2025 curves. For passenger cars, which have minimal change in slope from the MY 2012-2016 rulemakings and no change in cut points, there were no curve crossing issues in the proposed (or final) standards. The minimum stringency determination was done using the two-cycle curves. Stringency adjustments for air conditioning and other credits were calculated after curves that did not cross were determined in two-cycle space. The year over year increase in these adjustments cause neither the GHG nor CAFE curves (with A/C) to contact the 2016 curves when charted. The agencies received some comments on the selection of these cutpoints. ACEEE commented that the extension of the light truck cutpoint upward from 66 s.f to 74 s.f. would reduce stringency for large trucks even though there is no safety-related reason to discourage downsizing of these trucks. Sierra Club and Volkswagen commented that moving this cutpoint could encourage trucks to get larger and may be detrimental to societal fatalities. Global Automakers commented that the cutpoint for the smallest light trucks should be set at approximately ten percent of sales (as for passenger cars) rather than at 41 square feet. Conversely, IIHS commented that, for both passenger cars and light trucks, the 41 s.f. cutpoint should be moved further to the left (i.e., to even smaller footprints), to reduce the incentive for manufacturers to downsize the lightest vehicles. The agencies have considered these comments regarding the cutpoint applied to the high footprint end of the target function for light trucks, and we judge there to be minimal risk that manufacturers would respond to this upward extension of the cutpoint by deliberately increasing the size of light trucks that are already at the upper end of marketable vehicle sizes, particularly as gasoline prices may continue to increase in the future. Such vehicles have distinct size, maneuverability, fuel consumption, storage, and other characteristics which differ from vehicles between 43 and 48 square feet, and are likely not be suited for all consumers in all usage scenarios. Further, larger vehicles typically also have additional production costs that make it unlikely that the sales of these vehicles will increase in response to changes in the cutpoint. Therefore, we remain concerned that not to extend this cutpoint to 74 s.f. would fail to take into adequate consideration the challenges to improving fuel economy and CC>2 emissions to the levels required by this final rule for vehicles with footprints larger than 66 s.f, given their increased utility, As noted above, while manufacturers are not required to ensure that every vehicle model meets its target, the agencies are concerned that standards with more stringent targets for large trucks would unduly burden full-line manufacturers active in the market for full-size pickups and other large light trucks, as discussed earlier, and evidenced by the agencies' estimates of differences between compliance burdens faced by OEMs active and not active in the market for full-size 2-53 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting pickups. While some manufacturers have recently indicated1111" that buyers are currently willing to pay a premium for fuel economy improvements, the agencies are concerned that disparities in long-term regulatory requirements could lead to future market distortions undermining the economic practicability of the standards. Absent an upward extension of the cutpoint, such disparities would be even greater. For these reasons, the agencies do not expect that gradually extending the cutpoint to 74 s.f will incentivize the upsizing of large trucks and, thus, believe there will be no adverse effects on societal safety. Therefore, we are promulgating standards that, as proposed, gradually extend the truck curve cutpoint to 74 s.f. We have also considered the above comments by Global Automakers and IIHS on the cutpoints for the smallest passenger cars and light trucks. In our judgment, placing these cutpoints at 41 square feet continues to strike an appropriate balance between (a) not discouraging manufacturers from introducing new small vehicle models in the U.S. and (b) not encouraging manufacturers to downsize small vehicles. 2.5.3 Once the agencies determined the complete mathematical function shape, how did the agencies adjust the curves to develop the proposed standards and regulatory alternatives? The curves discussed above all reflect the addition of technology to individual vehicle models to reduce technology differences between vehicle models before fitting curves. This application of technology was conducted not to directly determine the proposed standards, but rather for purposes of technology adjustments, and set aside considerations regarding potential rates of application (i.e., phase-in caps), and considerations regarding economic implications of applying specific technologies to specific vehicle models. The following sections describe further adjustments to the curves discussed above, that affect both the shape of the curve (section 2.5.3.1), and the location of the curve (2.5.3.2), that helped the agencies determine curves that defined the proposed standards. 2.5.3.1 Adjusting for Year over Year Stringency As in the MYs 2012-2016 rules, the agencies developed curves defining regulatory alternatives for consideration by "shifting" these curves. For the MYs 2012-2016 rules, the agencies did so on an absolute basis, offsetting the fitted curve by the same value (in gpm or g/mi) at all footprints. In developing the proposal for MYs 2017-2025, the agencies reconsidered the use of this approach, and concluded that after MY 2016, curves should be offset on a relative basis—that is, by adjusting the entire gpm-based curve (and, equivalently, the CC>2 curve) by the same percentage rather than the same absolute value. The agencies' estimates of the effectiveness of these technologies are all expressed in relative terms—that is, each technology (with the exception of A/C) is estimated to reduce fuel consumption (the inverse of fuel economy) and CC>2 emissions by a specific percentage of fuel consumption without the technology. It is, therefore, more consistent with the agencies' estimates of 111111 For example, in its June 11, 2012 edition, Automotive News quoted a Ford sales official saying that "fuel efficiency continues to be a top purchaser driver." ("More MPG - ASAP", Automotive News, Jun 11, 2012.) 2-54 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting technology effectiveness to develop standards and regulatory alternatives by applying a proportional offset to curves expressing fuel consumption or emissions as a function of footprint. In addition, extended indefinitely (and without other compensating adjustments), an absolute offset would eventually (i.e., at very high average stringencies) produce negative (gpm or g/mi) targets. Relative offsets avoid this potential outcome. Relative offsets do cause curves to become, on a fuel consumption and CC>2 basis, flatter at greater average stringencies; however, as discussed above, this outcome remains consistent with the agencies' estimates of technology effectiveness. In other words, given a relative decrease in average required fuel consumption or CC>2 emissions, a curve that is flatter by the same relative amount should be equally challenging in terms of the potential to achieve compliance through the addition of fuel-saving technology. On this basis, and considering that the "flattening" occurs gradually for the regulatory alternatives the agencies have evaluated, the agencies conclude that this approach to offsetting the curves to develop year-by-year regulatory alternatives neither re-creates a situation in which manufacturers are likely to respond to standards in ways that compromise highway safety, nor undoes the attribute-based standard's more equitable balancing of compliance burdens among disparate manufacturers. The agencies sought comment on these conclusions, and on any other means that might avoid the potential negative outcomes discussed above. As indicated earlier, ACEEE and the Alliance both expressed support for the application of relative adjustments in order to develop year-over-year increases in the stringency of fuel consumption and CC>2 targets, although the Alliance also commented that this approach should be revisited as part of the mid-term evaluation. 2.5.3.2 Adjusting for anticipated improvements to mobile air conditioning systems The fuel economy values in the agencies' market forecasts are based on the 2-cycle (i.e., city and highway) fuel economy test and calculation procedures that do not reflect potential improvements in air conditioning system efficiency, refrigerant leakage, or refrigerant Global Warming Potential (GWP). Recognizing that there are significant and cost effective potential air conditioning system improvements available in the rulemaking timeframe (discussed in detail below in Chapter 5), the agencies are increasing the stringency of the target curves based on the agencies' assessment of the capability of manufacturers to implement these changes. For the proposed CAFE standards and alternatives, an offset was included based on air conditioning system efficiency improvements, as these improvements are the only improvements that effect vehicle fuel economy. For the proposed GHG standards and alternatives, a stringency increase was included based on air conditioning system efficiency, leakage and refrigerant improvements. As discussed in Chapter 5 of the joint TSD, the air conditioning system improvements affect a vehicle's fuel efficiency or CC>2 emissions performance as an additive stringency increase, as compared to other fuel efficiency improving technologies which are multiplicative. Therefore, in adjusting target curves for improvements in the air conditioning system performance, the agencies adjusted the target curves by additive stringency increases (or vertical shifts) in the curves. For the GHG target curves, the offset for air conditioning system performance is being handled in the same manner as for the MYs 2012-2016 rules. For the CAFE target curves, NHTSA for the first time is accounting for potential improvements in air conditioning system 2-55 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting performance. Using this methodology, the agencies first use a multiplicative stringency adjustment for the sloped portion of the curves to reflect the effectiveness on technologies other that air conditioning system technologies, creating a series of curve shapes that are "fanned" based on two-cycle performance. Then the curves are offset vertically by the air conditioning improvement by an equal amount at every point. 2.6 What does the agencies' updated analysis indicate? As discussed above in Chapter 1, the agencies have used two different market forecasts to conduct analyses supporting today's final rule. The first, referred to here as the "MY 2008-Based Fleet Projection," is largely identical to that used for analysis supporting the NPRM, but includes some corrections (in particular, to the footprint of some vehicle models) discussed in Chapter 1 of this TSD. The second, referred to here as the "MY 2010-Based Fleet Projection," is a post-proposal market forecast based on the MY 2010 fleet of vehicles; the development of this 2010 based fleet projection is discussed in Chapter 1. Having made these changes, the agencies repeated the normalization and statistical analyses describe above, following the same approaches as used in the analysis supporting the NPRM. The tables and charts that follow compare the results of NHTSA's updated analysis to those of NHTSA's prior analysis, and compare the resultant fitted lines to the lines (one each for passenger cars and light trucks) selected for purposes of developing the proposed attribute-based standards. The charts below present details of the results in graphical form. 2-56 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Tialized for Technology Differences i_ o 2 Yes Yes Yes Yes Yes Yes Yes Yes No No No No No No No No Yes Yes Table 2-8 Fitted Coefficients (Slope in gpm/sf, Intercept in gpm), Passenger Cars 4-1 *-• E :§ & tt tt QJ Q rtJ TO ^ £ is is oo QJ tuO QJ QJ u_ u_ 5 ';r *— *— 4-< 4-< n -2! O O QJ QJ 0 g u. u. ^ _* ,j_j ,j_j •"— •"— in in i- i- QJ QJ TO TO to ~o ~O " " ^ ^ 'w QJ QJ £- £- >s to to £ £ QJ .^ T3 T3 -_ TO TO OJOJ S " >" ^^•^(--2 0 0 ^ ^ ^ "a"a"&CQ: > > , , , MM'QJO^ ^ ^ •£ ^ ^ — — >~^ ^ ^ CL CL CL TO TO > to ' ' ' QJ QJ QJ EFw^QJ QJ QJ H f H i_ O ^ No No No No Yes Yes Yes Yes No No No No Yes Yes Yes Yes No No i_ o ^ No No No No No No Yes Yes No No No No No No Yes Yes Yes Yes QJ TO 0) Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No bfl QJ C£. OLS OLS MAD MAD OLS OLS OLS OLS MAD MAD OLS OLS OLS OLS OLS OLS OLS OLS CL O 0) 0.000648 0.000513 0.000725 0.000359 0.000431 0.000399 0.000161 0.000264 0.001486 0.000942 0.001345 0.001109 0.000984 0.000920 0.000481 0.000669 0.000378 0.000378 CL O 0) 0.000510 0.000464 0.000560 0.000334 0.000293 0.000351 0.000131 0.000250 0.001220 0.000959 0.001175 0.001085 0.000800 0.000890 0.000452 0.000673 0.000348 0.000362 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. CL O 0) 000472 000502 000427 000445 000248 000398 000093 000268 001058 000995 001096 001099 000737 000933 000403 000654 000316 000371 QJ — -0.01027 0.00009 -0.01408 0.00610 -0.00052 0.00336 0.01155 0.00844 -0.03401 -0.00507 -0.02766 -0.01122 -0.01144 -0.00579 0.01103 0.00367 0.00181 0.00517 QJ 4-1 f— -0.00450 0.00184 -0.00699 0.00650 0.00520 0.00508 0.01238 0.00873 -0.02131 -0.00572 -0.01974 -0.00983 -0.00299 -0.00425 0.01242 0.00358 0.00268 0.00550 QJ 4-1 f— -0.00376 -0.00076 -0.00210 0.00076 0.00643 0.00221 0.01349 0.00736 -0.01670 -0.00944 -0.01806 -0.01259 -0.00176 -0.00785 0.01336 0.00319 0.00330 0.00440 Note 1: Coefficients selected for NPRM shown underlined. Note 2: "MY2008-Based Fleet Projection" refers to market forecast developed using (a) MY2008 vehicle models and characteristics, (b) AEO2011-based overall passenger car and light truck volumes, and (c) manufacturer- and segment-level shares from forecast provided late 2009 by CSM (now owned by Global Insight). Note 3: "MY2010-Based Fleet Projection" refers to market forecast developed using (a) MY2010 vehicle models and characteristics, (b) AEO2012-based overall passenger car and light truck volumes, and (c) manufacturer- and segment-level shares from forecast provided late 2011 by J.D. Power (automotive forecasting service now owned by LMC). 2-57 ``````------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting zed for Technology Differences TO E o 2 Yes Yes Yes Yes Yes Yes Yes Yes No No No No No No No No Yes Yes 4-1 tuO '01 Ol o CL- IO Ol u CU CU it Q o M— T3 Ol M TO E o 2 No No No No Yes Yes Yes Yes No No No No Yes Yes Yes Yes No No H 1 - 1 zed for Differences in Weight/Footprint ^ TO E o 2 No No No No No No Yes Yes No No No No No No Yes Yes Yes Yes 2-9 Fitted Coefficients (Slope in gpm/sf, Intercept in gpm), Light Trucks 4-1 10 TO 1/11/1 cu TO TO ;: <_> u O cu cu u- O O OJ LL. LL. -^ OJ OJ ro TO TO in -o 5 5 's 1/1 Ol .W) ~O "O -^ TO T 4-> 10 cu TO I/) Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No to 10 cu tuO CU C£ OLS OLS MAD MAD OLS OLS OLS OLS MAD MAD OLS OLS OLS OLS OLS OLS OLS OLS ------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting iO 35 40 45 50 55 60 65 70 75 80 -sales-weighted OLS -OLS -sales-weighted MAO -MAD -sales-weighted OLS w/ pw norm -OLS w/pw norm -sales-weighted OLS w/pw andIh/'sf norm -OLS w/ pw and ll>/sf norm - sales-weighted MAD, no tech. -MAD, no tech -sales-weighted OLS, no tech. -OLS, no tech. "sales-weighted OLS, no tech., w/ pw norm - OLS, no tech., w/ pw norm sales-weigh ted OLS, no tech.. w/' pw and Ib/sf norm -OLS, no tech., w/ pw and Ib/sf norm sales-weigh ted OLS w/ Ih/sf norm OLS w/ Ib/sf norm •coefficients selected for NPRM Figure 2-20 Fitted Lines, Passenger Cars, NPRM Analysis 2-59 ------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting iO 35 40 45 50 55 60 65 70 75 80 -sales-weighted OLS -OLS -sales-weighted MAO -MAD -sales-weighted OLS w/ pw norm -OLS w/pw norm -sales-weighted OLS w/pw andIh/'sf norm -OLS w/ pw and ll>/sf norm - sales-weighted MAD, no tech. -MAD, no tech -sales-weighted OLS, no tech. -OLS, no tech. -sales-weighted OLS, no tech., w/ pw norm - OLS, no tech., w/ pw norm sales-weigh ted OLS, no tech.. w/' pw and Ib/sf norm -OLS, no tech., w/ pw and Ib/sf norm sales-weigh ted OLS w/ Ih/sf norm OLS w/ Ib/sf norm •coefficients selected for NPRM Figure 2-21 Fitted Lines, Passenger Cars, Corrected MY2008-Based Market Forecast Note 1: Line based on coefficients selected for NPRM shown for comparison. Note 2: "MY2008-Based Fleet Projection" refers to market forecast developed using (a) MY2008 vehicle models and characteristics, (b) AEO2011-based overall passenger car and light truck volumes, and (c) manufacturer- and segment-level shares from forecast provided late 2009 by CSM (now owned by Global Insight). 2-60 ------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Si 0.06 30 35 40 45 50 55 60 65 70 75 80 -sales-weighted OLS -OLS -sales-weighted MAD -MAD -sales-weighted OLS w/ pw noun -OLS w/pw norm - sales-weighted OLS w/' pw and Ib/sf norm -OLS w/ pw and Ib/sf norm - sales-weighted MAD, no tech. -MAD, no tech -sales-weighted OLS, no tech. - OLS, no tech. - sales-weighted OLS, no tech., w/ pw norm -OLS, no tech., w/ pw norm sales-weighted OLS, no tech., w/ pw and Ib/sf norm - OLS, no tech., w/ pw and Ib/sf norm sales-weighted OLS w/ Ib/sf norm OLS w/ H)/sf norm •coefficients selected forNPRM Figure 2-22 Fitted Lines, Passenger Cars, MY2010-Based Market Forecast Note 1: Line based on coefficients selected for NPRM shown for comparison. Note 2: "MY2010-Based Fleet Projection" refers to market forecast developed using (a) MY2010 vehicle models and characteristics, (b) AEO2012-based overall passenger car and light truck volumes, and (c) manufacturer- and segment-level shares from forecast provided late 2011 by J.D. Power (automotive forecasting service now owned by LMC). 2-61 ------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting ^2 0.06 30 35 40 45 50 55 60 65 70 75 80 -sales-weighted OLS -OLS -sales-weighted MAD -MAD -sales-weighted OLS w/ pw noun -OLS w/pw norm -sales-weighted OLS w/' pw and Ib/sf norm -OLS w/ pw and Ib/sf norm - sales-weighted MAD, no tech. -MAD, no tech -sales-weighted OLS, no tech. - OLS, no tech. - sales-weighted OLS, no tech., w/ pw norm -OLS, no tech., w/ pw norm sales-weighted OLS, no tech., w/ pw and Ib/sf norm - OLS, no tech.. w/ pw and Ib/sf norm sales-weigh ted OLS w/ Ib/sf norm OLS w/ Ib/sf norm •coefficients selected forNPRM Figure 2-23 Fitted Lines, Light Trucks, NPRM Analysis 2-62 ------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Si 0.06 30 35 40 45 50 55 60 65 70 75 80 -sales-weighted OLS -OLS -sales-weighted MAD -MAD -sales-weighted OLS w/ pw noun -OLS w/pw norm - sales-weighted OLS w/' pw and Ib/sf norm -OLS w/ pw and Ib/sf norm - sales-weighted MAD, no tech. -MAD, no tech -sales-weighted OLS, no tech. - OLS, no tech. - sales-weighted OLS, no tech., w/ pw norm -OLS, no tech., w/ pw norm sales-weighted OLS, no tech., w/ pw and Ib/sf norm - OLS, no tech., w/ pw and Ib/sf norm sales-weighted OLS w/ Ib/sf norm OLS w/ H)/sf norm •coefficients selected forNPRM Figure 2-24 Fitted Lines, Light Trucks, Corrected MY2008-Based Market Forecast Note 1: Line based on coefficients selected for NPRM shown for comparison. Note 2: "MY2008-Based Fleet Projection" refers to market forecast developed using (a) MY2008 vehicle models and characteristics, (b) AEO2011-based overall passenger car and light truck volumes, and (c) manufacturer- and segment-level shares from forecast provided late 2009 by CSM (now owned by Global Insight). 2-63 ------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Si 0.06 •ill }5 40 45 50 55 60 65 70 75 80 Footprint (sf| -sales-weighted OLS -OLS -sales-weighted MAD -MAD -sales-weighted OLS w/ pw noun -OLS w/pw norm - sales-weighted OLS w/ pw and Ib/sf norm -OLS w/ pw and Ib/sf norm - sales-weighted MAD, no tech. -MAD, no tech -sales-weighted OLS, no tech. -• OLS, no tech. - sales-weighted OLS, no tech., w/ pw norm -OLS, no tech., w/ pw norm sales-weighted OLS, no tech., w/ pw and Ib/sf norm - OLS, no tech., w/' pw and Ib/sf norm sales-weighted OLS w/ Ib/sf norm OLS w/ Ib/sf norm •coefficients selected forNPRM Figure 2-25 Fitted Lines, Light Trucks, MY2010-Based Market Forecast Note 1: Line based on coefficients selected for NPRM shown for comparison. Note 2: "MY2010-Based Fleet Projection" refers to market forecast developed using (a) MY2010 vehicle models and characteristics, (b) AEO2012-based overall passenger car and light truck volumes, and (c) manufacturer- and segment-level shares from forecast provided late 2011 by J.D. Power (automotive forecasting service now owned by LMC). As discussed above, the selection of a calibrated functional form—in this case, a specific line expressing a relationship between fuel consumption and footprint—upon which to base attribute-based fuel economy and related GHG standards involves considering not just the apparent range of the relevant technical relationship, but also the potential implications for affected policy issues. The approaches described above provide a range of reasonable means of estimating relationships between observed or adjusted fuel consumption and footprint. Having made corrections to the MY 2008-based fleet projection, and having developed a new MY 2010-based fleet projection, the agencies have obtained results generally similar, albeit not identical, to those obtained for the NPRM analysis. For any given method of estimating these lines, it is unlikely that the agencies could have obtained identical results after changing inputs. Also, there is no reason to expect that the MY 2008- and MY 2010-based fleet projections should produce identical results. Still, these differences were mostly small. Using both the corrected MY 2008-based passenger car market forecast and the new MY 2010-based forecast, three techniques produced fitted passenger car lines very close—in terms of average squared differences within the range of footprints between the 2-64 ------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting selected outpoints discussed above—to those selected for the NPRM: sales-weighted OLS without normalizations for differences in power/weight or weight/footprint, sales-weighted OLS with normalization for differences weight/footprint, and unweighted OLS with normalizations for differences in both power/weight and weight/footprint. For light trucks, two techniques did so for both the corrected MY 2008-based passenger car market forecast and the post-proposal MY 2010-based forecast: unweighted OLS with normalizations for differences in both power/weight and weight/footprint, and unweighted OLS with normalization for differences weight/footprint. Without any normalizations applied to the set of footprint and fuel economy values, unweighted OLS produced fitted slopes within 2% of the values obtained through the corresponding unweighted OLS analysis conducted in support of the NPRM. Also, as the above charts show, the resultant ranges (i.e.., areas in fuel consumption - footprint space) spanned by these methods are similar across the NPRM analysis and the updated analyses using the MY 2008- and MY 2010-based fleet projections. Considering that the agencies have adopted an approach whereby regulatory alternatives are developed by shifting fitted curves on a multiplicative basis, results of several of the techniques evaluated here thus would produce regulatory alternatives virtually identical to those developed for the NPRM. For the method that produced results selected for development of the NPRM, relative adjustment of lines fitted to the corrected MY 2008-based market forecast and the MY 2010-based market forecast produces lines that are, between the footprint cutpoints discussed above (41-56 ft2 and 41-74 ft2 for passenger cars and light trucks, respectively), very close to the lines fitted for the NPRM (FIGURE Label): 2-65 ------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting 0.005 'coefficients selected forNPRM -MY2008-Biise------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting 0030 0.025 £ 0.015 0.010 0005 'coefficients selected forNPRM -MY2008-Biise------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting vehicle attributes (e.g., power/weight or weight/footprint or, plausibly, seating capacity, interior volume, towing capacity, etc.), and statistical techniques (e.g., unweighted, sales- weighted, MAD, OLS). Considering (a) that the reasonable analytical techniques examined by the agencies produce a range of fitted lines, (b) that the future composition of the light vehicle market is subject to some uncertainty, and (c) that other aspects of the agencies' analysis are informed by policy implications, in the agencies' judgment, there is no single analytical method that is the sole "correct" way to establish the two fitted lines (one for passenger cars, one for light trucks) the agencies use to specify final standards. The agencies' updated analysis shows newly-fitted lines producing regulatory alternatives very close to the corresponding regulatory alternatives considered in the NPRM. This confirms that the standards are within the range of technically supportable possibilities. While the agencies' analysis indicates that slopes spanning relatively wide ranges could be technically supportable, the agencies note that the final car standard is very similar to the slope of the MY 2016 standard, despite being based on a different analytical approach than the previous rule. As explained above, the agencies have selected a truck curve differing from that adopted for the previous rule (both slope and upper cut-point); the agencies expect that doing so will account for the future characteristics of the larger (work) trucks, and the manufacturers serving the future market for such trucks. The upper size cut-points for cars, and the lower size cut-point for both cars and trucks, are the same as in the previous rule. Without these adjustments, the agencies' believe that there would either be incentives for manufacturers to reduce the utility of these trucks, or that the manufacturer's compliance costs for reaching the targets would be disproportionately high (Preamble Sections III.C.5 and HID). Thus, in the agencies' judgment, the curves strike a reasonable and appropriate balance between the affected policy considerations—better reflecting the reasonable penetration rates of the technologies needed to achieve the standards and the lead time needed for implementation of those technologies, minimizing the incentive for manufacturers to respond to standards in ways that may either result in decreased utility or compromise safety (by downsizing vehicles with footprints on the sloped portion of mathematical functions defining fuel economy and GHG targets), and encouraging widespread penetration of technologies throughout both the car and light truck fleets at reasonable cost while achieving very significant energy and environmental benefits. Having repeated the analysis documented in the NPRM, and having done so based on two fleets (the corrected MY 2008-based market forecast, and the MY 2010-based market forecast), the agencies have demonstrated that, as proposed, the passenger car and light truck curves are well within technically supportable ranges. Slightly flatter standards would directionally have a potentially compromising effect on the safety-related incentives reflected by the promulgated curves, and potentially force more aggressive penetration of advanced technologies into work trucks in a way that raises issues of both increased cost and consumer acceptance. Conversely, slightly steeper standards would tend to increase the potential that manufacturers would respond to the standards by increasing vehicle size beyond levels the market would otherwise demand, in lieu of applying some fuel-saving technologies. For these reasons, the agencies are today promulgating standards using lines matching those used to develop proposed standards for the NPRM. 2-68 ------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting Additional discussion of the feasibility of the final standards is available in Preamble section HID and IV.F. 2-69 ------- Chapter 2: What are the Attribute-Based Curves the Agencies are Adopting References: UQU.S.C. 32902(a)(3)(A). 2 69 FR 38958 (June 29, 2004). 3 76 FR 57106, 57162-64, (Sept. 15,2011). 4 See 74 FR at 14359 (Mar. 30, 2009). 5 75 FR at 25362. 6 See generally 74 FR at 49491-96; 75 FR at 25357-62. 7 68 FR 74920-74926. 8 74 FR 14359. 9 See 75 FR at 25458 10 75 FR at 25363 1' See 75 FR at 25359. 12 Mat 25362-63. 13 Mat 25363. 14 75 FR at 25362 and n. 64 15 75 FR at 25632/3. 16 See 75 FR at 25354 17 74 Fed. Reg. at 14370 (Mar. 30, 2009). 2-70 ------- Technologies Considered in the Agencies' Analysis Chapter 3: Technologies Considered in the Agencies' Analysis This Chapter of the joint TSD describes the technologies NHTSA and EPA evaluated as potential inputs in their respective models and provides estimates of the technologies' costs, effectiveness and availability. This Chapter also describes, in general terms, how the agencies use these inputs in their respective models. The agencies assume, in this analysis, that manufacturers will add a variety of technologies to each of their vehicle model platforms in order to improve their fuel economy and GHG performance. In order to evaluate CAFE and GHG standards and regulatory alternatives, it is essential to understand what is feasible within the timeframe of the final rule. Determining the technological feasibility of the MYs 2017-2025 standards requires a thorough study of the technologies available to the manufacturers during that timeframe. This chapter includes an assessment of the cost, effectiveness, and the availability, development time, and manufacturability of the technologies within either the normal redesign periods of a vehicle line or in the design of a new vehicle. As we describe below, when a technology can be applied can affect the costs as well as the technology penetration rates (or phase-in caps) that are assumed in the analysis. The agencies considered technologies in many categories that manufacturers could use to improve the fuel economy and reduce CC>2 emissions of their vehicles during the MYs 2017-2025 timeframe. Many of the technologies described in this chapter are available today, are well known, and could be incorporated into vehicles once product development decisions are made. These are "nearer-term" technologies and are identical or very similar to those considered in the MYs 2012-2016 final rule analysis (of course, many of these technologies will likely be applied to the light-duty fleet in order to achieve the 2012-2016 CAFE and GHG standards; such technologies would be part of the 2016 reference case for this analysis51). Other technologies considered may not currently be in production, but are under development and are expected to be in production in the next five to ten years. Examples of these technologies are downsized and turbocharged engines operating at combustion pressures even higher than today's turbocharged engines, and an emerging hybrid architecture mated with an 8 speed dual clutch transmission (DCT)—a combination that is not available today. These are technologies which the agencies believe can, for the most part, be applied both to cars and trucks, and which are expected to achieve significant improvements in fuel economy and reductions in CC>2 emissions at reasonable costs in the MYs 2017 to 2025 timeframe. The agencies note that we did not consider in our analysis technologies that are currently in an initial stage of research because of the uncertainties involved in estimating their costs and effectiveness and in assessing whether the technologies will be ready to implement at significant penetration rates during the timeframe of the MY 2017-2025 standards. Examples a The technologies in the 2016 reference fleet are projections made by EPA's OMEGA model and NHTSA's CAFE model respectively. Some technologies may be significantly represented in this reference fleet and these details can be found in each agency's respective RIAs. 3-2 ------- Technologies Considered in the Agencies' Analysis of such technologies would be camless valve actuation and fuel cell vehicles.b The agencies acknowledge that due to the relatively long period between the date of this final rule and the timeframe of the MY 2017-2025 standards, the possibility exists that new and innovative technologies not considered in this analysis will make their way into the fleet (perhaps even in significant numbers). The agencies plan to assess these technologies afresh, along with all of the technologies considered in this final rule, as part of our mid-term evaluation. 3.1 What Technologies did the agencies consider for the final 2017-2025 standards? The technologies considered for this final rulemaking (FRM) analysis by NHTSA and EPA are briefly described below. They fit generally into five broad categories: engine, transmission, vehicle, electrification/accessory, and hybrid technologies. A more detailed description of each technology, and the technology's costs and effectiveness, is described in greater detail in section 3.4 of this TSD. Types of engine technologies applied in this FRM analysis, consistent with the proposal, analysis to improve fuel economy and reduce CC>2 emissions include the following: • Low-friction lubricants - low viscosity and advanced low friction lubricants oils are now available with improved performance and better lubrication. • Reduction of engine friction losses - can be achieved through low-tension piston rings, roller cam followers, improved material coatings, more optimal thermal management, piston surface treatments, and other improvements in the design of engine components and subsystems that improve engine operation. • Second level of low-friction lubricants and engine friction reduction - As technologies advance between now and the rulemaking timeframe, there will be further developments enabling lower viscosity and lower friction lubricants and more engine friction reduction technologies available. • Cylinder deactivation - deactivates the intake and exhaust valves and prevents fuel injection into some cylinders during light-load operation. The engine runs temporarily as though it were a smaller engine which substantially reduces pumping losses. • Variable valve timing - alters the timing or phase of the intake valve, exhaust valve, or both, primarily to reduce pumping losses, increase specific power, and control residual gases. b Fuel cell vehicles may be especially useful in lieu of full battery electric technology for the larger trucks. However, the agencies are not including this technology in the final rule due to the maturity level of the technology. 3-3 ------- Technologies Considered in the Agencies' Analysis Discrete variable valve lift - increases efficiency by optimizing air flow over a broader range of engine operation which reduces pumping losses. Accomplished by controlled switching between two or more cam profile lobe heights. Continuous variable valve lift - is an electromechanically controlled system in which cam period and phasing is changed as lift height is controlled. This yields a wide range of performance optimization and volumetric efficiency, including enabling the engine to be valve throttled. Stoichiometric gasoline direct-injection technology - injects fuel at high pressure directly into the combustion chamber to improve cooling of the air/fuel charge within the cylinder, which allows for higher compression ratios and increased thermodynamic efficiency. Turbocharging and downsizing - increases the available airflow and specific power level, allowing a reduced engine size while maintaining performance. This reduces pumping losses at lighter loads in comparison to a larger engine. In this FRM, the agencies considered three levels of boosting, 18 bar brake mean effective pressure (BMEP), 24 bar BMEP and 27 bar BMEP, as well as four levels of downsizing, from 14 to smaller 14 or 13, from V6 to 14 and from V8 to V6 and 14. 18 bar BMEP is applied with 33 percent downsizing, 24 bar BMEP is applied with 50 percent downsizing and 27 bar BMEP is applied with 56 percent downsizing. To achieve the same level of torque when downsizing the displacement of an engine by 50 percent, approximately double the manifold absolute pressure (2 bar) is required. Accordingly, with 56 percent downsizing, the manifold absolute pressure range increases up to 2.3 bar. Ricardo states in their 2011 vehicle simulation project report that advanced engines in the 2020- 2025 timeframe can be expected to have advanced boosting systems that increase the pressure of the intake charge up to 3 bar1. Refer to Section 3.3.1.2.24.2 for examples of Ricardo-modeled displacements used for turbocharged and downsized engines in each vehicle class. Exhaust-gas recirculation boost - increases the exhaust-gas recirculation used in the combustion process to increase thermal efficiency and reduce pumping losses. Levels of exhaust gas recirculation approach 25% by volume in the highly boosted engines modeled by Ricardo (this, in turn raises the boost requirement by approximately 25%). This technology is only applied to 24 bar and 27 bar BMEP engines in this FRM. Diesel engines - have several characteristics that give superior fuel efficiency, including reduced pumping losses due to lack of (or greatly reduced) throttling, and a combustion cycle that operates at a higher compression ratio, with a very lean air/fuel mixture, than an equivalent-performance gasoline engine. This technology requires additional enablers, such as NOX trap catalyst after-treatment or selective catalytic reduction NOX after-treatment. 3-4 ------- Technologies Considered in the Agencies' Analysis Types of transmission technologies applied in this FRM, consistent with the proposal, include: • Improved automatic transmission controls - optimizes shift schedule to maximize fuel efficiency under wide ranging conditions, and minimizes losses associated with torque converter slip through lock-up or modulation. • Six- and seven-speed automatic transmissions - the gear ratio spacing and transmission ratio are optimized to enable the engine to operate in a more efficient operating range over a broader range of vehicle operating conditions. • Dual clutch transmission (DCT) - are similar to a manual transmission, but the vehicle controls shifting and launch functions. A dual-clutch automated shift manual transmission uses separate clutches for even-numbered and odd-numbered gears, so the next expected gear is pre-selected, which allows for faster, smoother shifting. • Eight-speed automatic transmissions - the transmission gear ratios are optimized to enable the engine to operate in a more efficient operating range over a broader range of vehicle operating conditions. This technology is applied after 2016. • Shift Optimization - tries to keep the engine operating near its most efficient point for a given power demand. The shift controller emulates a traditional Continuously Variable Transmission by selecting the best gear ratio for fuel economy at a given required vehicle power level to take full advantage of high BMEP engines. • Manual 6-speed transmission - offers an additional gear ratio, often with a higher overdrive gear ratio, than a 5-speed manual transmission. • High Efficiency Gearbox (automatic, DCT or manual) - continuous improvement in seals, bearings and clutches, super finishing of gearbox parts, and development in the area of lubrication, all aimed at reducing frictional and other parasitic load in the system for an automatic, DCT or manual type transmission. Types of vehicle technologies applied in this FRM analysis, consistent with the proposal, analysis include: • Low-rolling-resistance tires - have characteristics that reduce frictional losses associated with the energy dissipated in the deformation of the tires under load, thereby reducing the energy needed to move the vehicle. There are two levels of rolling resistance reduction considered in this FRM analysis targeting at 10 percent and 20 percent rolling resistance reduction respectively. • Low-drag brakes - reduce the sliding friction of disc brake pads on rotors when the brakes are not engaged because the brake pads are pulled away from the rotors. • Front or secondary axle disconnect for four-wheel drive systems - provides a torque distribution disconnect between front and rear axles when torque is not required for the non-driving axle. This results in the reduction of associated parasitic energy losses. 3-5 ------- Technologies Considered in the Agencies' Analysis • Aerodynamic drag reduction - is achieved by changing vehicle shape or reducing frontal area, including skirts, air dams, underbody covers, and more aerodynamic side view mirrors. There are two levels of aerodynamic drag reduction considered in this FRM analysis targeting 10 percent and 20 percent aerodynamic drag reduction respectively. • Mass reduction- Mass reduction encompasses a variety of techniques ranging from improved design and better component integration to application of lighter and higher-strength materials. Mass reduction can lead to collateral fuel economy and GHG benefits due to downsized engines and/or ancillary systems (transmission, steering, brakes, suspension, etc.). The maximum mass reduction level considered in this FRM is 20 percent. Types of electrification/accessory and hybrid technologies applied in this FRM include: • Electric power steering (EPS) and electro-hydraulic power steering (EHPS) - is an electrically-assisted steering system that has advantages over traditional hydraulic power steering because it replaces a continuously operated hydraulic pump, thereby reducing parasitic losses from the accessory drive. • Improved accessories (IACC) - There are two levels of IACC applied in this FRM analysis, consistent with the proposal. The first level may include high efficiency alternators, electrically driven (i.e., on-demand) water pumps and cooling systems. This excludes other electrical accessories such as electric oil pumps and electrically driven air conditioner compressors. The second level of IACC includes alternator regenerative braking on top of what are included in the first level of IACC. • Air Conditioner Systems - These technologies include improved hoses, connectors and seals for leakage control. They also include improved compressors, expansion valves, heat exchangers and the control of these components for the purposes of improving tailpipe CC>2 emissions and fuel economy when the A/C is operating. These technologies are covered separately in Chapter 5 of this joint TSD. • 12-volt Stop-start - also known as idle-stop or 12V micro hybrid and commonly implemented as a 12-volt belt-driven integrated starter-generator, this is the most basic hybrid system that facilitates idle-stop capability. Along with other enablers, this system replaces a common alternator with an enhanced power starter- alternator, both belt driven, and a revised accessory drive system. • Higher Voltage Stop-Start/Belt Integrated Starter Generator (BISG) - sometimes referred to as a mild hybrid, BISG provides idle-stop capability and launch assistance and uses a high voltage battery with increased energy capacity over typical automotive batteries. The higher system voltage allows the use of a smaller, more powerful electric motor and reduces the weight of the motor, inverter, and battery wiring harnesses. This system replaces a standard alternator with an enhanced power, higher voltage, higher efficiency belt-driven starter- 3-6 ------- Technologies Considered in the Agencies' Analysis alternator which can recover braking energy while the vehicle slows down (regenerative braking). An example of a BISG system is the GM eAssist introduced in MY 2012. This technology was not included in the analysis for the proposal because we had incomplete information on the technology at that time. Since the proposal, the agencies have obtained better data on the costs and effectiveness of this technology (see 3.4.3.5 of this joint TSD). Therefore, the agencies have revised their technical analysis on both and found that the technology is now competitive with the others in the CAFE model technology decision trees and EPA's technology packages. Further, this technology has been used for "game changing" credit for pick-up trucks and can act as a bridge technology for strong hybrid. For these reasons, the technology is now included in the analysis. • P2 Hybrid- P2 hybrid is a hybrid technology that uses a transmission integrated electric motor placed between the engine and a gearbox or CVT, with a wet or dry separation clutch which is used to decouple the motor/transmission from the engine. In addition, a P2 Hybrid would typically be equipped with a larger electric machine than a mild hybrid system but smaller than a power-split or 2-mode hybrid architecture. Disengaging the clutch allows all-electric operation and more efficient brake-energy recovery. Engaging the clutch allows efficient coupling of the engine and electric motor and based on simulation, when combined with a DCT transmission, provides similar or improved fuel efficiency to other strong hybrid systems with reduced cost. • Plug-in hybrid electric vehicles (PHEV) - are hybrid electric vehicles with the means to charge their battery packs from an outside source of electricity (usually the electric grid). These vehicles have larger battery packs than non-plug-in hybrid electric vehicles with more energy storage and a greater capability to be discharged. They also use a control system that allows the battery pack to be substantially depleted under electric-only or blended mechanical/electric operation, allowing for reduced fuel use during "charge depleting" operation. • Electric vehicles (EV) - are vehicles with all-electric drive and with vehicle systems powered by energy-optimized batteries charged primarily from grid electricity. EVs with 75 mile, 100 mile and 150 mile ranges have been included as potential technologies. Types of accessory/hybridization/electrification technologies discussed but not applied in this FRM analysis, consistent with the proposal, include: • Integrated Motor Assist (IMA)/Crank integrated starter generator (CISG) - provides idle-stop capability and uses a high voltage battery with increased energy capacity over typical automotive batteries. The higher system voltage allows the use of a smaller, more powerful electric motor and reduces the weight of the wiring harness. This system replaces a standard alternator with an enhanced power, higher voltage and higher efficiency starter-alternator that is crankshaft mounted and can recover braking energy while the vehicle slows down 3-7 ------- Technologies Considered in the Agencies' Analysis (regenerative braking). The IMA technology is not included as an enabling technology in this analysis as the industry trends toward more cost effective hybrid configurations, although it is included as a baseline technology because it exists in the baseline fleet. • Power-split Hybrid (PSHEV) - is a hybrid electric drive system that replaces the traditional transmission with a single planetary gearset and two motor/generators. The smaller motor/generator uses the engine to either charge the battery or supply additional power to the drive motor. The second, more powerful motor/generator is permanently connected to the vehicle's final drive and always turns with the wheels, as well as providing regenerative braking capability. The planetary gearset splits engine power between the first motor/generator and the output shaft to either charge the battery or supply power to the wheels. The power-split hybrid technology is not included as an enabling technology in this analysis as the industry is expected to trend toward more cost-effective hybrid configurations, although it is included as a baseline technology because it exists in the baseline fleet. • 2-Mode Hybrid (2MHEV) - is a hybrid electric drive system that uses an adaptation of a conventional stepped-ratio automatic transmission by replacing some of the transmission clutches with two electric motors that control the ratio of engine speed to vehicle speed, while clutches allow the motors to be bypassed. This improves both the transmission torque capacity for heavy-duty applications and reduces fuel consumption and CC>2 emissions at highway speeds relative to other types of hybrid electric drive systems. The 2-mode hybrid technology is not included as an enabling technology in this analysis as the industry is expected to trend toward more cost effective hybrid configurations, although it is included as a baseline technology because it exists in the baseline fleet. 3.2 How did the agencies determine the costs of each of these technologies? 3.2.1 Direct Costs0 3.2.1.1 Costs from Tear-down Studies There are a number of technologies in this analysis that have been cost using the rigorous tear-down method described in this section. As a general matter, the agencies believe that the best method to derive technology cost estimates is to conduct studies involving tear-down and analysis of actual vehicle components. A "tear-down" involves breaking down a technology into its fundamental parts and manufacturing processes by completely disassembling actual vehicles and vehicle subsystems and precisely determining 0 Note that only battery pack and non-battery costs for HEVs, EVs and PHEVs have changed since proposal. All other direct costs are unchanged except for adjustments from 2009 to 2010 dollars. Battery pack and non- battery cost changes are detailed in Section 3.4.3.6. 3-8 ------- Technologies Considered in the Agencies' Analysis what is required for its production. The result of the tear-down is a "bill of materials" for each and every part of the vehicle or vehicle subsystem. This tear-down method of costing technologies is often used by manufacturers to benchmark their products against competitive products. Historically, vehicle and vehicle component tear-down has not been done on a large scale by researchers and regulators due to the expense required for such studies. While tear- down studies are highly accurate at costing technologies for the year in which the study is intended, their accuracy, like that of all cost projections, may diminish over time as costs are extrapolated further into the future because of uncertainties in predicting commodities (and raw material) prices, labor rates, and manufacturing practices. The projected costs may be higher or lower than predicted. Over the past several years, EPA has contracted with FEV, Inc. and its subcontractor Munro & Associates to conduct tear-down cost studies for a number of key technologies evaluated by the agencies in assessing the feasibility of future GHG and CAFE standards. The analysis methodology included procedures to scale the tear-down results to smaller and larger vehicles, and also to different technology configurations. FEV's methodology was documented in a report published as part of the MY 2012-2016 rulemaking process, detailing the costing of the first tear-down conducted in this work (#1 in the below list).2 This report was peer reviewed by experts in the industry and revised by FEV in response to the peer review comments.3 Subsequent tear-down studies (#2-5 in the below list) were documented in follow-up FEV reports made available in the public docket for the MY 2012-2016 rulemaking.4 Since then, FEV's work under this contract has continued. Additional cost studies have been completed for mild hybrid technology and are available for public review.5 The most extensive study, performed after the MY 2012-2016 Final Rule, involved whole-vehicle tear-downs of a 2010 Ford Fusion power-split hybrid and a conventional 2010 Ford Fusion. (The latter served as a baseline vehicle for comparison.) In addition to providing power-split HEV costs, the results for individual components in these vehicles were subsequently used to cost another hybrid technology, the P2 hybrid, which employs similar hardware. This approach to costing P2 hybrids was undertaken because P2 HEVs were not yet in volume production at the time of hardware procurement for tear-down. Finally, an automotive lithium-polymer battery was torn down and costed to provide supplemental battery costing information to that associated with the NiMH battery in the Fusion, because we think automakers are moving to Li-ion battery technologies due to the higher energy and power density of these batteries. This HEV cost work, including the extension of results to P2 HEVs, has been extensively documented in a new report prepared by FEV.6 Because of the complexity and comprehensive scope of this HEV analysis, EPA commissioned a separate peer review focused exclusively on the new tear down costs developed for the HEV analysis. Reviewer comments generally supported FEV's methodology and results, while including a number of suggestions for improvement, many of which were subsequently incorporated into FEV's analysis and final report. The peer review comments and responses are available in the 3-9 ------- Technologies Considered in the Agencies' Analysis rulemaking docket.d'eOver the course of this contract between EPA and FEV, FEV performed teardown-based studies on the technologies listed below. These completed studies provide a thorough evaluation of the new technologies' costs relative to their baseline (or replaced) technologies. 1. Stoichiometric gasoline direct injection (SGDI) and turbocharging with engine downsizing (T-DS) on a DOHC (dual overhead cam) 14 engine, replacing a conventional DOHC 14 engine. 2. SGDI and T-DS on a SOHC (single overhead cam) on a V6 engine, replacing a conventional 3-valve/cylinder SOHC V8 engine. 3. SGDI and T-DS on a DOHC 14 engine, replacing a DOHC V6 engine. 4. 6-speed automatic transmission (AT), replacing a 5-speed AT. 5. 6-speed wet dual clutch transmission (DCT) replacing a 6-speed AT. 6. 8-speed AT replacing a 6-speed AT. 7. 8-speed DCT replacing a 6-speed DCT. 8. Power-split hybrid (Ford Fusion with 14 engine) compared to a conventional vehicle (Ford Fusion with V6). The results from this tear-down were extended to address P2 hybrids. In addition, costs from individual components in this tear- down study were used by the agencies in developing cost estimates for PHEVs and EVs. 9. Mild hybrid with stop-start technology (Saturn Vue with 14 engine), replacing a conventional 14 engine. 10. Fiat Multi-Air engine technology. (Although results from this cost study are included in the rulemaking docket, they were not used by the agencies in this rulemaking's technical analyses because the technology is under a very recently awarded patent and we have chosen not to base our analyses on its widespread use across the industry in the 2017-2025 timeframe.) In addition, FEV and EPA extrapolated the engine downsizing costs for the following scenarios that were based on the above study cases: • Downsizing a SOHC 2 valve/cylinder V8 engine to a DOHC V6. • Downsizing a DOHC V8 to a DOHC V6. • Downsizing a SOHC V6 engine to a DOHC 4 cylinder engine. • Downsizing a DOHC 4 cylinder engine to a DOHC 3 cylinder engine. The agencies have relied on the findings of FEV for estimating the cost of the technologies covered by the tear-down studies. However, we note that FEV based their costs on the assumption that these technologies would be mature when produced in large volumes d ICF, "Peer Review of FEV Inc. Report Light Duty Technology Cost Analysis, Power-Split and P2 Hybrid Electric Vehicle Case Studies", EPA-420-R-11-016, November 2011. e FEV and EPA, "FEV Inc. Report 'Light Duty Technology Cost Analysis, Power-Split and P2 Hybrid Electric Vehicle Case Studies', Peer Review Report - Response to Comments Document", EPA-420-R-11-017, November 2011. 3-10 ------- Technologies Considered in the Agencies' Analysis (450,000 units or more for each component or subsystem). If manufacturers are not able to employ the technology at the volumes assumed in the FEV analysis with fully learned costs, then the costs for each of these technologies would be expected to be higher. There is also the potential for stranded capital if technologies are introduced too rapidly for some indirect costs to be fully recovered. While the agencies consider the FEV tear-down analysis results to be generally valid for the 2017-2025 timeframe for fully mature, high sales volumes, we have had FEV perform supplemental analysis to consider potential stranded capital costs, and have included these in our primary analyses of program costs. The issue of stranded capital is discussed in detail in Section 3.2.2.3 of this TSD. 3.2.1.2 Costs of HEV, PHEV, EV, and FCEVs The agencies have also reconsidered the costs for HEVs, PHEVs, EVs, and FCEVs since the MY 2012-2016 rulemaking and the Technical Assessment Report (TAR) as the result of two issues. The first issue is that electrified vehicle technologies are developing rapidly and we sought to capture the results from the most recent analyses. The second issue is that the analysis for the MYs 2012-2016 final rule employed a single \$/kWh (\$ per kilowatt-hour) estimate, and did not consider the specific vehicle and technology application for the battery when we estimated the cost of the battery.g Specifically, batteries used in HEVs (high power density applications) versus EVs (high energy density applications) need to be considered appropriately to reflect the design differences, the chemical material usage differences, and the differences in cost per kWh as the power to energy ratio of the battery changes for different applications. To address these issues for this final rule, consistent with the proposal, the agencies have used a battery cost model developed by Argonne National Laboratory (ANL) for the Vehicle Technologies Program of the U.S. Department of Energy (DoE) Office of Energy Efficiency and Renewable Energy.7 The model developed by ANL allows users to estimate unique battery pack costs using user customized input sets for different types of electrified powertrains, such as strong hybrid, PHEV and EV. Since the publication of the TAR, ANL's battery cost model has been peer-reviewed and ANL has updated the model to incorporate suggestions from peer-reviewers.8 Further updates have been made to the model since the NPRM and this newly updated model is used in this FRM analysis.9 We discuss our updated battery costs in section in Section 3.4.3.9. As done in the proposal, the agencies developed costs and effectiveness values for the mild and P2 HEV configuration, two different all-electric mileage ranges for PHEVs (20 and 40 in-use miles) and three different mileage ranges for EVs (75, 100 and 150 in-use miles). Details regarding these vehicle technologies are discussed in sections 3.4.3.6.4 and 3.4.3.6.5. f The potential for stranded capital occurs when manufacturing equipment and facilities cannot be used in the production of a new technology. 8 However, we believe that this had little impact on the results of the cost analyses in support of the MYs 2012- 2016 final rule, as the agencies projected that the standards could be met with an increase of less than 2 percent penetration of hybrid technology and no increase in plug-in or full electric vehicle technology. 3-11 ------- Technologies Considered in the Agencies' Analysis 3.2.1.3 Direct Manufacturing Costs Used in the Rulemaking Analysis Building on the MYs 2012-2016 final rule, for the NPRM analysis, the agencies took a fresh look at technology cost and effectiveness values. For this final rule analysis, the direct manufacturing costs employed in the NPRM have been largely retained, although they were updated to 2010\$, and revisions were made to the costs of Li-ion batteries. The battery costs have been updated for the final rule using the latest ANL BatPaC model as discussed above. For costs, the agencies considered both the direct or "piece" costs and indirect costs of individual components of technologies. For the direct costs that were not developed through the FEV tear-down studies, the agencies generally followed a bill of materials (BOM) approach. A bill of materials, in a general sense, is a list of components that make up a system—in this case, an item of fuel economy-improving technology. In order to determine what a system costs, one of the first steps is to determine its components and what they cost. NHTSA and EPA estimated these components and their costs based on a number of sources for cost-related information. The objective was to use those sources of information considered to be most credible for projecting the costs of individual vehicle technologies. For those cost estimates that are fundamentally unchanged since the 2012-2016 final rule and/or the 2010 TAR (we make note of these in Section 3.4, below), we have a full description of the sources used in Chapters of the final joint TSD supporting that rule.10'11 For those costs that have been updated since those analyses (e.g., battery pack cost, costs based on more recent tear down analyses, etc.), we note their sources in Section 3.4, below. We have also considered input from manufacturers and suppliers gathered either through meetings following the 2010 TAR or in comment submitted in response to the 2010 TAR, some of which cannot be shared publicly in detailed form but, where used, we make note of it while protecting its confidentiality. In this final rule analysis, the agencies have not updated the costs based on any confidential information. Note that a summary of comments on the 2010 TAR, with the agencies' responses, was published as a "Supplemental Notice of Intent" in December of 2010.12 As discussed throughout this chapter, the agencies have reviewed, revalidated or updated cost estimates for individual components based on the latest information available. Once costs were determined, they were adjusted to ensure that they were all expressed in 2010 dollars (the NPRM was in 2009 dollars) using the GDP price deflator as described in section 3.2.4. Indirect costs were accounted for using the ICM approach developed by EPA and explained below. NHTSA and EPA also considered how costs should be adjusted to reflect manufacturer learning as discussed below. Additionally, costs were adjusted by modifying or scaling content assumptions to account for differences across the range of vehicle sizes and functional requirements, and the associated material cost impacts were adjusted to account for the revised content, although these adjustments were different for each agency due to the different vehicle subclasses used in their respective models. h The conversion to 2010 dollars has very little impact on costs (the conversion factor to convert from 2009 to 2010 dollars is 1.01). 3-12 ------- Technologies Considered in the Agencies' Analysis 3.2.2 Indirect Costs' 3.2.2.1 Indirect Cost Multiplier Changes since the 2012-2016 FRM and 2010 TAR As discussed in greater detail below, the agencies have revised the markups used to estimate indirect costs. The first change was to normalize the ICM values to be consistent with the historical average retail price equivalent (RPE) of 1.5, rather than the single year that the RTI study examined. This was done by applying a factor of .57.46 to all indirect cost elements. The second change was to re-consider the markup factors and the data used to generate them. The result on this new thinking is to increase the markup in all cases. The final change is the way in which the ICM factors are applied. In previous analyses ICMs were applied to the learned value of direct costs. However, since learning influences direct costs only, the agencies were concerned that this could overstate the impact of learning on total costs. Indirect costs are thus now established based on the initial value of direct costs and held constant until the long-term ICM is applied. This is done for all ICM factors except warranties, which are influenced by the learned value of direct costs. 3.2.2.2 Cost markups to account for indirect costs To produce a unit of output, auto manufacturers incur direct and indirect costs. Direct costs include the cost of materials and labor costs. Indirect costs may be related to production (such as research and development [R&D]), corporate operations (such as salaries, pensions, and health care costs for corporate staff), or selling (such as transportation, dealer support, and marketing). Indirect costs are generally recovered by allocating a share of the costs to each unit of goods sold. Although it is possible to account for direct costs allocated to each unit of goods sold, it is more challenging to account for indirect costs allocated to a unit of goods sold. To make a cost analysis process more feasible, markup factors, which relate total indirect costs to total direct costs, have been developed. These factors are often referred to as retail price equivalent (RPE) multipliers. Cost analysts and regulatory agencies including EPA and NHTSA have frequently used these multipliers to estimate the resultant impact on costs associated with manufacturers' responses to regulatory requirements. The best approach to determining the impact of changes in direct manufacturing costs on a manufacturer's indirect costs would be to actually estimate the cost impact on each indirect cost element. However, doing this within the constraints of an agency's time or budget is not always feasible, and the technical, financial, and accounting information to carry out such an analysis may simply be unavailable. RPE multipliers provide, at an aggregate level, the relative shares of revenues (Revenue = Direct Costs + Indirect Costs + Net Income) to direct manufacturing costs. Using RPE multipliers implicitly assumes that incremental changes in direct manufacturing costs produce common incremental changes in all indirect cost contributors as well as net income. Note that our approach to estimating indirect costs remains unchanged since the proposal. 3-13 ------- Technologies Considered in the Agencies' Analysis A concern in using the RPE multiplier in cost analysis for new technologies added in response to regulatory requirements is that the indirect costs of vehicle modifications are not likely to be the same for different technologies. For example, less complex technologies could require fewer R&D efforts or less warranty coverage than more complex technologies. In addition, some simple technological adjustments may, for example, have no effect on the number of corporate personnel and the indirect costs attributable to those personnel. The use of RPEs, with their assumption that all technologies have the same proportion of indirect costs, is likely to overestimate the costs of less complex technologies and underestimate the costs of more complex technologies. To address this concern, EPA has developed modified multipliers. These multipliers are referred to as indirect cost multipliers (ICMs). In contrast to RPE multipliers, ICMs assign unique incremental changes to each indirect cost contributor ICM = (direct cost + adjusted indirect cost + profit)/(direct cost) Developing the ICMs from the RPE multipliers requires developing adjustment factors based on the complexity of the technology and the time frame under consideration. This methodology was used in the cost estimation for the MYs 2012-2016 final rule. The ICMs were developed in a peer-reviewed report from RTI International and were subsequently discussed in a peer-reviewed journal article.13 Note that the cost of capital (reflected in profit) is included because of the assumption implicit in ICMs (and RPEs) that capital costs are proportional to direct costs, and businesses need to be able to earn returns on their investments. The capital costs are those associated with the incremental costs of the new technologies. As noted above, for the analysis supporting this final rulemaking, consistent with the proposal, the agencies are again using the ICM approach but have made some changes to both the ICM factors and to the method of applying those factors to arrive at a final cost estimate. The first of these changes was done in response to continued thinking among the EPA- NHTSA team about how past ICMs have been developed and what are the most appropriate data sources to rely upon in determining the appropriate ICMs. The second change has been done both due to staff concerns and public feedback suggesting that the agencies were inappropriately applying learning effects to indirect costs via the multiplicative approach to applying the ICMs. Regarding the first change - to the ICM factors themselves - a little background must first be provided. In the original work done under contract to EPA by RTI International,14 EPA staff with extensive experience in the auto industry had undertaken a consensus approach to determining the impact of specific technology changes on the indirect costs of a company. Subsequent to that effort, EPA staff, again with extensive experience in the auto industry, conducted a blind survey to make this determination on a different set of technology changes. This subsequent effort, referred to by EPA as a modified-Delphi approach, resulted in slightly different ICM determinations. This effort is detailed in a memorandum contained in the docket for this rule.15 Upon completing this effort, the EPA team determined that the original RTI values should be averaged with the modified-Delphi values to arrive at the final 3-14 ------- Technologies Considered in the Agencies' Analysis ICMs for low and medium complexity technologies and that the original RTI values would be used for high complexity level 1 while the modified-Delphi values would be used for high complexity level 2. These final ICMs as described were used in the MYs 2012-2016 light- duty GHG/CAFE rulemaking. More recently, EPA and NHTSA decided that the original light-duty RTI values, because of the technologies considered for low and medium complexity, should no longer be used and that we should rely solely on the modified-Delphi values for these complexity levels. The original light-duty RTI study used low rolling resistance tires as a low complexity technology example and a dual clutch transmission as a medium complexity technology. Upon further thought, the technologies considered for the modified Delphi values (passive aerodynamic improvements for low complexity and turbocharging with downsizing for medium complexity) were considered to better represent the example technologies. As a result, the modified-Delphi values became the working ICMs for low and medium complexity rather than averaging those values with the original RTI report values. NHTSA and EPA staff also re-examined the technology complexity categories that were assigned to each light-duty technology and modified these assignments to better reflect the technologies that are now used as proxies to determine each category's ICM value. A secondary-level change was also made as part of this ICM recalculation to the light- duty ICMs. That change was to revise upward the RPE level reported in the original RTI report from an original value of 1.46 to 1.5 to reflect the long term average RPE. The original RTI study was based on 2007 data. However, an analysis of historical RPE data indicates that, although there is year to year variation, the average RPE has remained roughly 1.5. ICMs will be applied to future year's data and therefore NHTSA and EPA staff believe that it would be appropriate to base ICMs on the historical average rather than a single year's result. Therefore, ICMs in this final rulemaking, consistent with the proposal, were adjusted to reflect this average level. As a result, the High 1 and High 2 ICMs have also changed. Table 3-1 shows both the ICM values used in the MYs 2012-2016 final rule and the new ICM values used for the analysis supporting these final rules. Near term values account for differences in the levels of R&D, tooling, and other indirect costs that will be incurred. Once the program has been fully implemented, some of the indirect costs will no longer be attributable to the standards and, as such, a lower ICM factor is applied to direct costs. Table 3-1 Indirect Cost Multipliers Used in this Analysis" Complexity Low Medium Highl High2 20 12-20 16 Rule Near term 1.17 1.31 1.51 1.70 Long term 1.13 1.19 1.32 1.45 This Final rule Near term 1.24 1.39 1.56 1.77 Long term 1.19 1.29 1.35 1.50 a Rogozhin, A., et. al., "Using indirect cost multipliers to estimate the total cost of adding new technology in the automobile industry," International Journal of Production Economics (2009); "Documentation of the Development of Indirect Cost Multipliers for Three Automotive Technologies," Helfand, G., and Sherwood, T., Memorandum dated August 2009; "Heavy Duty Truck Retail Price Equivalent and Indirect Cost 3-15 ------- Technologies Considered in the Agencies' Analysis Multipliers," Draft Report prepared by RTI International and Transportation Research Institute, University of Michigan, July 2010 The second change made to the ICMs has to do with the way in which they are applied. To date, we have applied the ICMs, as done in any analysis that relied on RPEs, as a pure multiplicative factor. This way, a direct manufacturing cost of, say, \$100 would be multiplied by an ICM of 1.24 to arrive at a marked up technology cost of \$124. However, as learning effects (discussed below) are applied to the direct manufacturing cost, the indirect costs are also reduced accordingly. Therefore, in year two the \$100 direct manufacturing cost might reduce to \$97, and the marked up cost would become \$120 (\$97 x 1.24). As a result, indirect costs would be reduced from \$24 to \$20. Given that indirect costs cover many things such as facility-related costs, electricity, etc., it is perhaps not appropriate to apply the ICM to the learned direct costs, at least not for those indirect cost elements unlikely to change with learning. The EPA-NHTSA team believes that it is appropriate to allow only warranty costs to decrease with learning, since warranty costs are tied to direct manufacturing costs (since warranty typically involves replacement of actual parts which should be less costly with learning). The remaining elements of the indirect costs should remain constant year-over- year, at least until some of those indirect costs are no longer attributable to the rulemaking effort that imposed them (such as R&D). As a result, the ICM calculation has become more complex with the analysis supporting this final rule, consistent with the proposal. We must first establish the year in which the direct manufacturing costs are considered "valid." For example, a cost estimate might be considered valid today, or perhaps not until high volume production is reached— which will not occur until MY 2015 or later. That year is known as the base year for the estimated cost. That cost is the cost used to determine the "non-warranty" portion of the indirect costs. For example, the non-warranty portion of the medium complexity ICM in the short-term is 0.343 (the warranty versus non-warranty portions of the ICMs are shown in Table 3-2). For the dual cam phasing (DCP) technology on an 14 engine we have estimated a direct manufacturing cost of \$70 in MY 2015. So the non-warranty portion of the indirect costs would be \$24.01 (\$70 x 0.343). This value would be added to the learned direct manufacturing cost for each year through 2018, the last year of short term indirect costs. Beginning in 2019, when long-term indirect costs begin, the additive factor would become \$18.13 (\$70 x 0.259). Additionally, the \$70 cost in 2015 would become \$67.90 in MY 2016 due to learning (\$70 x (1-3%)). So, while the warranty portion of the indirect costs would be \$3.15 (\$70 x 0.045) in 2015, indirect costs would decrease to \$3.06 (\$67.90 x 0.045) in 2016 as warranty costs decrease with learning. The resultant indirect costs for the DCP-I4 technology would be \$27.16 (\$24.01+\$3.15) in MY 2015 and \$27.07 (\$24.01+\$3.06) in MY2016, and so on for subsequent years. Table 3-2 Warranty and Non-Warranty Portions of ICMs Complexity Low Near term Warranty 0.012 Non-warranty 0.230 Long term Warranty 0.005 Non-warranty 0.187 3-16 ------- Technologies Considered in the Agencies' Analysis Medium Highl High2 0.045 0.065 0.074 0.343 0.499 0.696 0.031 0.032 0.049 0.259 0.314 0.448 There is some level of uncertainty surrounding both the ICM and RPE markup factors. The ICM estimates used in this final rule, consistent with the proposal, group all technologies into three broad categories and treat them as if individual technologies within each of the three categories (low, medium, and high complexity) will have exactly the same ratio of indirect costs to direct costs. This simplification means it is likely that the direct cost for some technologies within a category will be higher and some lower than the estimate for the category in general. Additionally, the ICM estimates were developed using adjustment factors developed in two separate occasions: the first, a consensus process, was reported in the RTI report; the second, a modified Delphi method, was conducted separately and reported in an EPA memorandum. Both these panels were composed of EPA staff members with previous background in the automobile industry; the memberships of the two panels overlapped but were not the same. The panels evaluated each element of the industry's RPE estimates and estimated the degree to which those elements would be expected to change in proportion to changes in direct manufacturing costs. The method and the estimates in the RTI report were peer reviewed by three industry experts and subsequently by reviewers for the International Journal of Production Economics.16 However, the ICM estimates have not yet been validated through a direct accounting of actual indirect costs for individual technologies. RPEs themselves are also inherently difficult to estimate because the accounting statements of manufacturers do not neatly categorize all cost elements as either direct or indirect costs. Hence, each researcher developing an RPE estimate must apply a certain amount of judgment to the allocation of the costs. Since empirical estimates of ICMs are ultimately derived from the same data used to measure RPEs, this affects both measures. However, the value of RPE has not been measured for specific technologies, or for groups of specific technologies. Thus applying a single average RPE to any given technology by definition overstates costs for very simple technologies, or understates them for advanced technologies. The International Council on Clean Transportation (ICCT) and the National Automobile Dealers Association (NADA) commented on our use of ICMs. ICCT supported the ICM approach as presented in the proposal, but argued for removal of sensitivity analyses examining RPEs in NHTSA's FRIA. NADA argued that the ICM approach is not valid and should be replaced with an RPE approach. Further, it argued that the RPE factor should be 2x rather than the 1.5x approach that is supported by filings to the Securities and Exchange Commission. We have conducted a thorough analysis of the NADA comments on the RPE vs. ICM approach. We disagree with NADA's arguments for both using the RPE approach and a 2x RPE factor, for the following reasons. NADA's objections to the ICM approach include: 1. There is no evidence that the RPE method is flawed. 2. The ICMs do not include the total costs of complying with the standards, because it does not include all the costs included in the RPE. 3-17 ------- Technologies Considered in the Agencies' Analysis 3. The ICMs use a subjective judgment to adjust indirect costs for different technologies, while the RPE uses one value for all components and does not rely on "nearly perfect foreknowledge." 4. The ICMs do not incorporate dealer and OEM profits. NADA's arguments for the RPE of 2x include: 5. Several scholarly papers support the use of RPEs in the 2.0 range. 6. A case study comparison of the added content of a 1971 Chevrolet Vega and 2011 Cruze shows that an RPE of 2.0 accounts for the change in retail price. The discussion above provides background on the issue of RPEs and ICMs, and on the agencies' decision to use ICMs to estimate indirect costs for this rulemaking. Our responses here address the specific points raised by NADA. First, the RPE approach applies the same average indirect cost markup across all technologies in the redesigned vehicle fleet, regardless of the source of the direct cost (i.e. whether a technology is simple or complex; whether the source of the additional cost is a new or a mature technology). The RPE methodology also assumes that an indirect cost is associated with the rule, even if no relation is apparent. For instance, the RPEs (until recent union contract changes) would have included the costs to the domestic auto companies of the health insurance for retired auto workers. Because the rulemaking would not affect the current retiree health care costs, (which account for about 1.5% of the RPE), they are irrelevant to the rulemaking. The ICM approach differs in that it allows indirect costs to vary with the complexity of the technology and the time frame.17 It is a reasonable assumption that simple technologies are expected to have fewer indirect costs per dollar than complex technologies. For instance, the use of low-rolling-resistance tires, considered by the EPA/NHTSA team to be a low-complexity technology, adds costs, but, because they require significantly less vehicle integration effort than for example, adding a hybrid powertrain would, the additional indirect costs per dollar of direct manufacturing costs may be very low. In contrast, converting a conventional vehicle to a hybrid-electric is a far more complex activity, involving increases in indirect costs such as research and development disproportionate to its direct costs. Shortly after product introduction, indirect costs for components such as warranty and research may be relatively high, but auto makers are expected to be able to reduce the costs of any specific technology over time, as they gain experience with them and, thus, redirect those expenditures to other areas of their choosing. Second, the ICM approach excludes some costs included in the RPE when those costs are expected not to be affected by the standards. The ICM approach, as discussed above, begins with the RPE and includes all the relevant cost categories. ICMs reflect the indirect costs judged by the EPA panel (see above for further explanation) to be incurred for each technology in response to regulatory imposed changes. Any "omissions", or instances where the ICM carries no costs for a given technology, are cases where the indirect costs are considered by the EPA panel not to be impacted by regulatory imposed changes for that technology. For instance, the costs of switching from a standard tire to a low-rolling- resistance tire (the example of a low-complexity technology in Rogozhin et al. (2009)) are not expected to lead to an increase in transportation costs (i.e., costs for transporting finished 3-18 ------- Technologies Considered in the Agencies' Analysis vehicles from production site to retail site) because it is not expected to be any more expensive to ship a new vehicle with the new tires than with the old tires.18 Third, the RPE approach relies on the assumption that applying the average RPE for the vehicle fleet as a whole will produce a reasonable average indirect cost for all technologies in the redesigned vehicle fleet resulting from these standards. The agencies believe that using the professional judgment and expertise of EPA staff with extensive experience in the auto industry provides useful insight into how a given regulation will impact indirect costs and is an improvement over ignoring differences among technologies. The agencies have therefore based their central analyses on the ICM method. Fourth, it is incorrect that the ICMs do not include profit. Although the initial ICM report reviewed by NRC did not include OEM profit, the ICM approach applied in this rulemaking does incorporate an allowance for profit, at the average corporate profit rate of 6% of sales. The inclusion of profit for the Joint NPRM is discussed in the draft Technical Support Document, and the agencies have included profit as an element of the indirect costs for the final rulemaking as well.19 Fifth, the papers cited to support the use of an RPE of 2x are only a subset of the literature. r\r\ The National Research Council (NRC) discusses the four studies that NADA's Exhibit A cites in its support of an RPE of 2.0. The NRC also notes that NHTSA used an RPE of 1.5 for its MY 2011 fuel economy rule; the NRC in 2002 used an RPE of 1.4, as did the California Air Resources Board; and EPA has used a markup factor of 1.3. The NRC report then discusses work done for the committee itself, doing a detailed analysis of a Honda Accord and a Ford F-150 truck; the former had an RPE of "1.39 to market transaction price and 1.49 to MSRP," and the latter had an RPE of "1.52 for market price and 1.54 for MSRP." Most significantly, the NRC does not recommend an RPE of 2.0. Rather, the NRC recommends, for technologies where the primary manufacturer of the technology is the automotive supply base, an RPE of 1.5, except for hybrid powertrain components from the automotive supply base, where it recommends an RPE of 1.3 due to the inclusion of several indirect costs in their 91 base estimate. Only in the case of technologies where an automotive OEM is the primary manufacturer does the NRC recommend an RPE of 2.0.J We note, without specifically commenting on the quality of the studies, that none of the papers NAD A cites in support of an RPE of 2x was published in a peer-reviewed journal, and none of the studies claim to have been peer-reviewed. In contrast, the research in Rogozhin et al. (2009) was peer-reviewed twice: as documented in the Peer Review Report, and when it was submitted (and accepted) for publication in the InternationalJournal of Production Economics. A full reading of the literature on RPEs thus shows little support for a value of 2x. Further support for an average J Importantly, application of the 2.0s RPE in the "OEM as primary manufacturer" case would be done to a smaller direct cost since the OEM has produced the part in-house and, thus, is not paying the full supplier-level indirect costs that would be included in a part purchased from a supplier. The end result should be a total cost roughly equivalent or less than a 1.5x RPE applied to the supplier-produced part. If not, the manufacturer should probably not produce in-house and should, instead, purchase parts since they would be less costly (all other considerations being equal). 3-19 ------- Technologies Considered in the Agencies' Analysis RPE lower than 2.0 comes from an examination of industry financial statements. NHTSA examined industry 10-K submissions to the Securities and Exchange Commission from the period 1972-1997.k The cost information in these submissions represents all industry operations, including both OEM and supplier-sourced technologies. During this period, the RPE averaged 1.5 while varying slightly, but never dropped below 1.4 or exceeded 1.6. At no time did the average RPE approach the 2.0 value advocated by NADA. The results are shown, together with the 2007 results from Rogozhin et al in the following figure: RPE History, 1972-1997, and 2007 2.00 1.90 1.80 1.70 1.60 1.50 1.40 1.30 1.20 1.10 1.00 V 4-RPE 1970 1975 1980 1985 1990 1995 2000 2005 2010 Sixth, the comparison of the Vega and the Cruze uses circular logic; it assumes its conclusion. The direct costs of the vehicles are calculated using an RPE of 2, and the NADA analysis then calculates a quality difference based on the change in direct costs. The magnitude of the quality difference is then discovered to correspond to an RPE of 2, although it is also an inevitable result of the initial assumption of an RPE of 2. The analysis provided can be replicated with any value of RPE. This argument thus provides no evidence on the value of the RPE. For these reasons, we do not accept NADA's request to use an RPE of 2x., and instead continue with our use of ICMs as the basis for our central analysis. However, the agencies recognize that there is uncertainty regarding the impact on indirect costs of regulatorily imposed changes. For this reason, both agencies have conducted sensitivity analyses using different indirect cost estimates. EPA presents its sensitivities in Chapter 3 of its final RIA. For its part, NHTSA rejects the ICCT proposal to eliminate sensitivity analyses examining the RPE and presents the impact of using the RPE as a basis for indirect costs in its analysis in Chapters 7 and 10 of NHTSA's FRIA. In addition, RPEs are incorporated into the Probabilistic Uncertainty analysis in Chapter 12 of NHTSA's FRIA. Spinney, B.C., Faigin, B.M, Bowie, N.N, Kratzke, S.R., Advanced Air Bag Systems Cost, Weight, and Lead Time Analysis Summary Report, Contract No. DTNH22-96-0-12003, Task Orders - 001, 003, and 005. 3-20 ------- Technologies Considered in the Agencies' Analysis 3.2.2.3 Stranded capital Because the production of automotive components is capital-intensive, it is possible for substantial capital investments in manufacturing equipment and facilities to become "stranded" (where their value is lost, or diminished). This would occur when the capital is rendered useless (or less useful) by some factor that forces a major change in vehicle design, plant operations, or manufacturer's product mix, such as a shift in consumer demand for certain vehicle types. It can also be caused by new standards that phase-in at a rate too rapid to accommodate planned replacement or redisposition of existing capital to other activities. The lost value of capital equipment is then amortized in some way over production of the new technology components. It is difficult to quantify accurately any capital stranding associated with new technology phase-ins under the final standards because of the iterative dynamic involved - that is, the new technology phase-in rate strongly affects the potential for additional cost due to stranded capital, but that additional cost in turn affects the degree and rate of phase-in for the same or other individual competing technologies. In addition, such an analysis is very company-, factory-, and manufacturing process-specific, particularly in regard to finding alternative uses for equipment and facilities. Nevertheless, in order to account for the possibility of stranded capital costs, the agencies asked FEV to perform an analysis, using conservative assumptions, of the potential stranded capital costs associated with rapid phase- in of technologies due to new standards, using data from FEV's primary teardown-based cost analyses.22 Since the direct manufacturing costs developed by FEV assumed a 10 year production life (i.e., capital costs amortized over 10 years) the agencies applied the FEV derived stranded capital costs whenever technologies were replaced prior to being utilized for the full 10 years. The other option would have been to assume a 5 year product life (i.e., capital costs amortized over 5 years), which would have increased the direct manufacturing costs. It seems only reasonable to account for stranded capital costs in the instances where the fleet modeling performed by the agencies replaced technologies before the capital costs were fully amortized. The agencies did not derive or apply stranded capital costs to all technologies only the ones analyzed by FEV. While there is uncertainty about the possible stranded capital costs (i.e., understated or overstated), their impact would not call into question the overall results of our cost analysis or otherwise affect the stringency of the standards, since costs of stranded capital are a relatively minor component of the total estimated costs of the rules. The assumptions made in FEV's stranded capital analysis with potential for major impacts on results are: • All manufacturing equipment was bought brand new when the old technology started production (no carryover of equipment used to make the previous components that the old technology itself replaced). 10-year normal production runs: Manufacturing equipment used to make old technology components is straight-line depreciated over a 10-year life. • Factory managers do not optimize capital equipment phase-outs (that is, they are assumed to routinely repair and replace equipment without regard to whether or not it will soon be scrapped due to adoption of new vehicle technology). 3-21 ------- Technologies Considered in the Agencies' Analysis • Estimated stranded capital is amortized over 5 years of annual production at 450,000 units (of the new technology components). This annual production is identical to that assumed in FEV's primary teardown-based cost analyses. The 5- year recovery period is chosen to help ensure a conservative analysis; the actual recovery would of course vary greatly with market conditions. FEV assembled a team of manufacturing experts to perform the analysis, using a methodology with the following key steps for each vehicle technology scenario: 1) Identify all of the old technology components that are no longer used or that are modified in the new technology vehicles (from the comparison bills of materials developed in the primary teardown-based analyses). 2) For each of these components identify the manufacturing equipment and tooling needed to make it. 3) Estimate the new-purchase \$ value of each item identified in step 2. 4) Assign an "Investment Category" to each equipment item identified in step 2, based on an assessment by FEV's experts of recoverable value: • Flexible: Equipment can be used to manufacture new technology or other parts (0% stranded) • Re-Useable: Equipment can be used in alternative industries, sold at 50% of its remaining value (50% stranded) • Semi-Dedicated: Estimate that 50% of equipment is flexible (50% stranded) • Dedicated: Custom manufacturing equipment (100% stranded) 5) Assign an "Investment Category" to each tooling item identified in step 2, based on an assessment by FEV's experts of recoverable value: • Flexible: Can be used for manufacturing new technology parts (0% stranded) • Perishable: Frequent replacement of tooling (0% stranded) Semi-Dedicated Tooling: Estimate that 50% of tooling is dedicated (50% stranded) • Dedicated: Commodity-specific (100% stranded) 6) Multiply the % stranding values from steps 4 and 5 by the \$ values from step 3. 7) Multiply the results in step 6 by 70%, 50%, and 20% for 3-, 5-, and 8-year stranding scenarios, respectively. That is, an old technology, for which production is truncated prematurely after only 8 years, will experience the stranding of 20% (the last 2 years of its 10-year normal production run) of its associated remaining capital value. 8) Sum the results in step 7 to obtain overall stranded capital costs. 3-22 ------- Technologies Considered in the Agencies' Analysis 9) Divide the results in step 8 by 2,250,000 (5 years x 450,000 units/year) to obtain \$/vehicle values, applicable to new technology vehicles for the 1st 5 years of their production due to the assumed 5-year recovery period. The stranded capital analysis was performed for three transmission technology scenarios, two engine technology scenarios, and one hybrid technology scenario, as shown in Table 3-3. The methodology used by EPA in applying these results to the technology costs is described in Chapter 3 of EPA's RIA. The methodology used by NHTSA in applying these results to the technology costs is described in NHTSA's RIA section V. Table 3-3 Stranded Capital Analysis Results (2010 dollars /vehicle) Replaced technology 6-speed AT 6-speed AT 6-speed DCT Conventional V6 Conventional V8 Conventional V6 New technology 6-speed DCT 8-speed AT 8-speed DCT DSTGDI 14 DSTGDI V6 Power-split HEV Stranded capital cost per vehicle when replaced technology's production is ended after: 3 years \$56 \$48 \$28 \$57 \$61 \$112 5 years \$39 \$34 \$20 \$40 \$43 \$80 8 years \$16 \$14 \$8 \$16 \$17 \$32 DSTGDI=Downsized, turbocharged engine with stoichiometric gasoline direct injection. 3.2.3 Cost reduction through manufacturer learningl For this final rule, consistent with the proposal, we have not changed our estimates of learning and how learning will impact costs going forward from what was employed in the analysis for the MYs 2012-2016 light-duty vehicle rule. However, we have updated our terminology in an effort to clarify that we consider there to be one learning effect—learning by doing—which results in cost reductions occurring with every doubling of production.™ In the past, we have referred to volume-based and time-based learning. Our terms were meant only to denote where on the volume learning curve a certain technology was—"volume-based learning" meant the steep portion of the curve where learning effects are greatest, while "time-based learning" meant the flatter portion of the curve where learning effects are less pronounced. Unfortunately, our terminology led some to believe that we were implementing two completely different types of learning—one based on volume of production and the other based on time in production. Our new terminology—steep portion of the curve and flat portion of curve—is simply meant to make more clear that there is one learning curve and some technologies can be considered to be on the steep portion while others are well into the Note that our approach to accounting for cost reduction through manufacturer learning is unchanged since the proposal. m Note that this new terminology was described in the recent heavy-duty GHG final rule (see 76 FR 57320). The learning approach used in this analysis is entirely consistent with that used and described for the heavy-duty analysis. 3-23 ------- Technologies Considered in the Agencies' Analysis flatter portion of the curve. These two portions of the volume learning curve are shown in Figure 3-1. Volume Learning Curve - Steep & Flat Portions 120% 100% 20% 0% Steep portion of volume learning curve Flat portion of volume learning curve Cumulative Production Figure 3-1 Steep & Flat Portions of the Volume Learning Curve For some of the technologies considered in this analysis, manufacturer learning effects would be expected to play a role in the actual end costs. The "learning curve" or "experience curve" describes the reduction in unit production costs as a function of accumulated production volume. In theory, the cost behavior it describes applies to cumulative production volume measured at the level of an individual manufacturer, although it is often assumed—as both agencies have done in past regulatory analyses—to apply at the industry-wide level, particularly in industries like the light duty vehicle production industry that utilize many common technologies and component supply sources. Both agencies believe there are indeed many factors that cause costs to decrease over time. Research in the costs of manufacturing has consistently shown that, as manufacturers gain experience in production, they are able to apply innovations to simplify machining and assembly operations, use lower cost materials, and reduce the number or complexity of component parts. All of these factors allow manufacturers to lower the per-unit cost of production. We refer to this phenomenon as the manufacturing learning curve. NHTSA and EPA included a detailed description of the learning effect in the MYs 2012-2016 light-duty rule and the more recent heavy-duty rule.23 Most studies of the effect of experience or learning on production costs appear to assume that cost reductions begin only after some initial volume threshold has been reached, but not all of these studies specify this 3-24 ------- Technologies Considered in the Agencies' Analysis threshold volume. The rate at which costs decline beyond the initial threshold is usually expressed as the percent reduction in average unit cost that results from each successive doubling of cumulative production volume, sometimes referred to as the learning rate. Many estimates of experience curves do not specify a cumulative production volume beyond which cost reductions would no longer occur, instead depending on the asymptotic behavior of the effect for learning rates below 100 percent to establish a floor on costs. In past rulemaking analyses, as noted above, both agencies have used a learning curve algorithm that applied a learning factor of 20 percent for each doubling of production volume. NHTSA has used this approach in analyses supporting recent CAFE rules. In its analyses, EPA has simplified the approach by using an "every two years" based learning progression rather than a pure production volume progression (i.e., after two years of production it was assumed that production volumes would have doubled and, therefore, costs would be reduced by 20 percent).11 In the MYs 2012-2016 light-duty rule and the heavy-duty GHG final rule, the agencies employed an additional learning algorithm to reflect the volume-based learning cost reductions that occur further along on the learning curve. This additional learning algorithm was termed "time-based" learning in the MYs 2012-2016 rule simply as a means of distinguishing this algorithm from the volume-based algorithm mentioned above, although both of the algorithms reflect the volume-based learning curve supported in the literature. As described above, we are now referring to this learning algorithm as the "flat portion" of the learning curve. This way, we maintain the clarity that all learning is, in fact, volume-based learning, and that the level of cost reductions depend only on where on the learning curve a technology's learning progression is. We distinguish the flat portion of the curve from the steep portion of the curve to indicate the level of learning taking place in the years following implementation of the technology (see Figure 3-1). The agencies have applied learning effects on the steep portion of the learning curve for those technologies considered to be newer technologies likely to experience rapid cost reductions through manufacturer learning, and learning effects on the flat portion learning curve for those technologies considered to be more mature technologies likely to experience only minor cost reductions through manufacturer learning. As noted above, the steep portion learning algorithm results in 20 n To clarify, EPA has simplified the steep portion of the volume learning curve by assuming that production volumes of a given technology will have doubled within two-years time. This has been done largely to allow for a presentation of estimated costs during the years of implementation, without the need to conduct a feedback loop that ensures that production volumes have indeed doubled. If we were to attempt such a feedback loop, we would need to estimate first year costs, feed those into OMEGA, review the resultant technology penetration rate and volume increase, calculate the learned costs, feed those into OMEGA (since lower costs would result in higher penetration rates, review the resultant technology penetration rate and volume increase, etc., until an equilibrium was reached. To do this for all of the technologies considered in our analysis is simply not feasible. Instead, we have estimated the effects of learning on costs, fed those costs into OMEGA, and reviewed the resultant penetration rates. The assumption that volumes have doubled after two years is based solely on the assumption that year two sales are of equal or greater number than year one sales and, therefore, have resulted in a doubling of production. This could be done on a daily basis, a monthly basis, or, as we have done, a yearly basis. 3-25 ------- Technologies Considered in the Agencies' Analysis percent lower costs after two full years of implementation {i.e.., the MY 2016 costs would be 20 percent lower than the MYs 2014 and 2015 costs). Once two steep portion learning steps have occurred, flat portion learning at 3 percent per year becomes effective for 5 years. Beyond 5 years of learning at 3 percent per year, 5 years of learning at 2 percent per year, then 5 at 1 percent per year become effective. Learning effects are applied to most but not all technologies because some of the expected technologies are already used rather widely in the industry and we therefore assume that learning impacts have already occurred. The steep portion learning algorithm was applied for only a handful of technologies that are considered to be new or emerging technologies. Most technologies have been considered to be more established given their current use in the fleet and, hence, the lower flat portion learning algorithm has been applied. The learning algorithms applied to each technology and the applicable timeframes are summarized in Table 3-4. Table 3-4 Learning Effect Algorithms Applied to Technologies Used in this Analysis Technology Engine modifications to accommodate low friction lubes Engine friction reduction - level 1 & 2 Lower rolling resistance tires - level 1 Low drag brakes Secondary axle disconnect Electric/Plug-in vehicle battery charger installation labor Variable valve timing Variable valve lift Cylinder deactivation Stoichiometric gasoline direct injection Aggressive shift logic - level 1 & 2 Early torque converter lockup 5/6/7/8 speed auto transmission 6/8 speed dual clutch transmission High efficiency gearbox Improved accessories - level 1 & 2 Electronic/electro-hydraulic power steering Aero improvements - level 1 & 2 Conversion to DOHC without reducing # of cylinders Air conditioner related hardware Air conditioner alternative refrigerant Cooled EGR Conversion to Atkinson cycle Turbocharging & downsizing Mass reduction Steep learning 2016-2020 Flat learning 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 2012-20205 2021-2025 2012-2025 2012-2025 2012-2025 2012-2025 No learning 2012-2025 2012-2025 2012-2025 2012-2025 2012-2025 3-26 ------- Technologies Considered in the Agencies' Analysis Advanced diesel Hybrid/Electric/Plug-in vehicle non-battery components P2 Hybrid vehicle battery-pack components Electric/Plug-in vehicle battery -pack components Electric/plug-in vehicle battery charger components Stop-start Lower rolling resistance tires - level 2 2012-2016 2012-20253 2012-2025a 2012-2015 2017-2021 2012-2025 2012-2025 2017-2025 2016-2025 2022-2025 Note that the steep learning effects have for EV and PHEV battery packs and charger components have been carried through 5 learning cycles but at a decelerated pace as described in the text. The learning effects discussed here impact the technology costs in that those technology costs for which learning effects are considered applicable are changing throughout the period of implementation and the period following implementation. For example, some of the technology costs considered in this analysis are taken from the MYs 2012-2016 light-duty rule. Many of the costs in the MYs 2012-2016 light-duty rule were considered "applicable" for the 2012 model year. If flat-portion learning were applied to those technologies, the 2013 cost would be 3 percent lower than the 2012 cost, and the 2014 model year cost 3 percent lower than the 2013 cost, etc. As a result, the MYs 2017-2025 costs for a given technology used in this analysis reflect those years of flat learning and would not be identical to the 2012 model year cost for that same technology presented in the MYs 2012-2016 light-duty rule. Because of the nature of battery pack development (i.e., we are arguably still in the research phase for the types of batteries considered in this final rule, and cost reduction through manufacturer-based learning has only just begun), the agencies have carried the learning curve through five steep learning steps although at a somewhat slower pace than every two years. This has been done in an effort to maintain the shape of a traditional learning curve. This curve was developed by using the ANL BatPaC model costs as direct manufacturing costs applicable in the 2025 MY. We have then unlearned those costs back to 2012 using the curve shown in Figure 3-2. This is the same curve used in the 2010 TAR (see 2010 TAR at page B-22). This allows the agencies to estimate costs in MYs 2017 through 2025, as well as those costs in each year back to MY 2012, if desired. As noted, this learning curve consists of 5 full learning steps on the steep portion of the learning curve, each of which results in costs being reduced 20 percent relative to the prior step. These learning steps are shown occurring every two years beginning in 2012 until 2020, at which time a 5 year gap is imposed until 2025 when the fifth steep learning step occurs. Beyond 2025, learning on the flat portion of the curve begins at 3 percent per year cost reductions. The smooth line shows a logarithmic curve fit applied to the learning curve as the agencies' cost model would apply learning. 3-27 ------- Technologies Considered in the Agencies' Analysis 4.000 3.500 0.000 2012 2016 •Cost Model 2020 2024 •Log. (Cost Model) 2028 Figure 3-2 Learning Curve used for EV & PHEV Battery-Packs and In-Home Charger Costs Note that the effects of learning on individual technology costs can be seen in the cost tables presented in section 3.3, below. For each technology, we show direct manufacturing costs for the years 2017 through 2025. The changes shown in the direct manufacturing costs from year-to-year reflect the cost changes due to learning effects. 3.2.4 Costs Updated to 2010 Dollars0 This change is simply to update any costs presented in earlier analyses to 2010 dollars using the GDP price deflator as reported by the Bureau of Economic Analysis on February 9, 2012. The factors used to update costs from 2007, 2008 and 2009 dollars to 2010 dollars are shown below. Price Index for Gross Domestic Product Factor applied to convert to 2010 dollars 2007 106.2 1.04 2008 108.6 1.02 2009 109.7 1.01 2010 111.0 1.00 Source: Bureau of Economic Analysis, Table 1.1.4. Price Indexes for Gross Domestic Product, downloaded 2/9/2012, last revised 1/27/2012. 0 Note that costs in the proposal were in terms of 2009 dollars. 3-28 ------- Technologies Considered in the Agencies' Analysis 3.3 How did the agencies determine effectiveness of each of these technologies? The agencies determined the effectiveness of each individual technology with a process similar to the one used for the 2012-2016 light duty vehicle GHG and CAFE standards. The individual effectiveness of several technologies discussed in this rule that were present in the earlier rule were left largely unchanged while others were updated. EPA and NHTSA reviewed recent confidential manufacturer estimates of technology effectiveness and found them to be generally consistent with our estimates. Additionally, EPA used vehicle simulation modeling to gain further insight on existing and new technologies for this rulemaking. EPA conducted a vehicle simulation project (described in 3.3.1) that included a majority of the technologies, the results of which: • informed existing individual technology effectiveness values, • provided data for newly introduced technologies, and • most importantly, provided an interactive data source with which to update and calibrate the new LP model The lumped parameter model then served as the primary tool in evaluating the individual technology effectiveness estimates the combined effectiveness of groups of technologies (or packages) and synergy factors, as described in 3.3.2. The effectiveness values, in conjunction with costs, were then applied to vehicles across the fleet for use in the Agencies' respective compliance models. For the final rule, NHTSA conducted a vehicle simulation project with Argonne National Laboratory (ANL), as described in NHTSA's FRIA that performed additional analyses on mild hybrid technologies and advanced transmissions to help NHTSA develop effectiveness values better tailored for the CAFE model's incremental structure. The effectiveness values that were developed by ANL for the mild hybrid vehicles were applied by both agencies for the final rule. Additionally, NHTSA updated the effectiveness values of advanced transmissions coupled with naturally-aspirated engines based on ANL's simulation work for the final rule. 3.3.1 Vehicle simulation modeling 3.3.1.1 Background For regulatory purposes, the fuel economy of any given vehicle is determined by placing the vehicle on a chassis dynamometer (akin to a large treadmill that puts the vehicle's wheels in contact with one or more rollers, rather than with a belt stretched between rollers) in a controlled environment, driving the vehicle over a specific driving cycle (in which driving speed is specified for each second of operation), measuring the amount of carbon dioxide emitted from the vehicle's tailpipe, and calculating fuel consumption based on the density and carbon content of the fuel. One means of determining the effectiveness of a given technology as applied to a given vehicle model would be to measure the vehicle's fuel economy on a chassis dynamometer, install the new technology, and then re-measure the vehicle's fuel economy. However, most technologies cannot simply be "swapped out," and even for those that can, 3-29 ------- Technologies Considered in the Agencies' Analysis simply doing so without additional engineering work may change other vehicle characteristics (e.g., ride, handling, performance, etc.), producing an "apples to oranges" comparison. Some technologies can also be more narrowly characterized through bench or engine dynamometer (i.e., in which the engine drives a generator that is, in turn, used to apply a controlled load to the engine) testing. For example, engine dynamometer testing could be used to evaluate the brake-specific fuel consumption (e.g., grams per kilowatt-hour) of a given engine before and after replacing the engine oil with a less viscous oil. However, such testing does not provide a direct measure of overall vehicle fuel economy or changes in overall vehicle fuel economy. For a vehicle that does not yet exist, as in the agencies' analyses of CAFE and GHG standards applicable to future model years, even physical testing can provide only an estimate of the vehicle's eventual fuel economy. Among the alternatives to physical testing, automotive engineers involved in vehicle design make use of computer-based analysis tools, including a powerful class of tools commonly referred to as "full vehicle simulation." Given highly detailed inputs regarding vehicle engineering characteristics, full vehicle simulation provides a means of estimating vehicle fuel consumption over a given drive cycle, based on the explicit representation of the physical laws governing vehicle propulsion and dynamics. Some vehicle simulation tools also incorporate combustion simulation tools that represent the combustion cycle in terms of governing physical and chemical processes. Although these tools are computationally intensive and required a great deal of input data, they provide engineers involved in vehicle development and design with an alternative that can be considerably faster and less expensive than physical experimentation and testing. Properly executed, methods such as physical testing and full vehicle simulation can provide reasonably (though not absolutely) certain estimates of the vehicle fuel economy of specific vehicles to be produced in the future. However, when analyzing potential CAFE and GHG standards, the agencies are not actually designing specific vehicles. In this rulemaking analysis, the agencies have considered the implications of new standards that will apply to the average performance of manufacturers' entire production lines. For this type of analysis, precision in the estimation of the fuel economy of individual vehicle models is not essential; although it is important that the agency avoid systematic upward or downward bias, uncertainty at the level of individual models is mitigated by the fact that compliance with CAFE and GHG standards is based on average fleet performance. DOT's CAFE model and EPA's OMEGA are not full vehicle simulation models. Both models use higher-level estimates of the efficacy of different technologies or technology packages. Both models apply methods to avoid potential double-counting of efficacy addressing specific energy loss mechanisms (e.g., pumping losses), and for this FRM, consistent with the proposal, both agencies applied estimates using EPA's lumped parameter model, which was updated using results of full vehicle simulation performed by Ricardo, PLC. Although full vehicle simulation could, in principle, be fully integrated into the agencies' model-by-model analyses of the entire fleet to be projected to be produced in future model years, this level of integration would be infeasible considering the size and complexity of the fleet. Also, considering the forward-looking nature of the agencies' analyses, and the amount of information required to perform full vehicle simulation, this level of integration 3-30 ------- Technologies Considered in the Agencies' Analysis would involve misleadingly precise estimates of fuel consumption and CC>2 emissions. Still, while the agencies have used results of full vehicle simulation to inform the development of model inputs for performing fleet-level analysis, information from other sources (e.g., vehicle testing) could be considered when developing such model inputs. 3.3.1.2 2011 Ricardo Simulation Study For this rule EPA built upon its 2008 vehicle simulation project24 used to support the 2012-2016 light duty vehicle GHG and CAFE standards. As in the initial project, the technical work was conducted by the global engineering consulting firm, Ricardo, Inc. (under subcontract to SRA Corporation), using its MSC.EASY5 dynamic vehicle simulation model. This section is intended to supplement the main report which has been recently published and peer-reviewed1. While this project represents a new round of full-scale vehicle simulation of advanced technologies, the scope has also been expanded in several ways to broaden the range of vehicle classes and technologies considered, consistent with a longer-term outlook through model years 2017-2025. The expanded scope also includes a new analytical tool (complex systems analysis tool) to assist in interpolating the response surface modeling (RSM) data and visualizing technology effectiveness. This tool was especially useful in isolating effectiveness trends during development of the updated Lumped Parameter model. The agencies try to use publicly available information as the basis for technical assessments whenever possible. Because these standards extend to MY 2025, and include some technologies that are not currently in production and for which there is limited information available in the literature, some of the technology inputs used to estimate effectiveness are based on confidential business information. This includes the inputs related to the technologies listed below which were based on confidential business information belonging to Ricardo, Inc, and their expert judgment that contributed to projecting how these technologies might improve in the future. The agencies have also considered information which is in the public domain, in particular for turbo-charged, downsized GDI engines as discussed in Section 3.4.1.8, as well as confidential information on engine and transmission technologies from automotive suppliers which directionally was in line with the information considered by Ricardo. In the draft TSD, the agencies encouraged commenters to submit technical information, preferably that may be released publicly, related to these technologies, particularly on their effectiveness and ability to be implemented in a way that maintains utility. The agencies sought comment and data on the following technologies individually or in combination: advanced turbocharged and downsized, atkinson, and advanced diesel (e.g. projected BSFC maps) engines, hybrid powertrain control strategies, optimized transmission shift control strategies, and transmission efficiency improvement. Few comments were received specific to these technologies, although the Alliance emphasized that the agencies should examine the progress in the development of powertrain improvements as part of the mid-term evaluation and determine if researchers are making the kind of breakthroughs anticipated by the agencies for technologies like high-efficiency transmissions. Additionally, Volkswagen commented that while high BMEP (27-31) bar engines with cooled EGR are currently the subject of research, Volkswagen believed that there are significant obstacles, such as thermal and mechanical loads and their impacts on costs and durability, low-end torque performance and part-load efficiency, which need to be overcome before these engines represent a viable option for improving fuel economy while maintaining customer 3-31 ------- Technologies Considered in the Agencies' Analysis satisfaction. The agencies recognize Volkswagen's comments, but note that the analysis for this final rule considered only high BMEP engines up to 27 bar, and will be monitoring the progress of this technology carefully and consider it at the mid-term evaluation. Moreover, since this technology does not reach significant levels in our modeling analyses of the final standards until after MY 2021, the agencies will evaluate industry experience with this technology at the mid-term evaluation and can adjust assumptions as appropriate. Below is a summary of the significant content changes from the 2008 simulation project to the 2011 simulation project that supports the final rule, consistent with the proposal. 3.3.1.2.1 More Vehicle Classes Two additional vehicle classes were considered, for a total of seven classes: a small car (subcompact) and a medium/heavy duty truck class. The inclusion of the small car class increased the fidelity of the results by capturing engineering differences unique to the smallest vehicles in the market. The inclusion of the medium/heavy duty truck was meant primarily to support EPA's analysis for the Heavy Duty GHG Rule25. It is worth noting that these vehicle classes are for simulation purposes only and are not be confused with regulatory classes, OMEGA classes, or NHTSA's technology subclasses for CAFE modeling. 3.3.1.2.2 More engine and vehicle technologies The original 2008 project modeled several engine and transmission technologies that were expected to become commercially available within the 2012-2016 time frame. These technologies included advanced valvetrain technologies (such as variable valve timing and lift, cylinder deactivation), turbocharged and downsized engines, as well as 6 speed automatic transmissions, CVTsp and dual-clutch transmissions. The current project built on top of this effort with the inclusion of several new engine and vehicle technologies. Highlighted examples included: • Advanced, highly downsized, high BMEPq turbocharged engines • High efficiency transmissions with 8 speeds and optimized shift strategies to maximize vehicle system efficiency • Atkinson-cycle engines for hybrids • Stop-start (or idle-off) technology A discussion of these technologies is included Section 3.3.1.2, and also in the 2011 vehicle simulation report1. p Continuously variable transmissions q BMEP refers to brake mean effective pressure, a common engineering metric which describes the specific torque of an engine, as a way of comparing engines of different sizes. It is usually expressed in units of bar, or kPa, Current naturally aspirated production engines typically average 10-12 bar BMEP, while modern turbocharged engines are now exceeding 20 bar BMEP with regularity. Simply put, a 20 bar BMEP turbocharged engine will provide twice the torque of an equivalently-sized engine that achieves 10 bar BMEP. 3-32 ------- Technologies Considered in the Agencies' Analysis 3.3.1.2.3 Includes hybrid architectures For the first time, this new work includes modeling of hybrid architectures for all vehicle classes. Two main classes of hybrids were considered: • Input powersplit hybrids. Examples of input powersplits in the market today include the Ford Fusion HEV and the Toyota Prius. • P2 hybrids. An example of the P2 hybrid is the Hyundai Sonata Hybrid. While input powersplit hybrids remain a very likely hybrid architecture choice for some manufacturers, the agencies focused solely on P2 hybrids compared to powersplit hybrids due to their apparent cost-effectiveness advantage in future years. Ricardo proprietary methodology was used to develop control strategies for each architecture, the details of which can be found in section 6.8 of the 2011 project report1. 3.3.1.2.4 Complex systems tool for data analysis In the original 2008 project, EPA staff selected unique technology packages, based on engineering judgment, to cover a representative subset of possible vehicle options ending in MY 2016. The expanded project time horizon (through MY 2025) and increased complexity of potential vehicle technology interactions (including hybrids) made package selection much more difficult. To account for unforeseen results and trends which might exist, EPA and Ricardo adopted a complex systems approach, which is a rigorous computational strategy designed to mathematically account for multiple input variables and determine the significance of each (the complex systems approach is described in further detail in the 2011 Ricardo report). As a comparison, in the 2008 study, twenty-six unique technology packages spanning five vehicle classes were selected by EPA staff and then modeled. For this project a set of core technology packages were chosen for each vehicle class, constituting a total of 107 unique vehicle packages ("nominal runs"), which are shown as Table 3-5 and Table 3-6 in 3.3.1.2.8. A neural network Complex Systems approach to design of experiments (DOE) was then applied to generate a set of response surface models (RSM), in which several input parameters were varied independently over a specified range to identify the complex relationship between these inputs and the vehicle performance. Using these methods, the vehicle simulation was run for a set of discrete input variables chosen based on a full factorial analysis, using a computationally efficient algorithm to select each input variable within the design space, allowing for subsequent statistical regression of the output variables. This approach resulted in an average of approximately two thousand independent simulation runs for each of the 100+ vehicle packages, the outputs of which were interpolated in the data analysis tool developed for this modeling activity. For each of these nominal and DOE runs Ricardo provided detailed 10-hz output data csv files for reviewr. r Stakeholders wishing to obtain this data may contact EPA to arrange for transfer of the data. Due to the considerable size of the files (2 terabytes), stakeholders must supply their own storage media. 3-33 ------- Technologies Considered in the Agencies' Analysis An interactive Complex Systems analysis and visualization tool was developed to interpret the vast arrays of RSM data generated as part of the project. It was created to sample a selected portion of the design space populated using the DOE approach described above, and then interpret the RSM data set in a form that could be used to calibrate the lumped parameter model (reference the equivalent-performance results in Section 3.3.1.2.18). For more detail on the use of the RSM tool, refer to the 2011 Ricardo report1. 3.3.1.2.5 Process The core technical work, completed in February 2011, consisted of the following steps: Definition of project scope Selection of vehicle classes and baseline vehicle characteristics Selection of vehicle architectures and individual technologies Selection of swept variables for use in the RSM matrix Selection of vehicle performance metrics Review and revision of the input assumptions and modeling process Build and run the baseline EASY5 vehicle models Review of baseline runs and checking for errors Build and run the nominal technology package EASY5 vehicle models Review results and debug Run complete DOE matrix for each technology package Incorporation of DOE results into RSM tool 3.3.1.2.6 Definition of project scope At project initiation, an advisory committee was formed and led by EPA to help guide the analysis. The advisory committee consisted of technical experts from CARB and The ICCT, the latter of which co-founded the project. A complete list of advisory committee members is found in the vehicle simulation project report1. The committee agreed upon the underlying ground rules, reviewed modeling assumptions and identified the desired vehicle architectures and selected technologies for review. The boundaries for the project are highlighted (quoted) below: • A total of seven vehicle classes will be included: small car, standard car, large car, small and large MPVs (multi-purpose vehicles), truck and HD truck • LDV technologies must have the potential to be commercially deployed in the MYs 2020-2025 timeframe • Vehicle sizes (footprint and interior space) for each class will be largely unchanged from MY 2010 to MYs 2020-2025 3-34 ------- Technologies Considered in the Agencies' Analysis • Hybrid vehicles will use an advanced hybrid control strategy, focusing on battery state-of-charge management, but will not compromise vehicle drivability • Ricardo simulation study uses certification gasoline and 40 cetane pump diesel to determine the effectiveness of engine technologies. The certification gasoline typically has an RON of approximately 95 versus approximately 91 for regular grade 87 anti-knocking index gasoline. • It is assumed that MYs 2020-2025 vehicles will meet future California LEV III requirements for criteria pollutants, approximately equivalent to current SULEV II (or EPA Tier 2 Bin 2) emissions levels • Changes in vehicle road loads including mass, aerodynamic drag, and rolling resistance, will not be accounted for in any of the modeled technologies. Instead, changes in vehicle road loads may be addressed through user-specified continuous input variables in the Complex Systems tool. The committee also decided that the following technologies fell outside the scope of the project, either due to project resource limitations, lack of sufficient input data, or a low potential to be commercially deployed in the timeframe considered: • Charge-depleting powertrains (e.g. plug-in hybrids and electric range-extended vehicles) and electric vehicles • Fuel cell-powered vehicles • Non-reciprocating internal combustion engines or external combustion engines • Manual transmissions and single-clutch automated manual transmissions (AMTs) • Kinetic energy recovery systems other than battery systems • Intelligent vehicle-to-vehicle (V2V) and vehicle-to-infrastructure optimization technology • Bottoming cycles (such as organic Rankine cycles) for energy recovery • Vehicle safety systems or structures will not be explicitly modeled for vehicles, as it is beyond the scope of the study The committee also selected a set of swept input variables (vehicle parameters) which were considered most important to vehicle fuel economy and performance (swept variables are continuously variable input values that affect vehicle output efficiency in a smooth function for the response surface model). These variables consisted of engine displacement, final drive ratio, electric drive motor size (for hybrids), as well as road load factors (vehicle mass, aerodynamic drag, and rolling resistance). All of these input variables were 3-35 ------- Technologies Considered in the Agencies' Analysis randomized in each vehicle design of experiment matrix and then incorporated into the post- processing RSM data visualization tool. 3.3.1.2.7 Selection of vehicle classes and baseline vehicle characteristics In order to estimate both technology costs and CO2 reduction estimates, it is necessary to describe the baseline vehicle characteristics as the basis from which comparisons may be drawn. In the MYs 2012-2016 light-duty vehicle rule the vehicle baseline was defined as having a naturally aspirated gasoline engine with a port-fuel injection system, two intake and two exhaust valves and fixed valve timing and lift; the baseline transmission was a conventional 4-speed automatic, with no hybrid systems. These vehicles are referred to throughout this section as the "2008 baselines." For the present study, EPA and Ricardo elected to include a set of "2010 baseline" technology vehicles, which reflect MY 2010 trends in engine and vehicle technology as well as some technologies that are expected to be widespread within a few years. It is important to note that the 2010 baseline vehicles in the Ricardo study do not reflect the technology content of the baseline fleet vehicles used by each agency in their respective compliance modeling. The Ricardo 2010 baseline vehicles are only used in the analysis required to establish effectiveness and synergies in the lumped parameter model. The 2010 baseline vehicles all include an engine with dual overhead camshaft and dual-independent intake/exhaust valve timing, a six-speed automatic transmission, 12-volt idle off (stop-start) functionality and an alternator with partial energy regeneration capability. There is no change in the engine displacement or vehicle road load coefficients between the 2008 baseline and the 2010 baseline vehicles. For a table showing the 2010 baseline vehicle characteristics refer to Appendix 3 of the 2011 Ricardo report1. In the Ricardo study, seven vehicle classes were selected for the analysis, in order to more fully represent the broad groupings of a wide variety of products offered in the US passenger car and light-duty truck market. The seven vehicle categories chosen were as follows: • Small car: a subcompact car typically powered by a small 4 cylinder engine. • Standard car: a midsize car typically powered by a small 6 cylinder engine. • Large car: a large passenger car typically powered by a large 6 cylinder engine. • Small MPV: a small multi-purpose vehicle (MPV) or "crossover" vehicle typically powered by a 4 cylinder engine • Large MPV: a minivan or large MPV or "crossover" unibody constructed vehicle with a large frontal area, typically powered by a 6 cylinder engine, capable of carrying ~ 6 or more passengers. • Large truck (1/2 ton): large sports-utility vehicles and large pickup trucks, typically a ladder-on-frame construction, and typically powered by an 8 cylinder engine. • Class 2b/3 truck (3/4 ton): a large pickup truck (although with a GVW no greater than 8.500 pounds) with a heavier frame intended to provide additional utility (a.k.a. "work" truck), typically powered by a larger 8 cylinder gasoline or diesel engine. 3-36 ------- Technologies Considered in the Agencies' Analysis 3.3.1.2.8 Technology selection Ricardo presented the committee with an array of potential technologies that might become commercially viable and present in the light-duty market by MY 2025. EPA and the Advisory Committee suggested additional other technologies, e.g. Atkinson engines for hybrids, fast engine warm-up strategies, etc, to consider in the selection process. The complete set of potential technologies can be found in Appendix 2 of the 2011 Ricardo report1. After further deliberation within the committee and by Ricardo, a subset of technologies considered most promising (from a technical feasibility and cost effectiveness standpoint) was selected by the committee and Ricardo for inclusion in the project test matrix. The technologies were distributed among four distinct vehicle architectures. These architectures represented unique EASY5 model structures, and are listed below: • 2010 Baseline vehicles: intended to represent physical replicas of existing vehicle models, although some minor additional content was included (as described in Section 3.3.1.2.7) • Conventional stop-start: vehicles for the MYs 2020-2025 timeframe that included advanced engines but did not incorporate an electric drive or braking energy recovery. These vehicles all contained a 12 volt stop-start (or idle-off) capability, along with the following technologies further detailed in the 2011 Ricardo simulation study8: o higher efficiency gearbox (2020 timeframe) o optimized shift strategy (best BSFC) o alternator regeneration (during braking) o high-efficiency alternator o advanced engine warmup technologies o engine friction reduction (+3.5% fuel consumption reduction over 2008 baseline) • P2 hybrid: represent a class of hybrids in which the electric drive motor is coupled via a clutch directly to the transmission input shaft. An existing vehicle in the market which most closely represents this architecture is the 2011 Hyundai Sonata Hybrid except that Ricardo recommended a P2 hybrid with a more efficient and cost effective dual clutch transmission in lieu of an automatic transmission. Additional examples of a P2 hybrid approach are the 2011 Volkswagen Touareg Hybrid, the 2011 Porsche S Hybrid, and the 2012 Infmiti M35 Hybrid. Each of these are examples of "first generation" P2 systems, as compared to for example the powersplit hybrid systems offered by Ford, Toyota and or the IMA systems s The technologies included in all of the conventional stop-start packages were expected to be widespread by years 2017-2025. Some "anytime technologies" such as aerodynamic drag and rolling resistance reduction were excluded from the nominal runs, but were incorporated in the complex systems portion of this project. 3-37 ------- Technologies Considered in the Agencies' Analysis from Honda which are in their second, third or even fourth generation. The agencies are aware of some articles in trade journals, newspapers and other reviews that some first generation P2 hybrid vehicles with automatic transmissions have trade-offs in NVH and drivability - though these reviews do not cover all of the P2 systems available today, and a number of reviews are very positive with respect to NVH and drivability. For this analysis we are projecting that these issues with some first generation P2 systems can be addressed with no hardware cost increase or reduction in efficiency for future generations of P2 systems developed for the 2017-2025 time frame. The agencies sought comment on our assumptions in this regard, and we requested comment on the applicability of DCTs to P2 hybrid applications, including any challenges associated with NVH or drivability. There were no comments submitted. Key technology assumptions included: o Lithium-ion battery o DCT transmission o Electric drive motor which provides, when combined with a less powerful engine, equivalent 0-60 performance to the baseline vehicle. o Engine displacement for the P2 hybrids were assumed to be 20% less than their conventional stop-start equivalents • Input powersplit hybrid: represent a class of hybrids with both an electric drive motor and a separate generator linked to a planetary gearset which effectively controls the overall gear ratio and distribution of tractive and electrical power. Example vehicles in the market include the Toyota Prius and the Ford Fusion hybrid. Key technology assumptions are consistent with those for the P2 hybrid, with the exception of the power split device, which functions as a CVT-type transmission (as is the case in real world examples), and replaces the DCT transmission in the P2 design. As stated previously while this technology was simulated it was not used in this FRM analysis, consistent with the proposal. Some architectures that seemed less appropriate for certain vehicle classes were omitted. For example, in the Ricardo modeling of the medium/heavy duty truck (a Class 3 vehicle with a GVWR >10,000 pounds, and thus not subject to the final standards in this rulemaking), no P2 or input powersplit hybrids were included. Other technologies that did not seem reasonable for some vehicle classes (such as dry-clutch DCTs for Large MPVs and Trucks) were also excluded in the Ricardo simulations. In summary, 4 distinct vehicle architectures (including the baselines as an "architecture"), across 7 vehicle classes, and a number of engine and transmission combinations, represented the complete set of vehicle combinations. The test matrices1 can be ' For each vehicle class, each advanced engine option is combined with each advanced transmission. Baseline runs are not combined with other transmissions. 3-38 ------- Technologies Considered in the Agencies' Analysis found below in Table 3-5 (for 2010 baselines and conventional stop-start vehicles) and Table 3-6 (for hybrids). Table 3-5: Nominal Package Matrix for Non-Hybrids Vehicle Class Small Car Standard Car Small MPV Full Size Car Large MPV LOT LHDT o3 o c* « m a. .°. ~ to E S o o w o 3 ffl ry tff X X X X X X X o ._ V ** ° E — "3 .2 .S> TJ -E Q O C LJ o v> 0 Q. C 5 °? 2 CM U> 1— X Ac o A 3 1- Q Q •c O ** ^ o t: ** '^ OT g X X X X X X X Ivance o i_ 3 1— (/) Q O C ^ CQ +j X X X X X X X d Engii o i^ H < Q Q ' C 0 -^ UJ | X X X X X X X ie o < X Ivance >» o ^ o o Q. 1- OT O u> O X d Iran _o ^ ts o f= o. o oo < X X X X X X smissic >» O •o o o Q. 1- OT O oo O X X X >n ^3 > •o o o Q. 1- OT O oo O X X X Table 3-6: Nominal package matrix for P2 and Input Powersplit hybrids Vehicle Class Small Car Standard Car Small MPV Full Size Car Large MPV LOT LHDT Hybrid Architecture 4-1 g O T O CM Q. CM X X X X X X Q. W Q < 5 X X X X X X 3-39 ------- Technologies Considered in the Agencies' Analysis 3.3.1.2.9 Selection of the swept input variables and their ranges The advisory committee agreed upon a set of continuous input variables to be swept in each vehicle package response surface. These variables consisted of both powertrain characteristics (engine displacement, final drive ratio, and electric machine size for hybrids) and road load parameters (rolling resistance coefficient, aerodynamic drag force, and vehicle mass). They were included in the DOE matrix for each vehicle architecture and powertrain configuration, and also serve as inputs to the complex systems visualization tool. Table 3-7 and Table 3-8 show the swept variables used (and their ranges) for the conventional stop-start and hybrid packages, respectively. The ranges represent a percentage of the default value used in the nominal runs. Table 3-7: Continuous input parameter sweep ranges for conventional stop-start vehicle Parameter Engine Displacement Final Drive Ratio Rolling Resistance Aerodynamic Drag Mass DoE Range (%) 50 125 75 125 70 100 70 100 60 120 Table 3-8: Continuous input parameter sweep ranges for P2 and Powersplit hybrid vehicles Parameter Engine Displacement Final Drive Ratio Rolling Resistance Aerodynamic Drag Mass Electric Machine Size DoE Ra P2 Hybrid 50 150 75 125 70 100 70 100 60 120 50 300 nge (%) Powersplit 50 125 75 125 70 100 70 100 60 120 50 150 The ranges were intended to include both the (unknown) optimal value for each technology case, but also wide enough to capture the range of values as they depart from the optimal value (in engineering parlance this is often referred to as finding the "knee" in the curve). From these variables, a user can determine the sensitivity of each input variable to the vehicle fuel economy and performance. For example, the effect of engine displacement on fuel economy was evaluated for several packages. A more elaborate discussion of engine displacement effects is provided in Section 3.3.1.2.24.2. 3-40 ------- Technologies Considered in the Agencies' Analysis 3.3.1.2.10 Selection of vehicle performance metrics For both effectiveness and cost estimates in these rulemakings, the agencies are assuming that vehicles will maintain utility (performance) comparable to the models in the baseline fleet". It was therefore important to maintain equivalent performance in the vehicle simulation modeling of future vehicle technology. The resulting effectiveness estimates were in the context of equivalent performance, which carried over into the lumped parameter model and into the OMEGA and CAFE model packages. Consistent with the 2008 simulation project, a set of vehicle (acceleration) performance metrics were selected by the advisory committee as a way of measuring "equivalent" vehicle performance. When quantifying vehicle efficiency, it is important that certain other vehicle performance metrics are maintained, such that there are no other competing factors contributing or detracting from the vehicle efficiency. Other vehicle characteristics that could impact or detract from vehicle efficiency (e.g., noise, vibration and harshness (NVH), drivability, durability, etc) were also considered during the generation of model inputs. However, they were not analyzed explicitly, with the expectation that manufacturers would ultimately be able to meet vehicle refinement levels necessary for commercial acceptability of these new technologies. These metrics, shown below in Table 3-9, include time at full load to reach given speeds (0-10 mph, 0-30 mph, etc), maximum grade capability, and distance traveled at a given time (e.g., after 3 seconds). Ultimately, the measure of equivalent performance is up to the reader or user of the Complex Systems tool. For EPA's analysis baseline vehicle 0-30 mph and 0-60 mph acceleration times were used as a benchmark for equivalent performance for the advanced vehicle packages. These estimated acceleration times are included in Table 3-11 through Table 3-18. Detailed results that include all performance metrics including those for baseline vehicles are provided in the full 2011 simulation report1. u The only exception to this is a subset of hybrids explicitly listed as "non-towing" vehicles. For further details and background, reference Section 1.3 of EPA's RIA. 3-41 ------- Technologies Considered in the Agencies' Analysis Table 3-9: Vehicle performance metrics produced by the EASY5 model Launch (WOT) 0-10mph 0-30mph 0-50mph 0-60mph 0-70mph Distance @ 1.3 sec Distance @ 3 sec Speed @ 1.3 sec Speed @ 3 sec Passing (WOT) 30-50mph 50-70mph Gradeability/ torque reserve Max Speed @ 5% grade Max Speed @ 10% grade Max Grade @ 70 mph (non-towing) Max Grade @ 60 mph (towing) 3.3.1.2.11 Review and revision of inputs For any system modeling in which the results extend beyond the bounds of known physical examples (and therefore direct data validation is impossible), it is imperative that the inputs be carefully constructed and thoroughly examined to minimize the potential for uncertainty-related errors. Prior to coding of the models, Ricardo presented the following inputs for review and approval to EPA. For each topic, EPA reviewed the material considering the rationale of Ricardo's technical experts, the appropriateness of the inputs in relation to the assumed time horizon, the required emissions levels, and the known literature in the field today. Listed below are several of the model inputs that were jointly reviewed by Ricardo and EPA: Engine maps o o o o Stoichiometric GDI turbo Lean-burn GDI turbo Cooled EGR turbo Advanced diesel maps Transmission efficiency tables (by gear) including torque converter efficiency Engine warm-up strategy (cold start modifiers) Alternator regeneration strategy Transmission shift optimizer Engine friction reduction level P2 hybrid controls 3-42 ------- Technologies Considered in the Agencies' Analysis • Input powersplit hybrid controls • Hybrid battery assumptions • Hybrid motor/generator efficiency maps EPA technical experts recommended several changes and iterated with Ricardo to establish a consensus set of inputs that were plausible and met the ground rules of the project. Some of these changes resulted in higher efficiencies, while others lowered efficiency. Highlighted below are a few key examples, starting with development of the engine maps: Engine maps carry perhaps the most significance of any of the sets of inputs needed to build vehicle simulation models. They provide the brake specific fuel consumption, or BSFC (typically in g/kWh) for a given engine speed and load. Typically these maps show an optimum speed and load band (or minimum BSFC "island") that is the most efficient condition in which to operate the engine. Ricardo generated engine maps for both the baseline vehicles (through benchmarking data) and proposed future engine maps for the various turbocharged and diesel engines. Figure 3-3 shows an example engine map for a baseline vehicle. It was constructed from EPA's analysis of a baseline vehicle model run output file. The contours represent lines of equivalent brake-specific fuel consumption/ 3.3.1.2.12 Engine Technologies Ricardo developed the engines for the 2012-2025 timeframe in two ways. The first was to take current boosted SI research engines and project these would represent the level of performance which could be achieved by production engines in the 2020-2025 timeframe. The second method took current production Atkinson cycle SI and diesel engines and then included 2020-2025 timeframe technology improvements. Both methods extrapolated current engine design and development trend to the 2020-2025 timeframe. These current trends include engine friction reduction, improved fuel injection systems (e.g., spray guided for the SI, and higher injection pressures for the diesels), more advanced engine controls, and improved engine design for faster engine warm-up. EPA reviewed the engine maps recommended by Ricardo and generally concurred they were appropriate for the study time frame based on EPA's review of maps for current production engines and for research engines described in the literature. v B SFC is measured in units of grams of fuel per kW-hour of energy and is an indicator of engine efficiency. Lower numbers indicate more efficient operating regions. As in this case, an engine typically has an "island' or region of best efficiency, in this case between 2000-3000 RPM and 150-180 Nm of torque. This island becomes much larger with the advent of advanced technologies such as boosting and downsizing, as well as advanced valvetrain technologies. 3-43 ------- Technologies Considered in the Agencies' Analysis 220 200 1BO 160 - 14O 12O - 1OO - SO - 40 1OOO 2OOO 3OOO 4OOO Speed (rpm) 5OOO BOOO Figure 3-3: Example baseline engine BSFC map 3.3.1.2.12.1 Stoichiometric GDI The original Stoichiometric GDI map that Ricardo proposed was based on laboratory data they had published in 2007, showing a peak brake-specific load of just under 20 bar BMEP and a minimum BSFC of approximately 235 g/kWh, obtained using a compression ratio of 10.5:1.26 However, based on input from manufacturers and from other, more recent published data on developmental and research engines, EPA asked Ricardo to raise the load _ 9*7 OQ OQ "20 capability of the engine to approximately 27 bar BMEP. ' ' ' This allowed a greater degree of engine downsizing, which resulted in a downsizing of a 1.5 liter engine to a 0.74 liter engine for the nominal small car and a 5.4 liter to a 1.94 liter engine for the nominal large truck. A compression ratio of 10.5:1 was maintained for improved efficiency. At the same time, EPA asked that Ricardo eliminate the use of high-load enrichment, since water-cooled exhaust manifolds, in some cases integrated into the cylinder head, can be incorporated in next-generation designs to mitigate the need for fuel enrichment in lowering turbine inlet temperatures to 950 degrees C and thus avoid the added costs of high-temperature materials in the turbocharger.31'32 By reducing the need for fuel enrichment fuel consumption is reduced over the more aggressive portions of the drive cycle, and PM emissions control at high load is improved. 3-44 ------- Technologies Considered in the Agencies' Analysis 3.3.1.2.13 Lean-burn GDI Ricardo's initial lean-burn GDI map was based on their single-cylinder research engine data, in which they operated in lean stratified charge mode at all speeds and loads, without due consideration of the potential limitations in lean exhaust NOx aftertreatment systems. To address concerns in this area, EPA examined the boundaries of operation of lean- NOx catalysts, assuming that manufacturers would adopt either LNTs or metal-zeolite urea SCR systems. EPA therefore asked Ricardo to place a constraint on the maximum allowable catalyst space velocity (at high engine power) and exhaust gas temperature entering the catalyst (at high load, low engine speed conditions) to maintain catalyst efficiency at high load and to reduce thermal sintering of PGM under high-temperature, lean operating conditions. More specifically, EPA recommended that engine operation switch away from lean operation (at air/fuel equivalence ratios up to approximately X=1.5) to stoichiometric operation at turbine outlet temperatures above 600C, and at total exhaust flows corresponding to space velocities of 60,000/hour, assuming a catalyst volume of 2.5 times engine displacement. This marginally diminished the engine brake thermal efficiency to stoichiometric GDI levels over this region of the map, but it provided more certainty that the engine would be able to adhere to the emissions levels as assumed in the project ground rules by the Advisory Committee. Figure 3-4 shows the engine speed and load region EPA proposed as suitable for lean stratified operation. 3-45 ------- Technologies Considered in the Agencies' Analysis -700- - 40000 - Temperature NOx Catalyst Space Velocity Transition from lean to A=l operation 0 1000 2000 3000 4000 5000 6000 7000 Engine Speed [rpm] Figure 3-4 Proposed lean/stoichiometric operating threshold for lean-burn GDI engines 3.3.1.2.13.1 Cooled EGR GDI EPA provided technical information from the literature which enabled Ricardo to assume a dual loop (both low pressure and high pressure EGR loops), cooled EGR system in addition to the stoichiometric turbocharged engine. The development of engine maps for this engine configuration was heavily informed by recently published data.30'31'32'33. Cooled EGR allowed the use of "X=l" operation at the same compression ratio with more aggressive spark timing at high load and reduced pumping losses at part load while maintaining acceptable turbocharger inlet temperatures. 3.3.1.2.13.2 Motor/generator and power inverter efficiency maps EPA recommended that Ricardo update the efficiency maps of the motor and generator (referred to as "electric machines" throughout the project), which they had proposed 3-46 ------- Technologies Considered in the Agencies' Analysis based on current best-in-class technology. The baseline motor/generator+inverter efficiency map is taken from a 2007 Camry and shown in Figure 3-5 below. 300 250 200 a> g.150 100 50 92 90 88 86 84 82 80 78 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 Speed (RPM) Figure 3-5: 2007 Camry Hybrid motor-inverter efficiency map (Burress, et al, 200834) EPA requested that Ricardo provide their assessment of where they believed efficiency improvements might be made, based upon trends in research and development for both electric machines and power electronics. Ricardo and EPA generally agreed that these efficiency improvements were likely to be modest, particularly given the competitive pressures on manufacturers to reduce the cost of hybrid components. However, EPA and Ricardo assumed that today's best-in-class efficiency would likely be marginally improved through continuous incremental reductions in parasitic losses. To account for this, EPA and Ricardo agreed to reduce the losses in the motor/generator by 10% (in other words, raising the efficiency of a 90% efficient motor to 91%) and to reduce the losses in the power electronics by 25% (mainly through continued improvements in inverter development and electronic control systems). 3.3.1.2.13.3 Battery Battery packs were assumed to consist of spinel LiMnO2 cathode chemistry, which is consistent with the current state of technology. EPA recommended a maximum usable state of charge of 40% (from 30% charge to 70% charge) be incorporated as an operating window in Ricardo's hybrid control logic. This range may increase in subsequent real world examples as manufacturers gain more field experience with long term battery durability. Additionally there will likely be more advances in battery construction and chemistry by 2025, so EPA considers these assumptions as conservative in view of the long term research currently underway in many battery research companies. 3-47 ------- Technologies Considered in the Agencies' Analysis 3.3.1.2.14 Additional Technologies Modeled by Ricardo for 2011 Report The previous section discusses in detail those areas of the Ricardo simulation inputs which EPA provided recommendations to Ricardo on and which Ricardo agreed and made modifications to their initial suggestions. EPA did review modeling inputs for many other technologies modeled by Ricardo, but for which we generally agreed with the reasonableness of Ricardo's approach and did not request any changes. This section summarizes at a high level some of the additional technologies considered by Ricardo. Additional detail on these technologies is contained in the 2011 Ricardo final report. Diesel engines - Ricardo started with existing production engines and identified technology advances that would lead to further advances in fuel consumption. These included many of the same technologies considered for advanced gasoline engines, such as engine friction reduction, improved fuel injection systems with higher injection pressures and more advanced controls, and better engine design to improve engine warm-up rate. Transmission Technologies - Taking a systems approach in the vehicle simulation modeling, Ricardo also introduced additional transmission and driveline oriented technologies that may be pathways to increased efficiency. Some of these key technological enablers include: shift optimization schedules, advanced clutches, torque converter design and lockup schedules. Automatic and Dual Clutch Transmissions - For the study timeframe, Ricardo assumed that eight-speed automatic transmissions will be in common use, as this supports more efficient operation, except for small cars, with energy losses expected to be about 20- 33% lower than in current automatic transmissions. Energy losses in both wet clutch and dry clutch DCTs are expected to be 40-50% lower than in current automatic transmissions. Transmission Shift Optimization - This advanced transmission shift optimization strategy tries to keep the engine operating near its most efficient point for a given power demand in effort to emulate a CVT. To protect against operating conditions out of normal range, several key parameters were identified, such as maximum engine speed, minimum lugging speed, and minimum delay between shifts. During development of this strategy, Ricardo estimated that fuel economy benefits of up to 5% can be obtained when compared to typical MY 2010 shift maps. Torque Converter Technology - Ricardo utilized a lockup clutch model with a multi- damper system to provide earlier torque converter clutch engagement. The advanced automatic transmission applications allow torque converter lockup in any gear except first gear, up to sixth for the Small Car or eighth for the other LDV classes. Shifting Clutch Technology - Shift clutch technology improves the thermal capacity of the shifting clutch to reduce plate count and lower clutch losses during shifting. Reducing the number of plates for the shifting process and reducing the hydraulic cooling requirements will increase the overall transmission efficiency for similar drivability characteristics. Dry Sump Technology - A dry sump lubrication system provides benefits by keeping the rotating members out of oil, which reduces losses due to windage and churning. This approach will provide a GHG emissions benefit across all vehicle classes, with the best benefits at higher speed. 3-48 ------- Technologies Considered in the Agencies' Analysis 3.3.1.2.15 Baseline models built and run Once all of the inputs were established, Ricardo built the baseline models: For these new (2010) baseline models Ricardo added a group of minor technologies, most of which already exist today in the market. The technologies included 12V stop-start, 6-speed automatic transmission, a high efficiency (70% efficient) alternator, and a strategy - "alternator regen" - that charges the 12V battery more aggressively by increasing the alternator field upon vehicle deceleration . In the 2008 study Ricardo validated their baseline models with 2008 MY certification data. Ricardo's 2010 baseline model results provided effectiveness data for EPA to calibrate the lumped parameter model for some of the newly applied technologies. These technologies included alternator regeneration, high efficiency alternator, and stop-start. For all model runs - the baselines and each of the advanced package nominal runs - EPA reviewed an extensive set of detailed intermediate output data for each model run. The parameters that were reviewed are shown in Table 3-10. Table 3-10: Vehicle simulation output data reviewed Ricardo outputs vehicle speed throttle position engine torque engine power transmission input shaft torque wheel torque transmission gear torque converter slip ratio current engine BSFC accessory power engine speed road load N/V electric power of motor generator mechanical power of motor generator motor generator speed motor generator torque motor generator current motor generator voltage power flow through battery battery state of charge battery voltage regenerative braking power vehicle foundation braking power driver braking force fuel mass flow rate transmission mechanical loss power idle off status EPA-calculated outputs engine operating point distribution engine load (BMEP) total accessory energy round-trip battery loop losses torque converter lockup time total road load total engine brake thermal energy EPA-calculated metrics cycle-average BSFC average brake thermal efficiency average engine power average engine speed average engine torque #of idle-off events % of engine time off average accessory power time in each gear average gear efficiency average torque converter efficiency battery state-of-charge statistics battery efficiency % of vehicle braking energy recovered average motor efficiency average generator efficiency average motor and generator operating speeds average motor and generator operating torque total vehicle tractive energy 3-49 ------- Technologies Considered in the Agencies' Analysis From this data, a set of summary statistics was generated to compare each baseline and nominal package run as a quality check. This information was used as the starting point in the dialogue between EPA and Ricardo to identify technical issues with the models. An example summary table (or "snapshot") for the 2010 Standard Car baseline is provided in Figure 3-6. 3-50 ------- Technologies Considered in the Agencies' Analysis Vehicle C02 Emissions (g/mi) Fuel Economy (mpg) 2007 Base Vehicle CO2 (g/mi) % CO2 Reduct on Engine Avg Brake Thermal Efficiency Cycle Avg BSFC (g/kWh) Avq Engine Power (HP) Avg Engine Speed (RPM) Avg Load (BMEP-barJ Avg Torque (Nm) Total Fuel (g) Idle Off Events % Time Off Accessory Loss Avg accessory power (W) Avg BSFC temp mult (20F) Avg BSFC temp mult (75F) Transmission Time in gear 1 Time in gea 2 Time in gea 3 Time in gea 4 Time in gea 5 Time in gea 6 Time in gea 7 Time in gea 8 Avg. n (gear) Avg. n (TC) Avg. n (driveline) lattery SOCAvg Std Deviat on Max SOC MinSOC Max SOCSwng Battery Efficiency (%) Average Voltage (V) Std Dev Voltage (V) Battery Energy Change (kWh) % ofbraking energy recovered %batt charge via brake recov %batt charge via eng ne MG1 Test-Avg Motor Power (hp) Avg Motor Eff Avg Generator Eff Avg Torque-Motor (N-m) Avg Torque-Generator (N-m) Avg RPM-Motor Avg RPM-Generator Mech Energy-Motor (kWh) Mech Energy-Gen (kWh) MG2 Avg Motor Power (hp) Avg Motor Eff Avg Generator Eff Avg Torque-Motor (N-m) Avg Torque-Generator (N-m) Avg RPM-Motor Avg RPM-Generator Mech Energy-Motor (kWh) Mech Energy-Gen (kWh) Round-trip MG efficiency' Buck/Boost Converter Avg Discharge Efl Avg Charging Eff Avg Bus Voltage (V) .HV (fuel) SG(fiel) Speciic C02 Vehicle Energy Audit (kWh) Total fuel energy Total indicated energy Engine pumping energy Engine friction energy Engine brakinq enerqy Total accessory energy Net brake thermal energy Torque converter losses Transmission losses Battery loop losses PE losses Losses to MG devices Total driveline losses Vehicle tractive energy Total road load energy Foundation braking energy Alternator regen decel energy Total reqd. braking energy FTP 303.8 29.9 337.8 10.1% FTP 21.7% 376 7.0 1993 2.21 42.1 1026.4 20 18.0% 0.0% 8.2 1.32 1.20 FTP 30% 9% 16% 27% 9% 9% 0% 0% 87.4% 88.9% 77.7% FTP n/a n/a n/a n/a n/a n/a n/a n/a 0.00 0.0% #DPV/0! #DIV/0! FTP n/a n/a n/a n/a n/a n/a n/a 0.00 0.00 FTP n/a n/a n/a n/a n/a n/a 0.00 0.00 SDIV/0! FTP n/a n/a n/a 44 0.739 9087 FTP 12.54 4.48 0.69 0.86 0.20 0.00 2.73 0.30 0.31 0.00 0.00 0.00 0.61 2.12 1.29 0.50 0.32 0.82 Hwy 209.0 43.5 217.5 3.9% Hwy 27.8% 295 14.1 1833 3.27 62.5 657.8 1 0.5% 0.5% 198.0 n/a n/a Hwy 2% 1% 2% 6% 35% 54% 0% 0% 88.0% 97.8% 86% Hwy n/a n/a n/a n/a n/a n/a n/a n/a 0.00 0.0% #DIV/0! #DIV/0! Hwy n/a n/a n/a n/a n/a n/a n/a 0.00 0.00 Hwy n/a n/a n/a n/a n/a n/a 0.00 0.00 #DIV/0! Hwy n/a n/a n/a kJ/g g/gal Hwy 8.04 3.38 0.57 0.48 0.03 0.04 2.23 0.05 0.26 0.00 0.00 0.00 0.31 1.92 1.76 0.11 0.06 0.16 Combined 261.2 34.8 283.7 7.9% Combined 23.8% 344 10.2 1921 2.69 51.3 860.5 n/a 10.1% 0.3% 93.6 n/a n/a Combined 17% 5% 10% 18% 21% 29% 0% 0% 87.7% 92.9% 81.5% Combined n/a n/a n/a n/a n/a n/a n/a n/a 0.00 0.0% #DIV/0! #DIV/0! Combined n/a n/a n/a n/a n/a n/a n/a 0.00 0.00 Combined n/a n/a n/a n/a n/a n/a 0.00 0.00 #DIV/0! Combined n/a n/a n/a Combined 10.52 3.98 0.63 0.69 0.12 0.02 2.50 0.19 0.29 0.00 0.00 0.00 0.47 2.03 1.50 0.32 0.20 0.53 US06 Power! 312.2 29.1 US06 30.6% 267 23.0 2453 5.19 99.1 764.8 5 6.5% 0.0% 12.4 n/a n/a US06 13% 5% 7% 8% 10% 57% 0% 0% 87.9% 95.4% 83.8% US06 n/a n/a n/a n/a n/a n/a n/a n/a 0.00 0.0% #DPV/0! #DIV/0! US06 n/a n/a n/a n/a n/a n/a n/a 0.00 0.00 US06 n/a n/a n/a n/a n/a n/a 0.00 0.00 " SDIV/0! n/a n/a n/a US06 9.35 4.22 0.76 0.52 0.07 0.00 2.86 0.13 0.33 0.00 0.00 0.00 0.46 2.40 1.75 0.49 0.12 0.62 tngme tngme Irans Disp Torque Type L Nm 2.4 220 base auto Perfo [ain Archi ecti # M of s gears k re G1 MG2 Battery ze size size W kW kWh 6 n/a n/a n/a rmance Metrics | 0-10mph | 0-30mph | 0-60mph | base 0-60 |30-50mph| 50-70mph| dist @ 3s 1.0 3.1 8.3 for using Ricardo maps % of FC s.s : Shift Optimizer Evaluation Gear Avg BMEP (bar) FTP Hwy 1 1.7 2.3 2 3.0 3.9 3 2.4 4.5 4 1.6 3.1 5 2.7 3.7 6 2.3 2.8 7 #DIV/0! #DPV/0! 8 #DIV/0! #DIV/0! US 06 F 4.2 1 7.1 2 6.5 2 6.7 2 6.7 2 4.0 1 #DIV/0! #DIV/0! Gear Avg BSFC (g/kWh) FTP Hwy 1 338 330 2 328 282 3 359 268 4 482 298 5 361 279 6 388 31 1 700 800 MG1=sun on planetary Recovered energy returned to whee s Gross recovered braking energy MG2=carrier (tractive) From alt regen braking (extra alternator load) x US 06 F 256 1 255 1 264 2 265 2 251 1 279 1 0 0 0 t ^ .2 5.1 20.5 Tables Avg RPM TP Hwy US06 121 1710 2155 309 2463 2881 188 2395 2974 60 1978 3209 D28 1869 2561 27 1737 2137 000 000 Total Energy (%) TP Hwy US06 3% 1% 8% 5% 1% 9% % 3% 10% 4% 7% 10% 2% 42% 16% % 46% 49% % 0% 0% % 0% 0% Figure 3-6 Sample output summary sheet for Standard Car (Camry) baseline 3-51 ------- Technologies Considered in the Agencies' Analysis Summary statistics were used as a first-order quality check on the model. Sample checks included: • were average engine speed and load within or close to the best BSFC region for the vehicle's engine map? • was transmission gear distribution reasonable and consistent between engine types? 3.3.1.2.16 Nominal runs The Ricardo "nominal" runs refer to the initial set of vehicle simulation models built for each vehicle architecture and vehicle class. These runs were used by EPA to assess the validity of the detailed model outputs (and hence the models themselves) prior to proceeding with the full design of experiment runs. Table 3-11 shows the summary results from the raw nominal runs for the conventional stop-start vehicles (including 12V stop-start, 70% efficient alternator, shift optimizer and alternator regen, as well as a 3.5% improvement due to engine friction reduction). Conventional automatic transmissions are assumed in all nominal runs. No road load reductions are included in these results. GHG reductions are in reference to the 2008 baseline vehicles. Table 3-11: Nominal Conventional Stop-Start modeling results Vehicle Engine Class Type Small Car Std Ca r La rge Ca r Small MPV LargeMPV Truck HD Truck STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB 2020 Diesel Displ. L 0.74 0.74 0.74 1.23 1.04 1.04 1.04 1.41 1.41 1.41 2.85 1.13 1.13 1.13 1.31 1.31 1.31 2.61 1.94 1.94 1.94 4.28 2.3 2.3 2.3 6.6 Torque Nm 157 157 157 221 220 220 220 298 298 298 503 239 239 239 277 277 277 460 410 410 410 694 486 486 486 895 Trans Type AT6 AT6 AT6 AT6 ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS FTP mpg 53.2 55.1 55.1 55.8 44.8 46.6 46.4 37.1 38.8 38.6 38.2 38.8 40.3 40.3 34.8 36.0 36.2 37.3 23.8 24.6 24.8 26.4 16.5 16.8 17.2 19.8 HW mpg 55.1 56.0 57.4 59.4 54.5 55.5 56.7 43.2 44.0 44.9 46.5 42.6 43.1 44.4 39.2 39.8 40.9 43.3 26.6 27.0 27.7 30.4 18.3 18.4 19.1 21.5 Comb mpg 54.0 55.5 56.1 57.4 48.7 50.2 50.5 39.6 41.0 41.2 41.5 40.4 41.5 42.0 36.7 37.6 38.2 39.8 25.0 25.6 26.0 28.1 17.3 17.5 18.0 20.5 0-30mph s 4.0 4.0 4.0 3.7 3.1 3.1 3.1 3.0 3.0 3.0 2.9 3.3 3.3 3.3 3.2 3.2 3.2 3.0 3.0 3.0 3.0 2.9 3.2 3.2 3.2 2.9 0-60mph s 10.0 10.0 10.0 9.8 8.5 8.5 8.5 7.4 7.4 7.4 7.5 8.9 8.9 8.9 8.6 8.6 8.6 8.6 8.1 8.1 8.1 8.0 9.8 9.8 9.8 8.8 %GHG Reduction 20% 22% 23% 16% 28% 31% 31% 31% 33% 33% 27% 25% 27% 28% 31% 33% 34% 30% 26% 28% 29% 26% 27% 28% 30% 31% 3-52 ------- Technologies Considered in the Agencies' Analysis Table 3-12 shows the results from the nominal runs for the P2 hybrid vehicles. Dual- clutch transmissions are assumed in all nominal runs. No road load reductions are included in these results. GHG reductions are in reference to the 2008 baseline vehicles. Table 3-12: Nominal P2 Hybrid modeling results Vehicle Engine Class Type Small Car Std Car La rge Ca r Small MPV LargeMPV Truck STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA Displ. L 0.59 0.59 0.59 1.66 1.66 0.83 0.83 0.83 2.4 2.4 1.13 1.13 1.13 3.8 3.8 0.9 0.9 0.9 2.6 2.6 1.05 1.05 1.05 3.15 3.15 1.55 1.55 1.55 4.6 4.6 Torque Nm 124 124 124 138 138 176 176 176 200 200 238 238 238 317 317 190 190 190 217 217 221 221 221 263 263 327 327 327 384 384 EM size kW 14 14 14 14 14 24 24 24 24 24 28 28 28 28 28 20 20 20 20 20 25 25 25 25 25 50 50 50 50 50 Battsize kWh 0.70 0.70 0.70 0.70 0.70 1.00 1.00 1.00 1.00 1.00 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.15 1.15 1.15 1.15 1.15 1.50 1.50 1.50 1.50 1.50 Trans Type DCT6 DCT6 DCT6 DCT6 DCT6 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 FTP mpg 68.2 68.4 70.2 70.8 71.7 61.9 62.9 65.1 64.6 65.9 49.8 50.4 51.7 49.9 51.1 50.1 50.8 52.0 52.9 54.1 47.7 47.4 47.6 48.3 48.8 32.5 33.0 33.8 33.2 33.9 HW mpg 57.3 57.7 59.9 59.0 60.5 57.2 58.0 59.7 59.7 61.0 46.5 46.8 48.3 46.2 47.4 44.2 44.5 46.1 45.5 46.8 42.2 42.6 43.0 42.4 43.5 28.4 28.6 29.6 29.0 29.7 Comb mpg 62.8 63.2 65.2 64.9 66.2 59.7 60.6 62.5 62.3 63.6 48.2 48.7 50.1 48.1 49.4 47.2 47.8 49.2 49.3 50.5 45.0 45.1 45.4 45.4 46.2 30.5 30.9 31.8 31.2 31.8 0-30mph s 3.8 3.8 3.8 3.7 3.7 3.6 3.6 3.6 3.4 3.4 3.4 3.4 3.4 3.0 3.0 3.9 3.9 3.9 3.7 3.7 3.8 3.8 3.8 3.6 3.6 3.3 3.3 3.3 3.1 3.1 0-60mph s 9.6 9.6 9.6 10.0 10.0 8.6 8.6 8.6 8.6 8.6 7.7 7.7 7.7 7.1 7.1 9.4 9.4 9.4 9.3 9.3 9.1 9.1 9.1 8.8 8.8 7.9 7.9 7.9 7.8 7.8 %GHG Reduction 31% 31% 33% 33% 35% 42% 42% 44% 44% 45% 43% 44% 45% 43% 44% 36% 36% 38% 38% 40% 44% 44% 44% 45% 45% 39% 40% 42% 40% 42% Table 3-13 shows the results from the nominal runs for the input powersplit vehiclesw. No road load reductions are included in these results. GHG reductions are in reference to the 2008 baseline vehicles. w While input powersplit hybrids remain a very likely hybrid architecture choice for some manufacturers, the Agencies focused on P2 hybrids compared to powersplits due to their apparent cost-effectiveness advantage in future years. As a result the powersplit nominal runs did not receive the same level of engineering scrutiny as the P2 hybrid nominal runs. 3-53 ------- Technologies Considered in the Agencies' Analysis Table 3-13: Nominal Powersplit hybrid modeling results Vehicle Engine Class Type Small Car Std Car La rge Ca r Small MPV LargeMPV STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA Displ. L 0.59 0.59 0.59 1.66 1.66 0.83 0.83 0.83 2.4 2.4 1.13 1.13 1.13 3.8 3.8 0.9 0.9 0.9 2.6 2.6 1.05 1.05 1.05 3.15 3.15 Torque Nm 124 124 124 138 138 176 176 176 200 200 238 238 238 317 317 190 190 190 217 217 221 221 221 263 263 EM size kW 14 14 14 14 14 80 80 80 80 80 28 28 28 28 28 20 20 20 20 20 25 25 25 25 25 Battsize kWh 0.70 0.70 0.70 0.70 0.70 1.00 1.00 1.00 1.00 1.00 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.15 1.15 1.15 1.15 1.15 Trans Type PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS PS FTP mpg 64.7 65.8 67.7 64.2 67.3 55.6 57.9 58.0 53.3 56.4 46.6 48.0 47.9 40.3 43.0 49.1 50.8 51.3 44.3 49.3 44.8 45.7 47.0 41.7 44.3 HW mpg 57.2 57.4 60.1 59.5 60.0 51.7 53.5 54.8 51.7 53.3 42.0 41.8 43.6 38.7 40.8 42.2 42.7 44.9 39.6 42.3 39.3 40.6 41.5 38.6 39.6 Comb mpg 61.1 61.7 64.0 62.0 63.8 53.8 55.8 56.5 52.6 55.0 44.4 45.0 45.9 39.6 42.0 45.8 46.8 48.2 42.1 45.9 42.1 43.3 44.4 40.3 42.0 0-30mph s 4.8 4.8 4.8 4.7 4.7 3.7 3.7 3.7 3.6 3.6 3.2 3.2 3.2 3.2 3.2 4.7 4.7 4.7 4.6 4.6 4.3 4.3 4.3 4.2 4.2 0-60mph s 10.4 10.4 10.4 9.8 9.8 8.7 8.7 8.7 8.0 8.0 7.8 7.8 7.8 7.1 7.1 10.3 10.3 10.3 9.1 9.1 9.7 9.7 9.7 8.8 8.8 %GHG Reduction 29% 30% 32% 30% 32% 35% 38% 38% 34% 37% 38% 39% 40% 31% 35% 33% 35% 37% 28% 34% 40% 42% 43% 37% 40% 3.3.1.2.17 Response Surface Model matrix runs After the nominal runs were completed according to the agreed-upon methodology, Ricardo set up a design of experiment matrix for each vehicle architecture. The continuously swept variables were randomized in a Latin hypercube fashion to achieve a representative sample within each matrix (reference the Ricardo report for more details on the complex systems modeling approach used). After a data review and removal of runs with errorsx (as needed) Ricardo then generated Response Surface Models (RSM) for use in the complex systems tool. EPA used the tool to evaluate a range of potential engine displacements, final drive ratios and electric motor sizes (hybrids only) for each vehicle package, in an effort to find the combination that would provide the greatest effectiveness while meeting EPA's definition of "equivalent performance". x e.g., model runs in which the vehicles were underpowered to the point where they could not follow the prescribed vehicle speed trace, rendering an invalid test or "error". These configurations were then excluded from the data sets. 3-54 ------- Technologies Considered in the Agencies' Analysis 3.3.1.2.18 Equivalent performance definition The Ricardo output data provides several performance metrics, as discussed in 3.3.1.2.10. For simplicity, EPA assumed that a range of acceleration times for both a 0-60 mph test and also a 0-30 mph test (emphasizing launch character) would provide a simple yet representative measure of a vehicle's equivalent performance. A range was chosen rather than assuming a single point value equal to the baseline. This provided more acceptable data points and reduced error due to "noise" in the datasets. The acceptable acceleration times were as follows with respect to the baseline: 0-60 mph: 5 percent slower to 15 percent faster as compared to baseline 0-30 mph: 10 percent slower to 20 percent faster as compared to baseline The range above reflects a deviation from the actual baseline value that is well within the normal variation of acceleration times for different vehicle models within a given vehicle class. 3.3.1.2.19 Treatment of "turbo lag" in performance runs for turbocharged engines A common critique of comparisons of the modeled performance of highly turbocharged engines with naturally-aspirated engines is that consideration must be given to the delay in producing full engine load associated with the turbocharger, commonly referred to as "turbo lag". In technical discussions, Ricardo's engine experts assured EPA that the dual-sequential designs of the turbocharger systems in the engines in this study should mitigate most of this phenomenon often seen on older-model vehicles. However, due to the heavy reliance on turbocharged engines as a significant source of motive force for the high BMEP engines evaluated in this project, EPA took this sensitivity further into account. Ricardo's initial model of WOT operation was based on a steady-state model of engine torque, assuming that the engine would be able to instantaneously reach a desired level of output torque, without consideration of the intake manifold filling dynamics or the mechanical inertia of the engine. EPA raised this as an issue, more in terms of properly representing vehicle performance than for effectiveness differences. EPA reviewed its own engine development data and proposed a somewhat conservative time constant for both the naturally aspirated engines (0.3 s) and the turbocharged engines (1.5 s), to apply to the engine torque response in the vehicle performance runs (these are shown below in Figure 3-7). In turn, Ricardo recalculated the acceleration times for the 0-30 and 0-60 mph runs to reflect the slower time constants. As a result, EPA used these two performance metrics exclusively in determining "equivalent performance". A transient engine/turbo model would have improved the accuracy of the model somewhat; however, it was beyond the scope of this project. 3-55 ------- Technologies Considered in the Agencies' Analysis 100 0% >? en no/ gine Output Torque ( * CTl C 3 O 1 D 0 i c^ c 1U 7O n% 0 0% -0 // r ^ 1 / 7 5 0.5 1 ^ / ^ ^~~~~ _ • — Ricarc — EPA-h EPA-B 5 2.5 3.5 4.5 5 Elapsed Time from WOT Command (sec) ' Jo-assumec at. Asp. oosted ^=^^=^^» I 5 6.5 7.5 Figure 3-7: EPA proposed time constants and resulting effect on torque rise time for turbocharging 3.3.1.2.20 Treatment of engine response and "turbo lag" in cycle simulations and control logic algorithms The EASY5 model used in the Ricardo simulations included engine and driveline inertia effects which account for some of the real-world transient torque delays. However, the simulation modeling did not include an adjustment to account for transient engine response delays (e.g. inclusion of time constant offsets), to simulate naturally aspirated and turbocharged engine response delays associated with intake manifold gas dynamics and turbocharger response delay. Consideration of engine response delay might affect how transmission shift optimization control logic and advanced HEV control logic is structured, and potentially affect GHG and fuel economy projections, particularly for boosted and downsized engines. EPA and Ricardo believe that the impact is small over the city and highway fuel economy test cycles. The agencies sought comment on the fuel economy impact of transient delays over the test cycles not accounted for in the Ricardo modeling, but there were no comments received, so the agencies have made no changes in this respect for the final rule analysis. 3.3.1.2.21 "Equivalent performance" results for conventional stop-start vehicles The following tables show the results from the complex systems tool, when displacement, final drive ratio and electric motor size are varied to optimize GHG and fuel consumption reduction effectiveness at equivalent performance for conventional stop-start, P2 and powersplit hybrids. Most of the vehicles show little change in performance between the nominal runs and the equivalent performance results from the complex systems tool. Table 3-56 ------- Technologies Considered in the Agencies' Analysis 3-14 through Table 3-18 illustrate the various effects of changing road loads on the various vehicle package configurations. Table 3-14, Table 3-16, and Table 3-16, respectively, show the equivalent performance results for the conventional stop-start (for both automatic transmissions and DCTs) and the P2 hybrid vehicles (modeled only as DCTs). No road load reductions are included in Table 3-14 through Table 3-16. For comparison, a second set of tables (Table 3-17 and Table 3-18) give equivalent performance results for conventional stop- start vehicles and P2 hybrids, each including example road load reductionsy of 20% mass reduction, 20% aerodynamic drag reduction and 10% rolling resistance reduction. The package effectiveness results from the equivalent performance runs were used in the datasets to calibrate the individual technology effectiveness values within the lumped parameter model. The development of the lumped parameter model is described in detail in Section 1.5 of EPA'sRIA. Table 3-14: Equivalent performance results for conventional-stop start vehicles (no road load reductions) Vehicle Engine Class Type Small Car StdCar La rge Ca r Small MPV LargeMPV Truck HD Truck STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB 2020 Diesel Displ. L 0.86 0.90 0.72 1.19 1.13 1.26 1.09 1.48 1.50 1.56 2.57 1.32 1.41 1.40 1.57 1.51 1.47 2.74 2.30 2.06 2.28 4.12 2.72 2.69 2.71 5.64 Torque Nm 183 190 154 213 240 266 230 314 317 330 454 280 297 296 332 319 312 483 486 435 482 669 575 568 573 764 Trans Type AT6 AT6 AT6 AT6 ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS FTP mpg 53.1 56.3 55.2 57.3 44.4 47.0 46.2 37.0 39.2 38.6 39.1 38.9 41.1 40.0 34.8 36.2 36.4 36.7 24.0 25.0 24.8 26.8 16.6 17.2 17.3 21.0 HW mpg 56.5 57.5 59.1 64.2 54.5 56.0 57.0 43.4 44.3 45.0 47.1 42.4 43.9 45.1 39.5 40.6 40.9 44.0 26.8 26.9 28.1 31.2 18.6 18.8 19.4 24.6 Comb mpg 54.6 56.9 56.9 60.2 48.5 50.6 50.5 39.6 41.3 41.2 42.3 40.4 42.3 42.1 36.8 38.0 38.3 39.7 25.2 25.8 26.2 28.6 17.4 17.9 18.2 22.5 0-30mph s 4.1 4.1 4.1 3.8 2.9 2.8 3.1 3.0 2.9 3.0 3.0 3.2 3.2 3.2 2.9 3.0 2.9 3.0 2.8 2.9 2.9 2.9 3.0 2.9 2.9 3.2 0-60mph s 9.1 8.9 10.1 10.0 7.9 7.2 8.3 7.2 7.1 7.0 8.1 8.0 7.7 7.7 7.4 7.7 7.6 8.4 7.0 7.6 7.2 8.3 8.4 8.4 8.4 10.3 %GHG Reduction 21% 24% 24% 20% 28% 31% 31% 31% 34% 34% 28% 25% 28% 28% 31% 34% 34% 29% 26% 28% 29% 28% 27% 29% 30% 37% y Note that in the regulatory fleet analysis, levels of road load reduction technologies (e.g., mass reduction) will vary by vehicle class. These tables are illustrative in nature. 3-57 ------- Technologies Considered in the Agencies' Analysis Table 3-15: Equivalent performance results for conventional-stop start vehicles with DCT transmissions (no road load reductions) Vehicle Engine Class Type Small Car Std Car La rge Ca r Small MPV LargeMPV Truck HD Truck STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB 2020 Diesel Displ. L 0.91 0.92 0.89 1.13 1.08 1.29 1.17 1.53 1.66 1.48 2.44 1.30 1.32 1.33 1.53 1.56 1.56 2.42 2.23 2.26 2.25 3.78 2.55 2.62 2.58 5.45 Torque Nm 193 196 188 204 229 273 248 324 352 313 431 276 280 282 324 330 330 427 472 478 475 613 538 554 544 739 Trans Type dry DCT6 dry DCT6 dry DCT6 dry DCT6 dry DCT8 dry DCT8 dry DCT8 dry DCT8 dry DCT8 dry DCT8 dry DCT8 dry DCT8 dry DCT8 dry DCT8 wet DCT8 wet DCT8 wet DCT8 wet DCT8 wet DCT8 wet DCT8 wet DCT8 wet DCT8 wet DCT8 wet DCT8 wet DCT8 wet DCT8 FTP mpg 55.0 58.0 57.2 61.4 46.4 48.7 48.1 38.4 40.5 40.0 41.0 40.1 42.1 41.7 36.0 38.0 37.6 39.2 24.8 25.9 25.8 28.1 17.3 17.8 18.0 21.8 HW mpg 58.8 59.8 61.3 69.4 55.0 57.5 57.6 44.0 45.4 45.6 48.4 43.6 44.7 45.6 40.2 41.1 41.8 45.2 27.1 27.7 28.1 32.1 18.1 18.7 19.0 24.2 Comb mpg 56.7 58.8 59.0 64.8 49.9 52.3 51.9 40.7 42.6 42.3 44.0 41.6 43.2 43.3 37.8 39.4 39.4 41.7 25.8 26.7 26.8 29.8 17.6 18.2 18.4 22.8 0-30mph s 3.9 3.9 3.9 3.9 3.1 3.0 3.0 2.9 2.9 3.0 3.0 3.1 3.2 3.1 3.1 3.0 3.0 3.1 3.0 3.0 3.0 3.0 3.1 3.1 3.1 3.3 0-60mph s 8.6 8.5 8.7 10.4 8.0 7.1 7.6 6.8 6.5 7.0 8.1 7.7 7.7 7.6 7.4 7.3 7.3 9.0 7.1 7.0 7.0 8.6 8.5 8.4 8.5 10.3 %GHG Reduction 23% 26% 27% 26% 30% 33% 33% 33% 36% 35% 31% 27% 30% 30% 33% 36% 36% 33% 28% 31% 31% 31% 28% 30% 31% 38% 3-58 ------- Technologies Considered in the Agencies' Analysis Table 3-16: Equivalent performance results for P2 hybrids (no road load reductions) Vehicle Engine Class Type Small Car Std Car La rge Ca r Small MPV LargeMPV Truck STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA Displ. L 0.68 0.68 0.67 1.72 1.68 1.00 0.95 1.04 2.54 2.31 1.39 1.37 1.38 3.73 3.33 1.40 1.39 1.41 3.87 3.59 1.31 1.30 1.29 3.13 3.00 1.87 1.92 1.92 5.34 5.34 Torque Nm 144 144 142 143 140 213 202 219 212 193 292 289 291 311 278 295 293 297 322 299 276 274 272 262 250 394 404 405 445 445 EM size kW 21 21 21 17 19 26 27 26 27 28 29 29 29 30 30 34 37 38 38 39 30 31 32 34 34 50 48 48 53 56 Battsize kWh 0.70 0.70 0.70 0.70 0.70 1.00 1.00 1.00 1.00 1.00 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.15 1.15 1.15 1.15 1.15 1.50 1.50 1.50 1.50 1.50 Trans Type DCT6 DCT6 DCT6 DCT6 DCT6 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 FTP mpg 68.9 70.1 72.0 72.0 74.4 62.2 63.2 64.8 64.6 65.7 50.6 51.3 52.6 48.6 50.7 52.3 53.0 54.4 53.6 55.2 48.5 49.0 49.2 48.0 48.5 33.3 33.6 34.6 32.3 32.7 HW mpg 58.7 59.2 61.2 60.8 62.0 57.7 58.3 60.4 59.5 60.7 47.3 47.9 49.0 46.1 47.7 45.5 45.9 47.2 46.2 47.4 42.3 42.6 42.7 42.3 43.0 29.0 29.3 30.2 28.8 29.4 Comb mpg 63.9 64.7 66.7 66.5 68.2 60.1 60.9 62.7 62.2 63.4 49.1 49.7 50.9 47.5 49.3 49.0 49.6 50.9 50.0 51.4 45.5 45.9 46.0 45.3 45.9 31.2 31.5 32.4 30.6 31.1 0-30mph s 3.7 3.7 3.7 3.9 3.8 3.4 3.4 3.4 3.4 3.4 3.3 3.4 3.4 3.2 3.3 3.6 3.5 3.4 3.6 3.7 3.2 3.2 3.2 3.2 3.2 3.3 3.4 3.3 3.1 3.0 0-60mph s 8.5 8.5 8.5 9.6 9.6 7.9 8.0 7.8 8.6 8.7 7.2 7.3 7.2 7.5 8.0 8.1 8.0 7.9 9.0 9.3 7.4 7.4 7.5 8.2 8.3 7.3 7.2 7.2 7.2 7.1 %GHG Reduction 32% 33% 35% 35% 36% 42% 43% 44% 44% 45% 44% 45% 46% 42% 44% 38% 39% 40% 39% 41% 45% 45% 45% 44% 45% 40% 41% 43% 39% 40% 3-59 ------- Technologies Considered in the Agencies' Analysis Table 3-17 Equivalent performance results for conventional-stop start vehicles (with 20% mass, 20% aerodynamic drag and 10% rolling resistance reductions) Vehicle Engine Class Type Small Car Std Car La rge Ca r Small MPV LargeMPV Truck HD Truck STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB 2020 Diesel STDI LBDI EGRB 2020 Diesel Displ. L 0.68 0.89 0.69 0.91 1.04 1.27 0.98 1.00 1.49 1.00 2.05 1.20 1.40 1.13 1.00 1.26 1.02 1.98 1.44 1.89 1.44 3.20 2.21 2.24 2.19 4.45 Torque Nm 145 189 146 164 220 268 207 212 315 212 362 253 296 238 212 266 216 349 303 399 305 518 466 473 463 603 Trans Type AT6 AT6 AT6 AT6 ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS ATS FTP mpg 65.0 68.9 67.6 71.8 53.9 57.3 56.2 46.5 48.4 48.5 48.5 46.3 49.1 48.4 42.4 44.2 44.2 46.4 29.4 30.2 30.5 32.8 20.0 20.5 20.9 25.3 HW mpg 70.0 72.4 73.1 83.2 67.6 70.6 70.1 53.8 55.0 55.9 59.7 51.8 53.5 53.6 46.8 48.1 48.7 54.0 32.1 32.9 33.6 38.8 22.2 22.6 23.1 30.1 Comb mpg 67.2 70.4 70.0 76.5 59.3 62.6 61.7 49.5 51.2 51.6 53.0 48.6 51.0 50.6 44.3 45.9 46.2 49.6 30.6 31.3 31.8 35.3 20.9 21.4 21.8 27.3 0-30mph s 4.1 4.2 4.1 3.7 2.9 2.8 3.0 3.1 3.0 3.1 3.0 3.2 3.3 3.2 3.2 2.9 3.2 3.0 3.1 2.8 3.1 3.0 3.0 3.0 3.0 3.2 0-60mph s 9.2 8.4 9.2 10.4 7.2 6.4 7.6 8.1 6.5 8.1 8.1 7.4 6.9 7.7 8.8 7.3 8.7 9.0 8.6 7.0 8.6 8.6 8.4 8.4 8.4 10.3 %GHG Reduction 35% 38% 38% 37% 41% 44% 43% 45% 46% 47% 42% 37% 40% 40% 43% 45% 45% 43% 39% 41% 42% 41% 39% 41% 42% 48% 3-60 ------- Technologies Considered in the Agencies' Analysis Table 3-18: Equivalent performance results for P2 hybrids (with 20% mass, 20% aerodynamic drag and 10% rolling resistance reductions) Vehicle Engine Class Type Small Car Std Car La rge Car Small MPV LargeMPV Truck STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA STDI LBDI EGRB ATKCS ATKDVA Displ. L 0.68 0.68 0.68 1.60 1.52 0.90 0.91 0.92 2.36 2.03 1.21 1.25 1.25 3.52 3.29 1.25 1.22 1.24 3.71 3.44 1.01 1.04 1.02 2.91 2.84 1.57 1.60 1.58 4.16 4.15 Torque Nm 143 144 143 133 127 191 194 194 197 169 254 263 263 293 274 265 257 262 309 287 213 219 215 243 237 330 337 334 347 346 EM size kW 11 11 11 11 11 18 18 18 18 18 22 21 21 21 21 21 22 21 21 21 28 28 26 21 22 41 38 40 38 39 Battsize kWh 0.70 0.70 0.70 0.70 0.70 1.00 1.00 1.00 1.00 1.00 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.10 1.15 1.15 1.15 1.15 1.15 1.50 1.50 1.50 1.50 1.50 Trans Type DCT6 DCT6 DCT6 DCT6 DCT6 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 DCT8 FTP mpg 85.8 87.6 89.5 89.4 93.9 78.1 79.7 81.4 82.2 83.5 63.2 64.9 65.7 61.1 63.9 63.9 65.2 66.5 65.0 67.5 59.5 61.0 60.6 58.9 60.1 39.4 40.3 41.0 39.9 41.4 HW mpg 72.2 73.1 75.4 74.9 76.9 71.1 72.2 74.2 73.8 76.2 57.3 58.5 59.8 57.0 59.3 53.4 53.9 55.7 55.1 56.7 50.2 50.9 51.6 51.1 52.4 34.4 35.0 36.0 34.9 35.9 Comb mpg 79.1 80.4 82.5 82.2 85.4 74.8 76.2 78.0 78.2 80.0 60.4 61.9 62.9 59.2 61.7 58.7 59.5 61.1 60.1 62.1 54.9 56.0 56.2 55.1 56.3 37.0 37.7 38.6 37.5 38.7 0-30mph s 3.7 3.7 3.7 3.8 3.8 3.2 3.3 3.2 3.1 3.3 3.1 3.1 3.1 3.0 3.0 3.5 3.5 3.5 3.4 3.6 3.2 3.2 3.2 3.2 3.2 3.2 3.3 3.2 3.0 3.0 0-60mph s 7.9 7.9 8.0 8.9 9.0 7.2 7.2 7.1 7.5 8.3 6.6 6.5 6.6 6.7 6.8 7.7 7.7 7.8 8.2 8.7 7.4 7.3 7.3 7.5 7.7 7.0 7.0 7.0 7.1 7.2 %GHG Reduction 45% 46% 47% 47% 49% 53% 54% 55% 55% 56% 55% 56% 56% 54% 56% 48% 49% 50% 49% 51% 54% 55% 55% 54% 55% 50% 51% 52% 50% 52% 3.3.1.2.22 Validation of vehicle simulation results Ricardo described the process used to validate the baseline vehicles in its reportl. Ideally it would be desirable to validate the simulation results with actual vehicle certification test data. However, due to the nature and intended time frame (10+ years into the future) of the technologies modeled within the vehicle classes, it is difficult to find many real-world examples of specific technologies at the level of development reflected within the latest simulation models. Furthermore, there are no current vehicles in production that contain all (or even a majority) of the multiple advanced technologies embedded within the models so it is difficult to make meaningful direct comparisons between actual vehicles and model results. Finally, there is no direct way to disaggregate the various advanced technologies and isolate only the relevant pieces for evaluation (e.g., an advanced turbocharged engine at an interim BMEP level with a baseline-level transmission without stop-start): the lumped parameter model was developed for this very analytical capability. A full description of the lumped parameter model (including example comparisons of existing vehicle models to lumped parameter estimates) is provided in 3.3.2. 3-61 ------- Technologies Considered in the Agencies' Analysis 3.3.1.2.23 The "efficient frontier" capability in Complex Systems tool A powerful feature of the Complex Systems tool is the "efficient frontier" function, which provides a graphical representation of the RSM data for the vehicle configuration of interest. The user can identify the combination of various attributes (engine displacement, final drive ratio, motor size, etc) that project the best model effectiveness. Figure 3-8 below is an example of the efficient frontier for a Standard Car with a cooled EGR turbocharged engine and a dry clutch DCT. The light red line along the top of the data set represents the best fuel economy at each 0-60 mph acceleration time within the desired window. The solid dark blue points represent the combinations that achieve both the desired 0-60 and 0-30 mph criteria for equivalent performance. In this way, it is easy to quantify the best effectiveness for a given technology package. Efficient Frontier: Camry_ConvSS-Camry EGRB_DCT 59,9 59,8 59.7 59.6 59,5 59,4 59,3 59,2 59,1 59.0 5:3,9 53,3 53,7 58.6 : 58.5 58.4 53,3 58,3 53,1 58.0 57.9 57,8 5" 57,6 57.5 57,4 57.3 57,2 57.1 57,1 Acceptable 0-60 mph time window 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 0-60 mph Figure 3-8: "Efficient Frontier" function in complex systems tool 3.3.1.2.24 Significance of the Complex Systems tool The complex systems tool was used not only to identify the optimal combination of input variables for each vehicle architecture, but also to analyze trends in the input variables for quality assurance (i.e., to make sure the response surface models made engineering sense), and to establish numerical relationships between these variables for the lumped parameter 3-62 ------- Technologies Considered in the Agencies' Analysis model calibration. Shown below are a few examples of the types of inquiries made via the complex systems tool: 3.3.1.2.24.1 Effects of motor size (HEVs) EPA reviewed the effects of motor size on hybrids. As motor size is increased, there is more opportunity to recapture energy during braking (because more powerful motors can recover all of the energy in more severe braking events). However, oversized motors also experience reduced efficiency as they operate in a less efficient operating region. This is shown in Figure 3-9 below, which shows a sweep of motor size vs. fuel economy for both the FTP/HWFE combined and also the high speed/load US06 cycle. Note that the optimum motor size increases with respect to the US06 cycle due to more severe braking and acceleration rates. Monte Carlo Results Plot 2: Monte Carlo Results 78 77 76 p75 | 74 •F 73 tl,72 "S 71 170! 69 68 67 60 49 48 47 46 45 44 43 4J 41 4n 0.5 0,6 0,7 0,8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 EMSIZE 0.5 0.6 0,7 0.3 0.9 1,0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 EMSIZE Figure 3-9: Electric motor sweeps for Standard Car class, P2 hybrid with stoichiometric GDI engine (left = FTP/HWFE test; right = US06 test) 3.3.1.2.24.2 Effects of engine displacement EPA reviewed the effects of engine displacement at equivalent performance to determine if there would be an "optimal" range of downsizing for best effectiveness. Surprisingly, there was little benefit beyond downsizing the engine past a minimal point. Shown in Figure 3-10 is an example complex systems tool graph with fuel economy plotted against engine displacement multiplier (compared to the "nominal" engine displacement) for the Truck class for three gasoline turbocharged engine packages and one diesel engine package (note all packages included 20% weight reduction, 20% aerodynamic drag reduction, and 10% rolling resistance reduction): • The diesel engine result shows that the nominal engine in this case was originally oversized because it was scaled on engine power not more accurately on engine torque and continued displacement reduction would improve fuel economy. For this package, the displacement for optimal fuel economy is smaller than 50% of 3-63 ------- Technologies Considered in the Agencies' Analysis the nominal value; however, when considering equivalent vehicle performance, the minimum diesel displacement increases to roughly 70% of the nominal value. In contrast, the gasoline turbo engine results shown reflect a relative insensitivity of displacement to fuel economy for these advanced vehicles. 391 33 37] LU ^ i LL " OJ 33- 32- 30- 29 28] F-150 Moiite Carlo Results Expected trend 0.5 0.6 0.7 0.8 0.9 1.0 1.1 DISPLACEMENT MULTIPLIER 1.2 1.3 Figure 3-10: Example displacement sweep for Truck class in complex systems tool Figure 3-10 shows that as modeled, the swept displacement range is not large enough for the advanced gasoline turbocharged engines. The displacement multiplier for these engines must be greater than 1.3x the nominal displacement before the fuel economy would degrade substantially. As the displacement drops below about 65% of the nominal (already downsized) value, the efficiency decreases, as the engine load must be much higher to provide the same required power. Regardless, the total fuel efficiency decrease from optimal is rather small compared to today's engines. A 27-bar cooled EGR turbocharged GDI engine map for 3-64 ------- Technologies Considered in the Agencies' Analysis a large car2 was reverse-engineered from the Ricardo 10 Hz output data, and is provided in Figure 3-11. The efficiency of this family of engines is very robust to changes in engine displacement because the highlighted BSFC region of interest (the second one out from the minimum BSFC "island") spans a large speed and load range. As a result, significant changes in displacement do not greatly reduce fuel efficiency. As displacement increases, the average operating points for the engine over a given test cycle will trend towards the lower left (lower speed, lower loadaa) portion of the map. In this case the points on the plot exist within the same BSFC contour, so there is little degradation in engine efficiency with increasing displacement (and drivetrain efficiency may improve at higher gears, potentially resulting in a fuel economy increase). Were the displacement to be increased much further, the operating region would cross the contour and fuel efficiency would begin to drop much more dramatically. z The 27 bar, cooled EGR turbocharged engine maps are similar for all classes as they originated from a common reference map and scaled according to engine displacement, as described in Section 6.3 of the 2011 Ricardo report. aa Load decreases as it is reflective of a % of the maximum achievable torque and torque is increasing with increased displacement. Speed decreases because of the greater torque available combined with the shift optimizer algorithm (allowing for a greater propensity to operate in higher gears). 3-65 ------- Technologies Considered in the Agencies' Analysis Avg. load (bmep) and speed decrease with increasing displacement 1000 2000 3000 4000 RPM 5000 6000 Figure 3-11: Advanced engine BSFC map (27-bar cooled EGR turbocharged GDI engine for large car) 3.3.1.2.25 Effects of mass reduction With the complex systems tool EPA isolated the effectiveness of mass reduction on advanced vehicle technology packages. Figure 3-12 below shows a mass reduction sweep plot of the Large MPV class for a conventional STDI and P2 hybrid vehicle with an Atkinson engine. 3-66 ------- Technologies Considered in the Agencies' Analysis Large MPV - weight sweep at equivalent performance 53 5J 51 50 -49 48 £ 47 1 16 45 44 4? 42 41 40 39 38 37 36 35 34 Atkinson-P2 hybrid -4.6% per 10% WR StoichGDI engine ~5.2% per 10% WR 0.725 0.750 0.775 O.SOO 0.825 0.850 0.875 0.900 0.925 0.950 0.975 1.000 1.025 1.050 1,075 1.100 1.125 Weight Factor Figure 3-12: Mass reduction sweep for Large MPV class at baseline equivalent performance. Engine displacement and motor size (hybrids) held constant. The mass reduction effectiveness, originally estimated at roughly 6% GHG reduction for a 10% reduction in mass, has been revised to reflect data such as that shown above. Isolated from benefits due to engine downsizing opportunities, the effectiveness of weight reduction for the non-hybrid packages is on the order of 5% per 10% weight reduction, while mass reduction for the P2 hybrid (or any hybrid) is reduced, on the order of 4.5% per 10% reduction due to the synergies with brake energy recovery (less braking energy is recoverable because the vehicle weighs less). The lumped parameter tool was also revised to incorporate the synergies of weight reduction and hybrids. 3.3.1.2.26 Vehicle simulation report peer review process As previously discussed, vehicle simulation modeling is a very detailed, mathematically intensive approach which relies heavily on numerical engineering inputs. These inputs (e.g., engine maps, transmission efficiency, control logic, etc.) are the heart of the model and are derived directly from proprietary engineering knowledge of components and subsystems. To simulate advanced engine and vehicle concepts, state-of-the-art knowledge must be applied and converted into modeling inputs. Public domain information is rarely at the forefront of technology, and of little use in modeling vehicles in the MYs 2017-2025 time frame. Engineering details on advanced vehicle technologies are closely guarded in industry, and engineering services companies which develop and generate this confidential information rely on it to remain competitive in the marketplace. Therefore, it is difficult, if not 3-67 ------- Technologies Considered in the Agencies' Analysis impossible, to be completely transparent with an advanced vehicle simulation model and make all of the inputs available for public review. EPA commissioned an external peer review of the 2011 Ricardo simulation project and report. The peer reviewers selected were highly respected members of academia and industry, all with substantial backgrounds in automotive technology. The list of peer reviewers and their credentials is provided in the associated peer review report35. EPA charged the peer reviewers to thoroughly evaluate the body of work with respect to the following topics: • Adequacy of the numerical inputs (engine technology selection, battery inputs, accessory load assumptions, etc.) and highlight any caveats or limitations that would affect the final results. • Validity and applicability of the simulation methodology, and if it adequately addresses synergies • The results, and their validity and applicability to the light-duty vehicle fleet in the 2020-2025 timeframe. • Completeness of the report (does it offer enough detail of the modeling process) • The overall adequacy of the report for predicting the effectiveness of these technologies, and suggest recommendations for improvement The first round of comments was reflective of the reviewer's lack of access to model inputs. Because the confidential inputs were initially withheld (for reasons described above), "lack of transparency" was a consistent theme amongst the reviewers, so much that they expressed frustration with their ability to evaluate the model methodology and the quality of the inputs. Additionally, due to the lack of access to Ricardo proprietary input data the peer reviewers expressed concern that they could not adequately judge the validity or accuracy of the input information or the simulation results. EPA worked with Ricardo to provide the peer reviewers with access to all of the detailed confidential modeling inputs under non-disclosure agreements. With this necessary information, 3 of the 5 peer reviewers submitted a second round of comments which were generally more specific. In turn, Ricardo modified the report to address some of the comments, and they developed a response to comments document which covered the comments from the peer review. One common theme called for increased detail in how the inputs were generated. To address these requests, Ricardo provided the detailed case studies that were used in the development of the engine maps for the cooled EGR boosted engines and the Atkinson engines for hybrids. Ricardo also elaborated on the hybrid control strategy, complete with state flow diagrams of operating modes, as well as a discussion of how hybrid control strategy was optimized. Additional transmission input details were provided, including an overview of the development of advanced gear efficiencies and how the optimized shift strategy was applied. The docket to this final rule contains Ricardo's response to comment document (which includes the first version of the Ricardo report that was peer reviewed and both rounds of peer review comments), and Ricardo's final report.36'37 The agencies sought comment on the all of these references and on the responsiveness of the final report to the peer review comments. 3-68 ------- Technologies Considered in the Agencies' Analysis 3.3.1.3 Argonne National Laboratory Simulation Study As discussed in the proposal, the U.S. D.O.T. Volpe Center has entered into a contract with Argonne National Laboratory (ANL) to provide full vehicle simulation modeling support for this MYs 2017-2025 rulemaking. While modeling was not complete in time for use in the NPRM, the ANL results were available for the final rule and were used to define the effectiveness of mild hybrids for both agencies, and NHTSA used the results to update the effectiveness of advanced transmission technologies coupled with naturally-aspirated engines for the CAFE analysis, as discussed above and more fully in NHTSA's RIA. This simulation modeling was accomplished using ANL's full vehicle simulation tool called "Autonomie," which is the successor to ANL's Powertrain System Analysis Toolkit (PSAT) simulation tool, and that includes sophisticated models for advanced vehicle technologies. The ANL simulation modeling process and results are documented in multiple reports that can be found in NHTSA's docket38. 3.3.2 Lumped parameter Modeling 3.3.2.1 Overview of the lumped parameter model As a more practical alternative to full vehicle simulation, EPA developed a "lumped parameter model" that estimates the effectiveness of various technology combinations or "packages," in a manner that accounts for synergies between technologies. In the analysis supporting the MYs 2012-2016 light duty vehicle GHG and CAFE rule, EPA built over 140 packages for use in its OMEGA model, which spanned 19 vehicle classes and over 1100 vehicle models. Vehicle simulation modeling performed for EPA by Ricardo, PLC, was used to calibrate the lumped parameter model. Although DOT's analysis supporting the MYs 2012-2016 CAFE rule applied technologies incrementally, rather than specifying packages in advance, DOT calibrated CAFE model inputs, using EPA's lumped parameter model, to harmonize as fully as practical with estimates produced by EPA's lumped parameter model. To support this rulemaking, EPA has updated its lumped parameter model and calibrated it with updated vehicle simulation work performed for EPA by Ricardo, PLC. As in the MYs 2012-2016 rulemaking, DOT has calibrated inputs, including synergy factors, to the CAFE model to as fully as practical align with estimates produced by EPA's lumped parameter model. Both agencies have continued to conduct and sponsor vehicle simulation efforts to improve inputs to the agencies' respective modeling systems. For the final rule, simulation results for the mild hybrid technology have been incorporated into the modeling systems for both agencies. Also, NHTSA updated the incremental effectiveness of advanced transmissions as applied to naturally-aspirated engines, a change which was only implemented in the CAFE model. The basis for EPA's lumped parameter analysis is a first-principles energy balance that estimates the manner in which the chemical energy of the fuel is converted into various forms of thermal and mechanical energy on the vehicle. The analysis accounts for the 3-69 ------- Technologies Considered in the Agencies' Analysis dissipation of energy into the different categories of energy losses, including each of the following: • Second law losses (thermodynamic losses inherent in the combustion of fuel), • Heat lost from the combustion process to the exhaust and coolant, • Pumping losses, i.e., work performed by the engine during the intake and exhaust strokes, • Friction losses in the engine, • Transmission losses, associated with friction and other parasitic losses of the gearbox, torque converter (when applicable) and driveline • Accessory losses, related directly to the parasitics associated with the engine accessories, • Vehicle road load (tire and aerodynamic) losses; • Inertial losses (energy dissipated as heat in the brakes) The remaining energy is available to propel the vehicle. It is assumed that the baseline vehicle has a fixed percentage of fuel lost to each category. Each technology is grouped into the major types of engine loss categories it reduces. In this way, interactions between multiple technologies that are applied to the vehicle may be determined. When a technology is applied, the lumped parameter model estimates its effects by modifying the appropriate loss categories by a given percentage. Then, each subsequent technology that reduces the losses in an already improved category has less of a potential impact than it would if applied on its own. Using a lumped parameter approach for calculating package effectiveness provides necessary grounding to physical principles. Due to the mathematical structure of the model, it naturally limits the maximum effectiveness achievable for a family of similar technologies'*. This can prove useful when computer-simulated packages are compared to a "theoretical limit" as a plausibility check. Additionally, the reduction of certain energy loss categories directly impacts the effects on others. For example, as mass is reduced the benefits of brake energy recovery decreases because there is not as much inertia energy to recapture. Figure 3-13 is an example spreadsheet used by EPA to estimate the package effectiveness and the synergistic impacts of a technology package for a standard-size car. bb For example, if only 4% of fuel energy is lost (in a baseline engine) to pumping work, leveraging multiple technologies to theoretically eliminate all pumping losses would yield an aggregate reduction of no more than 15% in fuel consumption. 3-70 ------- Technologies Considered in the Agencies' Analysis Vehicle Energy Effects Estimator i :; r I "M .". '• ' : : : . . j i L : : :.-^: t-. = r '-:: -:::.: IiU | Iri- ' :: : Uni:.-.; .%;i 7;.T.; -Sir:::. Trj.-j -u.^.: T:--: I : ; 7. ;•; -. - : - Trx =cr. 7 • jrr- r-l Mar L:i=T: T: : j,.-.r 1--I3J ITJVJX^XZJ i;^"^. ".7, :*:•- *•* :•: K. X" >. OJ? 071 077 •-^3.-i| :.l:i. I i.i^c |3r.-.;T»r.|Z;..:iI T^i TiJt T«ao Zto-»»a Qgi« r^ijari -. : : : r - : : ; : : - «." • .- --. : -• * '••'• .- ^ i ; r r vs : L: : : —£-.-. Figure 3-13 Sample lumped parameter model spreadsheet The LP model has been updated from the MYs 2012-2016 final rule to support the MYs 2017-2025 final standards. Changes were made to include new technologies for 2017 and beyond, improve fidelity for baseline attributes and technologies, and better represent hybrids based on more comprehensive vehicle simulation modeling. EPA RIA Chapter 1 provides details of the methodology used to update and refine the model. 3.3.2.2 Calibration of Lumped Parameter model to vehicle simulation data The lumped parameter model includes a majority of the new technologies being considered as part of this proposed rulemaking. The results from the Ricardo vehicle simulation project (See section 3.3.1 for additional information) were used to successfully calibrate the predictive accuracy and the synergy calculations that occur within the lumped 3-71 ------- Technologies Considered in the Agencies' Analysis parameter model. When the vehicle packages Ricardo modeled are estimated in the lumped parameter model, the results are comparable. All of the baselines for each vehicle class, as predicted by the lumped parameter model, fall within 3% of the Ricardo-modeled baseline results. With a few exceptions (discussed in Chapter 1 of EPA's RIA the lumped parameter results for the 2020-2025 "nominal" technology packages are within 5% of the vehicle simulation results. Shown below in Figure 3-14 through Figure 3-19 are Ricardo's vehicle simulation package results (for conventional stop-start and P2 hybrid packages00) compared to the lumped parameter estimates. Small Car Nominal Results Ricardo LP results Figure 3-14: Comparison of LP to simulation results for Small Car class ! Refer to 3.3. Ifor definitions of the baselines, "conventional stop-start" and "P2 hybrid" vehicle architectures. 3-72 ------- Technologies Considered in the Agencies' Analysis Standard Car Nominal Results Ricardo LP results Figure 3-15: Comparison of LP to simulation results for Standard Car class Large Car Nominal Results 60 50 40 30 20 10 Illllllll Illllllll I Ricardo I LP results ^ ------- Technologies Considered in the Agencies' Analysis Small MPV Nominal Results 60 50 40 30 20 10 n • • • • • • i • • • • • • • i • • • • • • • I Ricardo I LP results 4? ^ c9» | Conventional SS P2 Hybrid Figure 3-17: Comparison of LP to simulation results for Small MPV class Large MPV Nominal Results Ricardo LP results Figure 3-18: Comparison of LP to simulation results for Large MPV class 3-74 ------- Technologies Considered in the Agencies' Analysis Truck Nominal Results Figure 3-19: Comparison of LP to simulation results for Truck class The recent ANL modeling results for mild hybrids largely confirmed the effectiveness as originally predicted by the lumped parameter model, with minor differences for small cars and large trucks. A comparison of the ANL results to the original lumped parameter results (for comparable vehicle classes when modeled with a nominal 15 kW motor size) is shown below in Table 3-19 and Table 3-20. Table 3-19 ANL Effectiveness for Mild Hybrid FC reduction Compact 11.6% Midsize 11.6% Small SUV 10.2% Midsize SUV 10.5% Pickup 8.5% Table 3-20 Lumped Parameter Model Effectiveness for Mild Hybrid FC reduction Small Car 14.1% Std Car 11.8% Small MPV 10.1% Large MPV 10.1% Truck 6.9% The underlying structure of the lumped parameter model was not changed to accommodate this new information; instead, the nominal 15 kW motor sizes for small cars 3-75 ------- Technologies Considered in the Agencies' Analysis and pickup truck mild hybrids were adjusted (to 10 kW and 18 kW, respectively) to reflect the updated effectiveness results provided by the ANL simulation work. 3.3.2.3 Comparison of results to real-world examples To validate the lumped parameter model, representations of actual late-model production vehicles exhibiting advanced technologies were created. Shown in Table 3-21 are a set of select vehicle models containing a diverse array of technologies: included are the pertinent technologies and vehicle specifications, along with actual vehicle certification fuel economy test data compared to the lumped parameter fuel economy estimates. For the vehicles and technologies shown, the predicted fuel economy is within about 3% of the actual data. Table 3-21: Production vehicle certification data compared to lumped parameter predictions Vehicle Vehicle Class Engine Transmission HEV motor (kW) ETW (Ibs) City/HW FE (mpg) LP estimate (mpg) Key technologies applied in LP model 20 11 Chevy Craze ECO Small Car 1.4L 14 Turbo GDI 6 speed auto n/a 3375 40.3 40.2 GDI (stoich) Turbo (30% downsize) Ultra low R tires Active grill shutters 2011 Sonata Hybrid Standard Car 2.4L 14 Atkinson 6 speed auto 30 3750 52.2 51.7 P2 hybrid Aero improvements 20 11 Escape Hybrid Small MPV 2.5L 14 Atkinson CVT 67 4000 43.9 44.0 Powersplit hybrid 2011F-150 EcoBoost Track 3.5LV6TurboGDI 6 speed auto n/a 6000 22.6 21.9 GDI (stoich) Turbo (37% downsize) 3.4 What cost and effectiveness estimates have the agencies used for each technology? As discussed in the previous sections, many the effectiveness estimates for this final rule, consistent with the proposal, including the estimates for the technologies carried over from the MYs 2012-2016 final rule, are derived from the 2011 Ricardo study and corresponding updated version of the lumped-parameter model. It is important to note that the agencies used the average of the range presented when referencing the effectiveness 3-76 ------- Technologies Considered in the Agencies' Analysis estimates from the MYs 2012-2016 final rule. If, for example, the effectiveness range for technology X was determined to be 1 to 2 percent, the agencies used a value of 1.5 percent in their respective analyses. However, the effectiveness ranges that are presented for the MYs 2017-2025 analysis, as informed by the Ricardo 2011 study, define the range of estimates used by the agencies for the different vehicle types. Again using technology X as an example, if the range is now defined as 2.0 to 2.5 percent then for small passenger cars (subcompact or compact) the estimated effectiveness might be 2.0 percent but for large cars an estimate of 2.5 percent might be used. As noted in section 3.1.3, the effects of learning on individual technology costs can be seen in the cost tables presented throughout this section 3.3. For each technology, we show direct manufacturing costs for the years 2017 through 2025. The changes shown in the direct manufacturing costs from year-to-year reflect the cost changes due to learning effects. 3.4.1 Engine technologies As indicated in the cost tables that found in this section, the agencies updated the costing approach for some technologies in an effort to provide better granularity in our estimates. This is reflected in Table 3-23, among others, listing costs for technologies by engine configuration—in-line or "I" versus "V"—and/or by number of cylinders. In the MYs 2012-2016 final rule, we showed costs for identified vehicle classes such as small car, large car, large truck, etc. The identified challenges inherent with that approach are that different vehicle classes can have many different sized engines. This condition may become more prominent going forward as more turbocharged and downsized engines enter the fleet. For example, the agencies project that many vehicles in the large car class, have large displacement V8 or V6 engines would move to highly turbocharged 14 engines under the final rule, consistent with the proposal. As such, we would not want to estimate the costs of engine friction reduction for large cars—which have always and continue to be based on the number of cylinders—by assuming that all large cars have V8 or V6 engines. 3.4.1.1 Low Friction Lubricants A basic method of reducing fuel consumption in gasoline engines is using of lower viscosity engine lubricants. Advanced multi-viscosity engine oils are available today which yield improved performance in a wider temperature band and with better lubricating properties. These advances are accomplished by changes to the oil base stock (e.g., switching engine lubricants from a Group I base oils to lower-friction, lower viscosity Group III synthetic) and through changes to lubricant additive packages (e.g., friction modifiers and viscosity improvers). The use of 5W-30 motor oil is now widespread and auto manufacturers are introducing the use of lower viscosity oils, such as 5W-20 and OW-20, to improve cold- flow properties and reduce cold start friction. However, in some cases, changes to the crankshaft, connecting rod and main crankshaft bearing designs and/or materials along with the mechanical tolerances of engine components may be required. In all cases, durability testing would be required to ensure that durability is not compromised. Shifting to lower viscosity and lower friction lubricants can also improve the management of valvetrain technologies such as cylinder deactivation or variable valve timing, which rely on a minimum oil temperature (viscosity) for operation. 3-77 ------- Technologies Considered in the Agencies' Analysis Several manufacturers have previously commented confidentially that low friction lubricants could have an effectiveness value between 0 to 1 percent. The agencies used the average effectiveness of 0.5 in the MYs 2012-2016 final rule. For purposes of this final rule, consistent with the proposal, the agencies relied on the lumped parameter model and the range for the effectiveness of low friction lubricant is 0.5 to 0.8 percent. In the MYs 2012-2016 final rule, the 2010 TAR and the MYs 2014-2018 Medium and Heavy Duty GHG and Fuel Efficiency final rule, EPA and NHTSA used a direct manufacturing cost (DMC) of \$3 (2007\$) and considered that cost to be independent of vehicle class since the engineering work required should apply to any engine size. The agencies continue to believe that this cost is appropriate and, having adjusted for 2010\$, the cost remains the same for this analysis. No learning is applied to this technology so the DMC remains \$3 (2010\$) year-over-year. The agencies have used a low complexity ICM of 1.24 for this technology through 2018 and 1.19 thereafter. The resultant costs are shown in Table Note that low friction lubes are expected to exceed 85 percent penetration by the 2017 3-22. MY. dd Table 3-22 Costs for Engine Modifications to Accommodate Low Friction Lubes (2010\$) Cost type DMC 1C TC Engine type All All All 2017 \$3 \$1 \$4 2018 \$3 \$1 \$4 2019 \$3 \$1 \$4 2020 \$3 \$1 \$4 2021 \$3 \$1 \$4 2022 \$3 \$1 \$4 2023 \$3 \$1 \$4 2024 \$3 \$1 \$4 2025 \$3 \$1 \$4 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. 3.4.1.2 Engine Friction Reduction In addition to low friction lubricants, manufacturers can also reduce friction and improve fuel consumption by improving the design of engine components and subsystems. Approximately 10 percent of the energy consumed by a vehicle is lost to friction, and just over half is due to frictional losses within the engine.39 Example improvements include low- tension piston rings, piston skirt design, roller cam followers, improved crankshaft design and bearings, material coatings, material substitution, more optimal thermal management, and piston and cylinder surface treatments. Additionally, as computer-aided modeling software continues to improve, more opportunities for evolutionary friction reductions may become available. All reciprocating and rotating components in the engine are potential candidates for friction reduction where minute improvements in several components can result in a dd Note that the costs developed for low friction lubes for this analysis reflect the costs associated with any engine changes that would be required as well as any durability testing that may be required. 3-78 ------- Technologies Considered in the Agencies' Analysis measurable fuel economy improvement. In the MYs 2012-2016 final rule, the agencies relied on the 2002 NAS, NESCCAF and EEA reports, as well as, confidential manufacturer data that suggested a range of effectiveness for engine friction reduction (EFR1) to be between 1 to 3 percent. Because of the incremental technology application capability of the CAFE model, NHTSA used the narrower range of 1 to 2 percent, which resulted in an average effectiveness of 1.5 percent. Based on the 2011 Ricardo study results, the agencies have revised the effectiveness for engine friction reduction range to 2.0 to 2.7 percent for this analysis. For this final rule, consistent with the proposal, the agencies added a second level of incremental improvements in engine friction reduction (EFR2) applicable over multiple vehicle redesign cycles. This second level of engine friction reduction forecasts additional improvements to low friction lubricants relative to the low friction lubricant technology discussed above and is considered to be mature only after MY 2017. The effectiveness for this second level, relative to the base engine, is 3.4 to 4.8 percent based on the lumped parameter model. Because of the incremental technology application capability of the CAFE model, NHTSA used the effectiveness range of 0.83 to 1.37 percent incremental to the first level of engine friction reduction and low friction lubricants for a total effectiveness of 2.83 to 4.07 percent. In the MYs 2012-2016 rule, the 2010 TAR and the MYs 2014-2018 Medium and Heavy Duty GHG and Fuel Efficiency final rule, NHTSA and EPA used a EFR1 cost estimate of \$11 (2007\$) per cylinder DMC, or \$12 (2010\$) per cylinder in this analysis. No learning is applied to this technology so the DMC remains \$12 (2010\$) year-over-year. The agencies have used a low complexity ICM of 1.24 for this technology through 2018 and 1.19 thereafter. The resultant costs are shown in Table 3-23. Note that EFR1 is expected to exceed 85 percent penetration by MY 2017. Table 3-23 Costs for Engine Friction Reduction - Level 1-EFR1 (2010\$) Cost type DMC DMC DMC DMC 1C 1C 1C 1C TC TC TC TC Engine type 13 14 V6 V8 13 14 V6 V8 13 14 V6 V8 2017 \$36 \$48 \$71 \$95 \$9 \$11 \$17 \$23 \$44 \$59 \$89 \$118 2018 \$36 \$48 \$71 \$95 \$9 \$11 \$17 \$23 \$44 \$59 \$89 \$118 2019 \$36 \$48 \$71 \$95 \$7 \$9 \$14 \$18 \$43 \$57 \$85 \$113 2020 \$36 \$48 \$71 \$95 \$7 \$9 \$14 \$18 \$43 \$57 \$85 \$113 2021 \$36 \$48 \$71 \$95 \$7 \$9 \$14 \$18 \$43 \$57 \$85 \$113 2022 \$36 \$48 \$71 \$95 \$7 \$9 \$14 \$18 \$43 \$57 \$85 \$113 2023 \$36 \$48 \$71 \$95 \$7 \$9 \$14 \$18 \$43 \$57 \$85 \$113 2024 \$36 \$48 \$71 \$95 \$7 \$9 \$14 \$18 \$43 \$57 \$85 \$113 2025 \$36 \$48 \$71 \$95 \$7 \$9 \$14 \$18 \$43 \$57 \$85 \$113 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. The agencies have estimated the DMC of the second level of friction reduction and low friction lubricants at double the combined DMCs of EFR1 (double the DMC relative to the baseline). As a result, the costs of EFR2 are as shown in Table 3-24. For EFR2 the agencies have used a low complexity ICM of 1.24 through 2024 and 1.19 thereafter. 3-79 ------- Technologies Considered in the Agencies' Analysis Table 3-24 Costs for Engine Friction Reduction - Level 2 - EFR2 (2010\$) Cost type DMC DMC DMC DMC 1C 1C 1C 1C TC TC TC TC Engine type 13 14 V6 V8 13 14 V6 V8 13 14 V6 V8 2017 \$78 \$102 \$149 \$197 \$19 \$25 \$36 \$48 \$97 \$126 \$185 \$244 2018 \$78 \$102 \$149 \$197 \$19 \$25 \$36 \$48 \$97 \$126 \$185 \$244 2019 \$78 \$102 \$149 \$197 \$19 \$25 \$36 \$48 \$97 \$126 \$185 \$244 2020 \$78 \$102 \$149 \$197 \$19 \$25 \$36 \$48 \$97 \$126 \$185 \$244 2021 \$78 \$102 \$149 \$197 \$19 \$25 \$36 \$48 \$97 \$126 \$185 \$244 2022 \$78 \$102 \$149 \$197 \$19 \$25 \$36 \$48 \$97 \$126 \$185 \$244 2023 \$78 \$102 \$149 \$197 \$19 \$25 \$36 \$48 \$97 \$126 \$185 \$244 2024 \$78 \$102 \$149 \$197 \$19 \$25 \$36 \$48 \$97 \$126 \$185 \$244 2025 \$78 \$102 \$149 \$197 \$15 \$20 \$29 \$38 \$93 \$121 \$178 \$234 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. 3.4.1.3 Cylinder Deactivation In conventional spark-ignition engines, throttling the intake airflow controls engine torque output. At partial loads, efficiency can be improved by using cylinder deactivation instead of throttling. Cylinder deactivation (DEAC) can improve engine efficiency by disabling or deactivating (usually) half of the cylinders when the load is less than half of the engine's total torque capability. Cylinder deactivation is achieved by keeping specific cylinder valves closed and stopping fuel flow to the specified cylinder. As a result, the trapped air within the deactivated cylinders is simply compressed and expanded as an air spring, with reduced friction and heat losses. The active cylinders combust at almost double the load required if all of the cylinders were operating. Overall engine pumping losses are significantly reduced as long as the engine is operated in this "part-cylinder" mode. Cylinder deactivation control strategy relies on setting maximum manifold absolute pressures or predicted torque ranges where it is acceptable to deactivate engine cylinders. Noise and vibration issues reduce the operating range where cylinder deactivation is allowed, although manufacturers continue exploring vehicle changes that enable increasing the amount of time that cylinder deactivation might be suitable. Some manufacturers may choose to adopt active engine mounts and/or active noise cancellation systems to address NVH concerns and allow a greater operating range of activation which is also shown in the cost estimates for this technology. Most manufacturers have legitimately stated that use of DEAC on 4 cylinder engines would cause unacceptable NVH; therefore, as in the MYs 2012-2016 rule and the 2010 TAR, the agencies are not applying cylinder deactivation to 4-cylinder engines in evaluating potential emission reductions/fuel economy improvements and associated costs. Cylinder deactivation has seen a recent resurgence thanks to better valvetrain designs and engine controls. General Motors and Chrysler Group have incorporated cylinder deactivation across a substantial portion of their V8-powered vehicles and Honda offers V6 models with cylinder deactivation. 3-80 ------- Technologies Considered in the Agencies' Analysis Effectiveness improvements scale roughly with engine displacement-to-vehicle weight ratio: the higher displacement-to-weight vehicles, operating at lower relative loads for normal driving, have the potential to operate in part-cylinder mode more frequently. NHTSA and EPA reviewed estimates from the MYs 2012-2016 final rule, 2010 TAR, the RIA for the MYs 2014-2018 Medium and Heavy Duty GHG and Fuel Efficiency final rule. The lumped parameter model applied a 6 percent reduction in CC>2 emissions depending on vehicle class. The CAFE model, due to its incremental technology application capability, used a range depending on the engine valvetrain configuration. For example, DOHC engines already equipped with DCP and DVVLD achieve little benefit, 0.5 percent for DEACD, from adding cylinder deactivation since the pumping work has already been minimized and internal Exhaust Gas Recirculation (EGR) rates are maximized. However, SOHC engines, which have CCP and DWLS applied, achieve effectiveness ranging from 2.5 to 3 percent for DEACS. And finally, OHV engines, without VVT or VVL technologies, achieved effectiveness for DEACO ranging from 3.9 to 5.5 percent. For this final rule, consistent with the proposal, the agencies, taking into account the additional review and the work performed for the 2011 Ricardo study, have revised the effectiveness estimates for cylinder deactivation. The effectiveness relative to the base engine is 4.7 to 6.5 percent based on the lumped parameter model. Because of the incremental technology application capability of the CAFE model, NHTSA used the effectiveness range of 0.44 to 0.66 percent incremental for SOHC and DOHC applications. For OHV applications having no incremental application of VVT or VVL, the effectiveness was increased to a range of 4.66 to 6.30 percent. In the MYs 2012-2016 final rule and the 2010 TAR, the agencies used a DMC estimate of \$140 (2007\$) and \$157 (2007\$) for cylinder deactivation technology on V6 and V8 engines, respectively. Adjusted for 2010\$, the DMCs become \$146 (2010\$) and \$165 (2010\$) for this analysis and are considered applicable in MY 2015. This technology is considered to be on the flat-portion of the learning curve. The agencies have applied a low complexity ICM of 1.24 to this technology through 2018 and 1.19 thereafter. The resultant costs are shown in Table 3-25. Table 3-25 Costs for Cylinder Deactivation (2010\$) Cost type DMC DMC 1C 1C TC TC Engine type V6 V8 V6 V8 V6 V8 2017 \$139 \$157 \$56 \$63 \$196 \$220 2018 \$136 \$153 \$56 \$63 \$193 \$217 2019 \$134 \$150 \$42 \$47 \$176 \$198 2020 \$131 \$147 \$42 \$47 \$173 \$195 2021 \$128 \$144 \$42 \$47 \$170 \$191 2022 \$126 \$142 \$42 \$47 \$168 \$189 2023 \$123 \$139 \$42 \$47 \$165 \$186 2024 \$121 \$136 \$42 \$47 \$162 \$183 2025 \$118 \$133 \$42 \$47 \$160 \$180 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. There is potential that, on engines already equipped with the mechanisms required for cylinder deactivation capability, the cost of DEAC as applied to SOHC and DOHC engines could be as low as \$32 in MY 2017. This \$32 accounts for the potential additional 3-81 ------- Technologies Considered in the Agencies' Analysis application of active engine mounts on SOHC and DOHC engines that, while having the potential to apply cylinder deactivation, may or would require these additional NVH improving devices for consumer acceptance. For this analysis, this additional expanded application and expense is only applied on 50 percent of the vehicles. Further, this SOHC and DOHC engine estimate is relevant to the CAFE model only because the OMEGA model does not apply technologies in the same incremental fashion as the CAFE model. 3.4.1.4 Variable Valve Timing (WT) Variable valve timing (VVT) encompasses a family of valve-train designs that alter the timing of the intake valve, exhaust valve, or both, primarily to reduce pumping losses, increase specific power, and control the level of residual gases in the cylinder. VVT reduces pumping losses when the engine is lightly loaded by controlling valve timing closer to an optimum needed to sustain horsepower and torque. VVT can also improve volumetric efficiency at higher engine speeds and loads. Additionally, VVT can be used to alter (and optimize) the effective compression ratio where it is advantageous for certain engine operating modes (e.g., in the Atkinson Cycle). VVT has now become a widely adopted technology: in MY 2011, approximately 93.8 percent of all new cars and light trucks had engines with some method of variable valve timing.40 Manufacturers are currently using many different types of variable valve timing, which have a variety of different names and methods. Manufacturers are currently using many different types of variable valve timing, which have a variety of different names and methods. Therefore, the degree of further improvement across the fleet is limited by the level of valvetrain technology already implemented on the vehicles. Information found in the 2008 and 2010 baseline vehicle fleet files is used to determine the degree to which VVT technologies have already been applied to particular vehicles to ensure the proper level of VVT technology, if any, is applied. The three major types of VVT are listed below. Each of the three implementations of VVT uses a cam phaser to adjust the camshaft angular position relative to the crankshaft position, referred to as "camshaft phasing." The phase adjustment results in changes to the pumping work required by the engine to accomplish the gas exchange process. The majority of current cam phaser applications use hydraulically-actuated units, powered by engine oil pressure and managed by a solenoid that controls the oil pressure supplied to the phaser. 3.4.1.4.1 Intake Cam Phasing (ICP) Valvetrains with Intake Cam Phasing (ICP), which is the simplest of the cam phasing technologies, can modify the timing of the inlet valves by phasing the intake camshaft while the exhaust valve timing remains fixed. This requires the addition of a cam phaser on each bank of intake valves on the engine. An in-line 4-cylinder engine has one bank of intake valves, while V-configured engines have two banks of intake valves. In the MYs 2012-2016 final rule and 2010 TAR, NHTSA and EPA assumed an effectiveness range of 2 to 3 percent for ICP. Based on the additional information from the 3-82 ------- Technologies Considered in the Agencies' Analysis 2011 Ricardo study and updated lumped parameter model the agencies have been able to fine- tuned the effectiveness range to be 2.1 to 2.7 percent for this analysis. In the MYs 2012-2016 rule and the 2010 TAR, the agencies estimated the DMC of a single cam phaser for ICP at \$37 (2007\$). This DMC, adjusted for 2010\$, becomes \$39 (2010\$) for this analysis and is considered applicable in the 2015 MY. This cost would be required for each cam shaft controlling intake valves. As such an OHC 14 and OHV V6 or V8 would need one cam phaser while an OHC V6 or V8 would need two cam phasers. This technology is considered to be on the flat-portion of the learning curve. The agencies have applied a low complexity ICM of 1.24 to this technology through 2018 and 1.19 thereafter. The resultant costs are shown in Table 3-26. Table 3-26 Costs for WT-Intake Cam Phasing - ICP (2010\$) Cost type DMC DMC DMC 1C 1C 1C TC TC TC Engine type OHC-I4 OHC-V6/V8 OHV-V6/V8 OHC-I4 OHC-V6/V8 OHV-V6/V8 OHC-I4 OHC-V6/V8 OHV-V6/V8 2017 \$37 \$74 \$37 \$9 \$19 \$9 \$46 \$93 \$46 2018 \$36 \$72 \$36 \$9 \$19 \$9 \$46 \$91 \$46 2019 \$35 \$71 \$35 \$7 \$15 \$7 \$43 \$86 \$43 2020 \$35 \$70 \$35 \$7 \$15 \$7 \$42 \$84 \$42 2021 \$34 \$68 \$34 \$7 \$15 \$7 \$42 \$83 \$42 2022 \$33 \$67 \$33 \$7 \$15 \$7 \$41 \$82 \$41 2023 \$33 \$65 \$33 \$7 \$15 \$7 \$40 \$80 \$40 2024 \$32 \$64 \$32 \$7 \$15 \$7 \$40 \$79 \$40 2025 \$31 \$63 \$31 \$7 \$15 \$7 \$39 \$78 \$39 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; OHC=overhead cam; OHV=overhead valve; all costs are incremental to the baseline. 3.4.1.4.2 Coupled Cam Phasing (CCP) Valvetrains with coupled (or coordinated) cam phasing can modify the timing of both the inlet valves and the exhaust valves an equal amount by phasing the camshaft of a single overhead cam (SOHC) engine or an overhead valve (OHV) engine. For SOHC engines, this requires the addition of a cam phaser on each bank of the engine. Thus, an in-line 4-cylinder engine has one cam phaser, while SOHC V-engines have two cam phasers. For OHV engines, which have only one camshaft to actuate both inlet and exhaust valves, CCP is the only VVT implementation option available and requires only one cam phaser.ee The agencies' MYs 2012-2016 final rule estimated the effectiveness of CCP to be between 1 to 4 percent. Due to the incremental technology application capability of the ee It is also noted that coaxial camshaft developments would allow other VVT options to be applied to OHV engines. However, since they would potentially be adopted on a limited number of OHV engines NHTSA did not include them in the decision tree. 3-83 ------- Technologies Considered in the Agencies' Analysis CAFE model, NHTSA estimated the effectiveness for CCP to be 1 to 3 percent for a SOHC engine and 1 to 1.5 percent for an overhead valve engine. For this final rule, consistent with the proposal, the agencies, have revised the estimates for CCP taking into account the additional review and the work performed for the 2011 Ricardo study. The effectiveness relative to the base engine is 4.1 to 5.5 percent based on the lumped parameter model. Because of the incremental nature of the CAFE model, NHTSA used the incremental effectiveness range of 4.14 to 5.36 percent for SOHC applications; an increase over the MYs 2012-16 final rule and 2010 TAR. For OHV applications, CCP was paired with discrete variable valve lift (DVVL) to form a new technology descriptor called variable valve actuation (VVA). Effectiveness values for this new descriptor is discussed later in Section 3.4.1.6. In regard to CCP costs, the same cam phaser has been assumed for intake cam phasing as for coupled cam phasing, thus the DMCs for CCP is identical to those presented for ICP in Table 3-26. 3.4.1.4.3 Dual Cam Phasing (DCP) The most flexible VVT design is dual (independent) cam phasing (DCP), where the intake and exhaust valve opening and closing events are controlled independently. This allows the option of controlling valve overlap, which can be used as an internal EGR strategy. At low engine loads, DCP creates a reduction in pumping losses, resulting in improved fuel consumption/reduced CC>2 emissions. Increased internal EGR also results in lower engine-out NOx emissions. Fuel consumption and CC>2 emissions improvements enabled by DCP are dependent on the residual tolerance of the combustion system. Additional improvements are observed at idle, where low valve overlap could result in improved combustion stability, potentially reducing idle fuel consumption. For forward looking technology application, DCP is only applicable to dual overhead cam (DOHC) engines.ff For the MYs 2012-2016 final rule and 2010 TAR, the EPA and NHTSA assumed an effectiveness range for DCP of 3 to 5 percent relative to a base engine or 2 to 3 relative to an engine with ICP. The agencies have updated this range, based on the updated lumped- parameter model, to be 4.1 to 5.5 percent relative to a base engine or 2.0 to 2.7 percent relative to an engine with ICP. ff The agencies note at least one production implementation of an OHV dual cam phasing is included in the baseline fleet. This consisted of a single concentric camshaft (a "camshaft within a camshaft") and a single dual vane phaser assemblies enabling independent phasing of the intake and exhaust camshaft profiles. However, this technology was applied to a limited production sports car versus a mass market application with significant sales volume. The agencies are not aware of any similar application moving forward. 3-84 ------- Technologies Considered in the Agencies' Analysis The costs for DCP are the same per phaser as described above for ICP. However, for DCP, an additional cam phaser is required for each camshaft controlling exhaust valves. As a result, a dual overhead cam 14 would need two phasers and a dual overhead cam V6 or V8 would need four phasers, and an overhead valve V engine would need two.gg This technology is considered to be on the flat-portion of the learning curve. The agencies have applied a medium complexity ICM of 1.39 to this technology through 2018 and 1.29 thereafter. The resultant costs are shown in Table 3-27. Table 3-27 Costs for WT-Dual Cam Phasing (2010\$) Cost type DMC DMC DMC 1C 1C 1C TC TC TC Engine type OHC-I4 OHC-V6/V8 OHV-V6/V8 OHC-I4 OHC-V6/V8 OHV-V6/V8 OHC-I4 OHC-V6/V8 OHV-V6/V8 2017 \$68 \$146 \$74 \$27 \$59 \$30 \$95 \$205 \$104 2018 \$66 \$143 \$72 \$27 \$59 \$30 \$94 \$202 \$102 2019 \$65 \$140 \$71 \$20 \$44 \$22 \$86 \$184 \$93 2020 \$64 \$137 \$70 \$20 \$44 \$22 \$84 \$181 \$92 2021 \$62 \$134 \$68 \$20 \$44 \$22 \$83 \$178 \$90 2022 \$61 \$132 \$67 \$20 \$44 \$22 \$82 \$176 \$89 2023 \$60 \$129 \$65 \$20 \$44 \$22 \$80 \$173 \$88 2024 \$59 \$127 \$64 \$20 \$44 \$22 \$79 \$170 \$86 2025 \$58 \$124 \$63 \$20 \$44 \$22 \$78 \$168 \$85 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; OHC=overhead cam; OHV=overhead valve; all costs are incremental to the baseline. 3.4.1.5 Variable Valve Lift (VVL) Varying and controlling the amount of cylinder valve lift across and engine operating range provides a potential for further efficiency improvements. By optimizing the valve-lift profile for specific engine operating regions, the pumping losses can be reduced by reducing the amount of throttling required to produce the desired engine power output. By moving the throttling losses further downstream of the throttle valve, the heat transfer losses that occur from the throttling process are directed into the fresh charge-air mixture just prior to compression, delaying the onset of knock-limited combustion processes. Variable valve lift control can also be used to induce in-cylinder mixture motion, which improves fuel-air mixing and can result in improved thermodynamic efficiency. Variable valve lift control can also potentially reduce overall valvetrain friction. At the same time, such systems may also incur increased parasitic losses associated with their actuation mechanisms. A number of manufacturers have already implemented VVL into their fleets (Toyota, Honda, and BMW), but overall this technology is still available as an efficiency improving technology for most of the fleet. There are two major classifications of variable valve lift, described below: -Ibid. 3-85 ------- Technologies Considered in the Agencies' Analysis 3.4.1.5.1 Discrete Variable Valve Lift (DVVL) Discrete variable valve lift (DVVL) systems allow the selection between two or three discrete cam profiles by means of a hydraulically-actuated mechanical system. These cam profiles consist of a low and a high-lift lobe, and may include an inert or blank lobe to incorporate cylinder deactivation (in the case of a 3-step DVVL system). DVVL is normally applied together with VVT control. DWL is also known as Cam Profile Switching (CPS). DVVL is a mature technology with low technical risk. The effectiveness of DVVL has been estimated to range from 1 to 4 percent in addition to that realized by VVT systems. These values were based on the research supporting MYs 2012-16 final rule, confidential manufacturer data, and a research conducted by the Northeast States Center for a Clean Air Future (NESCCAF). Based on additional information contained in the 2011 Ricardo study, NHTSA and EPA have revised the effectiveness range of DVVL systems to 2.8 to 3.9 percent above that realized by VVT systems. In the MYs 2012-2016 rule and the 2010 TAR, the agencies estimated the DMC of DVVL at \$116 (2007\$), \$169 (2007\$) and \$241 (2007\$) for an 14, V6 and V8 engine, respectively. Adjusted for 2010\$, these DMCs become \$122 (2010\$), \$177 (2010\$) and \$253 (2010\$) for this analysis all of which are considered applicable in MY 2015. This technology is considered to be on the flat-portion of the learning curve and is applicable only to engines with overhead cam configurations. The agencies have applied a medium complexity ICM of 1.39 to this technology through 2018 and 1.29 thereafter. The resultant costs are shown in Table 3-28. Table 3-28 Costs for Discrete Variable Valve Lift - DWL (2010\$) Cost type DMC DMC DMC 1C 1C 1C TC TC TC Engine type OHC-I4 OHC-V6 OHC-V8 OHC-I4 OHC-V6 OHC-V8 OHC-I4 OHC-V6 OHC-V8 2017 \$116 \$168 \$240 \$47 \$68 \$97 \$163 \$236 \$338 2018 \$114 \$165 \$235 \$47 \$68 \$97 \$161 \$233 \$333 2019 \$111 \$161 \$231 \$35 \$51 \$73 \$146 \$212 \$303 2020 \$109 \$158 \$226 \$35 \$51 \$72 \$144 \$209 \$298 2021 \$107 \$155 \$222 \$35 \$51 \$72 \$142 \$206 \$294 2022 \$105 \$152 \$217 \$35 \$50 \$72 \$140 \$202 \$289 2023 \$103 \$149 \$213 \$35 \$50 \$72 \$137 \$199 \$285 2024 \$101 \$146 \$209 \$35 \$50 \$72 \$135 \$196 \$280 2025 \$99 \$143 \$204 \$35 \$50 \$72 \$133 \$193 \$276 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; OHC=overhead cam; OHV=overhead valve; all costs are incremental to the baseline. 3.4.1.5.2 Continuously Variable Valve Lift (CVVL) In CVVL systems, valve lift is varied by means of a mechanical linkage, driven by an actuator controlled by the engine control unit. The valve opening and phasing vary as the lift is changed and the relation depends on the geometry of the mechanical system. BMW has considerable production experience with CVVL systems and has sold port-injected "Valvetronic" engines since 2001. Fiat is now offering "MultiAir" engines enabling precise control over intake valve lift. CVVL allows the airflow into the engine to be regulated by 3-86 ------- Technologies Considered in the Agencies' Analysis means of intake valve opening reduction, which improves engine efficiency by reducing pumping losses from throttling the intake system further upstream as with a conventionally throttled engine. Variable valve lift gives a further reduction in pumping losses compared to that which can be obtained with cam phase control only, with CVVL providing greater effectiveness than DVVL, since it can be fully optimized for all engine speeds and loads, and is not limited to a two or three step compromise. There may also be a small reduction in valvetrain friction when operating at low valve lift, resulting in improved low load fuel consumption for cam phase control with variable valve lift as compared to cam phase control only. Most of the fuel economy effectiveness is achieved with variable valve lift on the intake valves only. CVVL is only applicable to double overhead cam (DOHC) engines. The MYs 2012-2016 final rule estimated the effectiveness for CVVL at 1.5 to 3.5 percent over an engine with DCP, but also recognized that it could go up as high as 5 percent above and beyond DCP to account for the implementation of more complex CVVL systems such as BMW's "Valvetronic" and Fiat "MultiAir" systems. Thus, the effectiveness range for CVVL in this Joint TSD ranges from 1.5 to 7 percent depending on the complexity level of the application . For this rulemaking, NHTSA has increased the incremental effectiveness values for this technology to a range of 3.6 to 4.9 percent from 1.5 to 3.5 percent in the MYs 2012-2016 final rule. In the 2012-2016 rule and the 2010 TAR, the agencies estimated the DMC of CVVL at \$174 (2007\$), \$320 (2007\$), \$349 (2007\$), \$866 (2007\$) and \$947 (2007\$) for an OHC- 14, OHC-V6, OHC-V8, OHV-V6 and OHV-V8 engine, respectively. Adjusted for 2010\$, these DMCs become \$183 (2010\$), \$335 (2010\$), \$366 (2010\$), \$893 (2010\$) and \$977 (2010\$) for this analysis all of which are considered applicable in MY 2015. As indicated in this section, CVVL is considered only applicable to DOHC engine designs. The DMCs for OHV engines are meant to reflect additional costs associated with moving to a DOHC engine design. This technology is considered to be on the flat-portion of the learning curve. The agencies have applied a medium complexity ICM of 1.39 to this technology through 2018 and 1.29 thereafter. The resultant costs are shown in Table 3-29. Table 3-29 Costs for Continuous Variable Valve Lift - CWL (2010\$) Cost type DMC DMC DMC DMC DMC 1C 1C 1C Engine type OHC-I4 OHC-V6 OHC-V8 OHV-V6 OHV-V8 OHC-I4 OHC-V6 OHC-V8 2017 \$174 \$319 \$348 \$857 \$937 \$70 \$129 \$141 2018 \$170 \$313 \$341 \$840 \$919 \$70 \$129 \$141 2019 \$167 \$306 \$334 \$823 \$901 \$53 \$96 \$105 2020 \$164 \$300 \$327 \$807 \$883 \$52 \$96 \$105 2021 \$160 \$294 \$321 \$791 \$865 \$52 \$96 \$105 2022 \$157 \$288 \$314 \$775 \$847 \$52 \$96 \$104 2023 \$154 \$283 \$308 \$760 \$830 \$52 \$96 \$104 2024 \$151 \$277 \$302 \$744 \$814 \$52 \$95 \$104 2025 \$148 \$271 \$296 \$729 \$798 \$52 \$95 \$104 3-87 ------- Technologies Considered in the Agencies' Analysis 1C 1C TC TC TC TC TC OHV-V6 OHV-V8 OHC-I4 OHC-V6 OHC-V8 OHV-V6 OHV-V8 \$347 \$380 \$244 \$448 \$489 \$1,205 \$1,317 \$346 \$379 \$241 \$441 \$482 \$1,187 \$1,298 \$259 \$283 \$220 \$403 \$439 \$1,083 \$1,184 \$259 \$283 \$216 \$396 \$432 \$1,066 \$1,166 \$258 \$282 \$213 \$390 \$426 \$1,048 \$1,147 \$258 \$282 \$209 \$384 \$419 \$1,032 \$1,129 \$257 \$281 \$206 \$378 \$412 \$1,016 \$1,112 \$257 \$281 \$203 \$372 \$406 \$1,001 \$1,095 \$256 \$280 \$200 \$367 \$400 \$986 \$1,078 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; OHC=overhead cam; OHV=overhead valve; all costs are incremental to the baseline. 3.4.1.6 Variable Valve Actuation (VVA) For this final rule, consistent with the proposal, NHTSA has combined two valve control technologies for OHV engines. Coupled cam phasing (CCPO) and discrete valve lift (DVVLO) into one technology defined as variable valve actuation (VVA). The agency estimates the incremental effectiveness for WA applied to and OHV engine as 2.71 to 3.59 percent. This effectiveness value is slightly lower than coupled cam phasing for overhead cam applications (CCPS) based on the assumption that VVA would be applied to an OHV engine after cylinder deactivation (DEAC). For more information on combining these technologies please refer to NHTSA's FRIA. 3.4.1.7 Stoichiometric Gasoline Direct Injection (SGDI) Stoichiometric gasoline direct injection (SGDI), or Spark Ignition Direct injection (SIDI), engines inject fuel at high pressure directly into the combustion chamber (rather than the intake port in port fuel injection). SGDI requires changes to the injector design, an additional high pressure fuel pump, new fuel rails to handle the higher fuel pressures and changes to the cylinder head and piston crown design. Direct injection of the fuel into the cylinder improves cooling of the air/fuel charge within the cylinder, which allows for higher compression ratios and increased thermodynamic efficiency without the onset of combustion knock. Recent injector design advances, improved electronic engine management systems and the introduction of multiple injection events per cylinder firing cycle promote better mixing of the air and fuel, enhance combustion rates, increase residual exhaust gas tolerance and improve cold start emissions. SGDI engines achieve higher power density and match well with other technologies, such as boosting and variable valvetrain designs. Several manufacturers are manufacturing vehicles with SGDI engines, including VW/Audi, BMW, Toyota, Ford, and General Motors. Additionally, BMW, GM, Ford and VW/Audi have announced plans to significantly increase the number of SGDI engines in their portfolios. NHTSA and EPA reviewed estimates from the MYs 2012-2016 final rule and 2010 TAR, which stated an effectiveness range of SGDI to be between 2 and 3 percent. NHTSA and EPA reviewed estimates from the Alliance of Automobile Manufactures, which projects 3 percent gains in fuel efficiency and a 7 percent improvement in torque. The torque increase provides the opportunity to downsize the engine allowing an increase in efficiency of up to a 5.8 percent. NHTSA and EPA also reviewed other published literature, reporting 3 percent effectiveness for SGDI.41 Confidential manufacturer data reported an efficiency effectiveness 3-88 ------- Technologies Considered in the Agencies' Analysis range of 1 to 2 percent. Based on data from the 2011 Ricardo study and reconfiguration of the new lumped parameter model, EPA and NHTSA have revised this value to 1.5 percent1111. Combined with other technologies (i.e.., boosting, downsizing, and in some cases, cooled EGR), SGDI can achieve greater reductions in fuel consumption and CC>2 emissions compared to engines of similar power output. The NHTSA and EPA cost estimates for SGDI take into account the changes required to the engine hardware, engine electronic controls, ancillary and Noise Vibration and Harshness (NVH) mitigation systems. Through contacts with industry NVH suppliers, and manufacturer press releases, the agencies believe that the NVH treatments will be limited to the mitigation of fuel system noise, specifically from the injectors and the fuel lines and have included corresponding cost estimates for these NVH controls. In the 2012-2016 FRM, the agencies estimated the DMC for SGDI at \$213 (2007\$), \$321 (2007\$) and \$386 (2007\$) for 13/14, V6 and V8 engines, respectively. Adjusted for 2010\$, these DMCs become \$222 (2010\$), \$334 (2010\$) and \$402 (2010\$) for this analysis all of which are considered applicable in MY 2012. This technology is considered to be on the flat-portion of the learning curve. The agencies have applied a medium complexity ICM of 1.39 to this technology through 2018 and 1.29 thereafter. The resultant costs are shown in Table 3-30. Table 3-30 Costs for Stoichiometric Gasoline Direct Injection (2010\$) Cost type DMC DMC DMC 1C 1C 1C TC TC TC Engine type 13/14 V6 V8 13/14 V6 V8 13/14 V6 V8 2017 \$192 \$290 \$348 \$84 \$127 \$153 \$277 \$417 \$501 2018 \$188 \$284 \$341 \$84 \$127 \$153 \$273 \$411 \$494 2019 \$185 \$278 \$335 \$63 \$95 \$114 \$248 \$373 \$449 2020 \$181 \$273 \$328 \$63 \$95 \$114 \$244 \$367 \$442 2021 \$177 \$267 \$321 \$63 \$95 \$114 \$240 \$362 \$435 2022 \$174 \$262 \$315 \$63 \$94 \$114 \$236 \$356 \$429 2023 \$170 \$257 \$309 \$63 \$94 \$113 \$233 \$351 \$422 2024 \$167 \$251 \$302 \$62 \$94 \$113 \$229 \$346 \$416 2025 \$164 \$246 \$296 \$62 \$94 \$113 \$226 \$340 \$409 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. 3.4.1.8 Turbocharging and Downsizing (TRBDS) The specific power of a naturally aspirated engine is primarily limited by the rate at which the engine is able to draw air into the combustion chambers. Turbocharging and supercharging (grouped together here as boosting) are two methods to increase the intake manifold pressure and cylinder charge-air mass above naturally aspirated levels. Boosting increases the airflow into the engine, thus increasing the specific power level, and with it the ability to reduce engine displacement while maintaining performance. This effectively reduces the pumping losses at lighter loads in comparison to a larger, naturally aspirated engine. 14 However, because GDI is a key enabler for modern, highly downsized turbocharged engines, this difference will be overshadowed by the higher effectiveness for turbocharging and downsizing when they are combined into packages. 3-89 ------- Technologies Considered in the Agencies' Analysis Almost every major manufacturer currently markets a vehicle with some form of boosting. While boosting has been a common practice for increasing performance for several decades, turbocharging has considerable potential to improve fuel economy and reduce CO2 emissions when the engine displacement is also reduced. Specific power levels for a boosted engine often exceed 100 hp/L, compared to average naturally aspirated engine power densities of roughly 70 hp/L. As a result, engines can be downsized roughly 30 percent or higher while maintaining similar peak output levels. In the last decade, improvements to turbocharger turbine and compressor design have improved their reliability and performance across the entire engine operating range. New variable geometry turbines and ball-bearing center cartridges allow faster turbocharger spool-up (virtually eliminating the once-common "turbo lag") while maintaining high flow rates for increased boost at high engine speeds. Low speed torque output has been dramatically improved for modern turbocharged engines. However, even with turbocharger improvements, maximum engine torque at very low engine speed conditions, for example launch from standstill, is increased less than at mid and high engine speed conditions. The potential to downsize engines may be less on vehicles with low displacement to vehicle mass ratios for example a very small displacement engine in a vehicle with significant curb weight, in order to provide adequate acceleration from standstill, particularly up grades or at high altitudes. Use of GDI systems with turbocharged engines and air-to-air charge air cooling also reduces the fuel octane requirements for knock limited combustion and allows the use of higher compression ratios. Ford's "Ecoboost" downsized, turbocharged GDI engines introduced on MY 2010 vehicles allow the replacement of V8 engines with V6 engines with improved in 0-60 mph acceleration and with fuel economy improvements of up to 12 percent.42 Recently published data with advanced spray-guided injection systems and more aggressive engine downsizing targeted towards reduced fuel consumption and CC>2 emissions reductions indicate that the potential for reducing CO2 emissions for turbocharged, downsized GDI engines may be as much as 15 to 30 percent relative to port-fuel-injected 97 9R 9Q ^0 ^ 1 engines. ' ' ' ' Confidential manufacturer data suggests an incremental range of fuel consumption and CC>2 emission reduction of 4.8 to 7.5 percent for turbocharging and downsizing. Other publicly-available sources suggest a fuel consumption and CO2 emission reduction of 8 to 13 percent compared to current-production naturally-aspirated engines without friction reduction or other fuel economy technologies: a joint technical paper by Bosch and Ricardo suggesting fuel economy gain of 8 to 10 percent for downsizing from a 5.7 liter port injection V8 to a 3.6 liter V6 with direct injection using a wall-guided direct injection system;43 a Renault report suggesting a 11.9 percent NEDC fuel consumption gain for downsizing from a 1.4 liter port injection in-line 4-cylinder engine to a 1.0 liter in-line 4- cylinder engine, also with wall-guided direct injection;44 and a Robert Bosch paper suggesting a 13 percent NEDC gain for downsizing to a turbocharged DI engine, again with wall-guided injection.45 These reported fuel economy benefits show a wide range depending on the SGDI technology employed. NHTSA and EPA reviewed estimates from the 2012-2016 final rule, the TAR, and existing public literature. The previous estimate from the MYs 2012-2016 suggested a 12 to 14 percent effectiveness improvement, which included low friction lubricant (level one), 3-90 ------- Technologies Considered in the Agencies' Analysis engine friction reduction (level one), DCP, DVVL and SGDI, over baseline fixed-valve engines, similar to the estimate for Ford's Ecoboost engine, which is already in production. Additionally, the agencies analyzed Ricardo vehicle simulation data for various turbocharged engine packages. Based on this data, and considering the widespread nature of the public estimates, the effectiveness of turbocharging and downsizing is highly dependent upon implementation and degree of downsizing. In alignment with these variances, for this final rule, consistent with the proposal, the agencies evaluated 4 different levels of downsized and turbocharged high Brake Mean Effective Pressure (BMEP)11. engines; 18-bar, 24-bar, 24-bar with cooled exhaust gas recirculation (EGR) and 27'-bar with cooled EGR All engines are assumed to include gasoline direct injection (SGDI) and effectiveness values include the benefits of this technology. In addition, the agencies believe to implement in production a 27 bar boost level, it is necessary to incorporate cooled exhaust gas recirculation (EGR) and also require a 2- stage turbocharger as well as engine changes to increase robustness. The cooled EGR technology is discussed later in this section. NHTSA and EPA have revised the effectiveness to reflect this new information and assume that turbocharging and downsizing, alone, will provide a 12 to 24.6 percent effectiveness improvement (dependent upon degree of downsizing and boost levels) over naturally aspirated, fixed-valve engines. More specifically, 12.1 to 14.9 percent for 18-bar engines, which is equal to the boost levels evaluated in the MYs 2012-2016 final rule, assuming 33 percent downsizing, 16.4 to 20.1 percent for 24-bar engines, assuming 50 percent downsizing, 19.3 to 23.0 percent for 24-bar engines with cooled EGR, assuming 50 percent downsizing and 20.6 to 24.6 percent for 27-bar engines with cooled EGR, assuming 56 percent downsizing. For comparison purposes an 18-bar engine with low friction lubricant (level one), engine friction reduction (level one), DCP, DVVL and SGDI, which is equivalent to MYs 2012-2016 assumed turbocharging and downsizing technology, now results in a 16.8 to 20.9 percent effectiveness improvement. Coupling turbocharging and downsizing with low friction lubricant (level one and two), engine friction reductions (level one and two), DCP, DVVL and SGDI, for the MYs 2017-2025 timeframe, yields 18.0 to 22.4 percent for 18-bar engines 20.4 to 25.2 percent for 24-bar engines, 23.2 to 27.9 percent for 24-bar engine with cooled EGR and 24.0 to 28.8 percent for 27-bar with cooled EGR over naturally aspirated, fixed-valve engines. As noted above, the agencies relied on engine teardown analyses conducted by EPA, FEV and Munro to develop costs for turbocharged GDI engines.46 In the 2012-2016 FRM, the agencies estimated the DMC for turbocharging to 18 bar BMEP at \$404 (2007\$) and \$681 (2007\$) for 14 and V6/V8 engines, respectively, where the higher cost for the V-configuration 11 Brake Mean Effective Pressure is the average amount of pressure in pounds per square inch (psi) that must be exerted on the piston to create the measured horsepower. This indicates how effective an engine is at filling the combustion chamber with an air/fuel mixture, compressing it and achieving the most power from it. A higher BMEP value contributes to higher overall efficiency. 3-91 ------- Technologies Considered in the Agencies' Analysis engines represents twin turbochargers versus the single turbocharger in the I-configuration engine. These DMCs become \$420 (2010\$) and \$708 (2010\$), respectively, for this analysis. In the 2010 TAR, the agencies presented costs for 24 bar BMEP turbocharging at 1.5x the cost of the 18 bar BMEP technology. This additional cost covered the incremental cost increase of a variable geometry turbocharger (see 2010 TAR at page B-12). Thus, the DMC for 24 bar BMEP would be \$630 (2010\$) and \$1,062 (2010\$) for I-configuration and V- configuration engines, respectively. Note also for this final rule, the agencies are estimating the DMC of the 27 bar BMEP technology at 2.5x the 18 bar BMEP technology, or \$1,050 (2010\$) and \$1,771 (2010\$) for I-configuration and V-configuration engines, respectively. All of these turbocharger-related DMCs are considered applicable in the 2012MY. The agencies consider each turbocharger technology to be on the flat portion of the learning curve and have applied a medium complexity ICM of 1.39 through 2018 for 18 bar and through 2024 for 24 and 27 bar, then 1.29 to each thereafter. The resultant costs are shown in Table 3-31. Table 3-31 Costs for Turbocharging (2010\$) Cost type DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC Technology (BMEP) 18 bar 18 bar 24 bar 24 bar 27 bar 27 bar 18 bar 18 bar 24 bar 24 bar 27 bar 27 bar 18 bar 18 bar 24 bar 24 bar 27 bar 27 bar Engine type I-engine V-engine I-engine V-engine I-engine V-engine I-engine V-engine I-engine V-engine I-engine V-engine I-engine V-engine I-engine V-engine I-engine V-engine 2017 \$365 \$614 \$547 \$922 \$911 \$1,536 \$160 \$270 \$240 \$405 \$401 \$675 \$525 \$885 \$787 \$1,327 \$1,312 \$2,211 2018 \$357 \$602 \$536 \$903 \$893 \$1,505 \$160 \$270 \$240 \$404 \$400 \$674 \$517 \$872 \$776 \$1,308 \$1,293 \$2,179 2019 \$350 \$590 \$525 \$885 \$875 \$1,475 \$120 \$202 \$239 \$403 \$399 \$672 \$470 \$792 \$765 \$1,289 \$1,274 \$2,148 2020 \$343 \$578 \$515 \$867 \$858 \$1,446 \$119 \$201 \$239 \$403 \$398 \$671 \$462 \$779 \$754 \$1,270 \$1,256 \$2,117 2021 \$336 \$567 \$504 \$850 \$841 \$1,417 \$119 \$201 \$238 \$402 \$397 \$670 \$455 \$768 \$743 \$1,252 \$1,238 \$2,087 2022 \$330 \$555 \$494 \$833 \$824 \$1,389 \$119 \$200 \$238 \$401 \$397 \$669 \$448 \$756 \$732 \$1,234 \$1,220 \$2,057 2023 \$323 \$544 \$484 \$816 \$807 \$1,361 \$119 \$200 \$238 \$400 \$396 \$667 \$442 \$744 \$722 \$1,217 \$1,203 \$2,028 2024 \$316 \$533 \$475 \$800 \$791 \$1,334 \$118 \$200 \$237 \$400 \$395 \$666 \$435 \$733 \$712 \$1,200 \$1,186 \$2,000 2025 \$310 \$523 \$465 \$784 \$775 \$1,307 \$118 \$199 \$177 \$299 \$296 \$499 \$428 \$722 \$643 \$1,083 \$1,071 \$1,805 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. The costs for the downsizing portion of the turbo/downsize technology is more complex. The agencies have described those cost and how they were developed—based primarily on FEV teardowns but some were scaled based on teardowns to generate costs for downsizing situations that were not covered by teardowns—in both the 2012-2016 FRM and the 2010 TAR. The DMCs used for this analysis are identical to those used in the 2010 TAR except that they have been updated to 2010 dollars. Notable is the fact that many of the downsizing costs are negative because they result in fewer parts and less material than the engine from which they are "derived." For example a V8 engine could be replaced by a turbocharged V6 engine having two fewer cylinders and as many as eight fewer valves (in the case of a V8 DOHC downsized to a V6 DOHC). Importantly, the agencies have used an approach to calculating indirect costs that results in positive indirect costs regardless of whether the DMC is positive or negative. This is done by calculating indirect costs based on the absolute value of the DMC, then adding the indirect cost to the DMC to arrive at the total 3-92 ------- Technologies Considered in the Agencies' Analysis cost. This way, the agencies are never making a negative DMC "more negative" when accounting for the indirect costs. This approach has been used in the 2012-2016 final rule and the 2010 TAR. Given the history of the downsizing costs used by the agencies, many are considered applicable in the 2012MY and many in the 2017MY.JJ All are considered to be on the flat portion of the learning curve. The agencies have applied a medium complexity ICM of 1.39 through 2018 and 1.29 thereafter. The resultant costs are shown in Table 3-32. Table 3-32 Costs for Engine Downsizing (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC Technology 14 DOHC to 13 14 DOHC to 14 V6 DOHC to 14 V6SOHC2Vto 14 V6OHVtoI4 V8 DOHC to 14 V8DOHCtoV6 V8SOHC2Vto 14 V8SOHC3Vto 14 V8SOHC2Vto V6 V8SOHC3Vto V6 V8OHVtoI4 V8OHVtoV6 14 DOHC to 13 14 DOHC to 14 V6 DOHC to 14 V6SOHC2Vto 14 V6OHVtoI4 V8 DOHC to 14 V8DOHCtoV6 V8SOHC2Vto 14 V8SOHC3Vto 14 V8SOHC2Vto V6 V8SOHC3Vto V6 V8OHVtoI4 V8OHVtoV6 14 DOHC to 13 2017 -\$174 -\$77 -\$494 -\$345 \$281 -\$854 -\$247 -\$656 -\$731 -\$76 -\$140 -\$242 \$328 \$77 \$34 \$217 \$152 \$109 \$331 \$109 \$254 \$283 \$33 \$62 \$94 \$127 -\$98 2018 -\$171 -\$75 -\$484 -\$338 \$272 -\$828 -\$242 -\$637 -\$709 -\$74 -\$137 -\$234 \$318 \$76 \$34 \$217 \$151 \$108 \$330 \$108 \$253 \$282 \$33 \$61 \$93 \$126 -\$94 2019 -\$167 -\$74 -\$474 -\$331 \$264 -\$804 -\$237 -\$617 -\$687 -\$73 -\$135 -\$227 \$308 \$57 \$25 \$162 \$113 \$81 \$246 \$81 \$189 \$210 \$25 \$46 \$70 \$94 -\$110 2020 -\$164 -\$72 -\$465 -\$325 \$256 -\$779 -\$233 -\$599 -\$667 -\$71 -\$132 -\$220 \$299 \$57 \$25 \$162 \$113 \$81 \$245 \$81 \$188 \$210 \$25 \$46 \$69 \$94 -\$107 2021 -\$161 -\$71 -\$455 -\$318 \$249 -\$756 -\$228 -\$581 -\$647 -\$70 -\$129 -\$214 \$290 \$57 \$25 \$161 \$113 \$80 \$244 \$81 \$188 \$209 \$25 \$46 \$69 \$94 -\$104 2022 -\$157 -\$69 -\$446 -\$312 \$241 -\$733 -\$223 -\$564 -\$627 -\$68 -\$127 -\$207 \$281 \$57 \$25 \$161 \$113 \$80 \$244 \$81 \$187 \$208 \$25 \$46 \$69 \$93 -\$101 2023 -\$154 -\$68 -\$437 -\$306 \$236 -\$719 -\$219 -\$552 -\$615 -\$67 -\$124 -\$203 \$276 \$57 \$25 \$161 \$112 \$80 \$243 \$80 \$187 \$208 \$25 \$46 \$69 \$93 -\$98 2024 -\$151 -\$67 -\$429 -\$300 \$232 -\$704 -\$215 -\$541 -\$603 -\$66 -\$122 -\$199 \$270 \$57 \$25 \$161 \$112 \$80 \$243 \$80 \$187 \$208 \$25 \$46 \$69 \$93 -\$95 2025 -\$148 -\$65 -\$420 -\$294 \$227 -\$690 -\$210 -\$530 -\$591 -\$64 -\$119 -\$195 \$265 \$57 \$25 \$160 \$112 \$80 \$242 \$80 \$186 \$207 \$25 \$45 \$69 \$93 -\$92 JJ The engine downsize costs based on actual FEV teardowns were considered applicable to the 2012MY, as was explained for some downsize costs in the 2012-2016 final rule and others in the 2010 TAR. For other downsize costs—the two changes from OHV engines to DOHC engines—the agencies did not use FEV teardowns or extrapolations from FEV teardowns, and instead used the methodology employed in the 2008 EPA Staff Report, a methodology determined by both agencies to result in cost estimates more appropriate for the 2017MY. The new downsize costs—those for V8 engines downsized to 14 engines—use a combination of V8 to V6 then V6 to 14 downsize costs and are considered applicable to the 2017MY within the context of this analysis. 3-93 ------- Technologies Considered in the Agencies' Analysis TC TC TC TC TC TC TC TC TC TC TC TC 14 DOHC to 14 V6DOHCtoI4 V6SOHC2Vto 14 V6OHVtoI4 V8 DOHC to 14 V8DOHCtoV6 V8SOHC2Vto 14 V8SOHC3Vto 14 V8SOHC2Vto V6 V8SOHC3Vto V6 V8OHVtoI4 V8OHVtoV6 -\$43 -\$277 -\$193 \$390 -\$523 -\$139 -\$402 -\$448 -\$42 -\$79 -\$148 \$454 -\$41 -\$267 -\$187 \$381 -\$499 -\$134 -\$383 -\$427 -\$41 -\$76 -\$141 \$444 -\$48 -\$312 -\$218 \$345 -\$558 -\$156 -\$429 -\$477 -\$48 -\$89 -\$158 \$403 -\$47 -\$303 -\$212 \$337 -\$534 -\$152 -\$411 -\$457 -\$46 -\$86 -\$151 \$393 -\$46 -\$294 -\$205 \$329 -\$512 -\$147 -\$393 -\$438 -\$45 -\$83 -\$145 \$384 -\$44 -\$285 -\$199 \$321 -\$490 -\$143 -\$376 -\$419 -\$44 -\$81 -\$139 \$375 -\$43 -\$277 -\$193 \$316 -\$476 -\$138 -\$365 -\$407 -\$42 -\$78 -\$134 \$369 -\$42 -\$268 -\$187 \$311 -\$462 -\$134 -\$355 -\$395 -\$41 -\$76 -\$131 \$363 -\$40 -\$260 -\$182 \$307 -\$448 -\$130 -\$344 -\$383 -\$40 -\$74 -\$127 \$358 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline; all resultant engines are DOHC. Note that the V8 to 14 engine downsize is new for this final rule, consistent with the proposal. This level of engine downsizing is considered for this analysis only if it also includes 27 bar BMEP turbo boost which, in addition, requires the addition of cooled EGR (discussed below). As a result, any 27 bar BMEP engine in this analysis will be 14 configuration and will include cooled EGR. With the information shown in Table 3-31 and Table 3-32, the costs for any turbo/downsize change can be determined. These costs are shown in Table 3-33. Table 3-33 Total Costs for Turbo/Downsizing (2010\$) Downsize Technology 14 DOHC to 13 14 DOHC to 13 14 DOHC to 13 14 DOHC to 14 14 DOHC to 14 14 DOHC to 14 V6 DOHC to 14 V6 DOHC to 14 V6 DOHC to 14 V6 SOHC 2V to 14 V6 SOHC 2V to 14 V6 SOHC 2V to 14 V6 OHV to 14 V6 OHV to 14 V6 OHV to 14 V8 DOHC to 14 V8 DOHC to 14 V8 DOHC to 14 V8 DOHC to V6 V8 DOHC to V6 V8 DOHC to V6 Turbo Technology (BMEP) 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar 2017 \$427 \$690 \$1,214 \$482 \$744 \$1,269 \$248 \$510 \$1,035 \$331 \$594 \$1,119 \$914 \$1,177 \$1,701 \$1 \$264 \$789 \$746 \$1,188 \$2,073 2018 \$423 \$681 \$1,199 \$476 \$734 \$1,251 \$250 \$508 \$1,026 \$330 \$589 \$1,106 \$898 \$1,156 \$1,674 \$18 \$277 \$794 \$738 \$1,174 \$2,045 2019 \$359 \$654 \$1,164 \$421 \$716 \$1,226 \$157 \$452 \$962 \$251 \$546 \$1,056 \$815 \$1,110 \$1,619 -\$88 \$207 \$716 \$635 \$1,132 \$1,991 2020 \$356 \$647 \$1,149 \$415 \$707 \$1,209 \$159 \$450 \$953 \$251 \$542 \$1,044 \$799 \$1,090 \$1,593 -\$72 \$219 \$722 \$628 \$1,118 \$1,965 2021 \$352 \$639 \$1,134 \$410 \$697 \$1,192 \$161 \$449 \$944 \$250 \$537 \$1,032 \$784 \$1,072 \$1,567 -\$56 \$231 \$726 \$620 \$1,105 \$1,940 2022 \$348 \$632 \$1,120 \$404 \$688 \$1,176 \$163 \$447 \$935 \$249 \$533 \$1,021 \$770 \$1,053 \$1,542 -\$41 \$243 \$731 \$613 \$1,092 \$1,914 2023 \$344 \$624 \$1,106 \$399 \$679 \$1,160 \$165 \$445 \$927 \$248 \$529 \$1,010 \$758 \$1,038 \$1,519 -\$34 \$246 \$728 \$606 \$1,078 \$1,890 2024 \$340 \$617 \$1,092 \$393 \$670 \$1,145 \$167 \$444 \$918 \$248 \$524 \$999 \$746 \$1,023 \$1,498 -\$27 \$250 \$725 \$599 \$1,066 \$1,866 2025 \$337 \$551 \$979 \$388 \$602 \$1,031 \$169 \$383 \$811 \$247 \$461 \$890 \$735 \$949 \$1,378 -\$19 \$195 \$623 \$592 \$953 \$1,675 3-94 ------- Technologies Considered in the Agencies' Analysis V8 SOHC 2V to 14 V8 SOHC 2V to 14 V8 SOHC 2V to 14 V8 SOHC 3V to 14 V8 SOHC 3V to 14 V8 SOHC 3V to 14 V8 SOHC 2V to V6 V8 SOHC 2V to V6 V8 SOHC 2V to V6 V8 SOHC 3V to V6 V8 SOHC 3V to V6 V8 SOHC 3V to V6 V8 OHV to 14 V8 OHV to 14 V8 OHV to 14 V8 OHV to V6 V8 OHV to V6 V8 OHV to V6 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar 18 bar 24 bar 27 bar \$123 \$385 \$910 \$77 \$339 \$864 \$842 \$1,284 \$2,169 \$806 \$1,248 \$2,133 \$377 \$639 \$1,164 \$1,339 \$1,781 \$2,666 \$134 \$392 \$910 \$90 \$349 \$866 \$831 \$1,267 \$2,138 \$796 \$1,232 \$2,103 \$376 \$635 \$1,152 \$1,316 \$1,752 \$2,623 \$41 \$336 \$846 -\$8 \$287 \$797 \$744 \$1,241 \$2,100 \$703 \$1,200 \$2,059 \$312 \$607 \$1,116 \$1,194 \$1,691 \$2,550 \$52 \$343 \$845 \$5 \$296 \$799 \$733 \$1,224 \$2,071 \$693 \$1,184 \$2,031 \$311 \$602 \$1,105 \$1,172 \$1,663 \$2,510 \$62 \$350 \$845 \$18 \$305 \$800 \$723 \$1,207 \$2,042 \$684 \$1,169 \$2,003 \$311 \$598 \$1,093 \$1,151 \$1,636 \$2,471 \$72 \$356 \$844 \$29 \$313 \$801 \$712 \$1,191 \$2,014 \$675 \$1,153 \$1,976 \$310 \$594 \$1,082 \$1,131 \$1,609 \$2,432 \$76 \$357 \$838 \$35 \$315 \$796 \$702 \$1,175 \$1,986 \$666 \$1,138 \$1,950 \$307 \$587 \$1,069 \$1,113 \$1,586 \$2,397 \$80 \$357 \$832 \$40 \$317 \$791 \$692 \$1,159 \$1,959 \$657 \$1,124 \$1,924 \$304 \$581 \$1,056 \$1,096 \$1,563 \$2,363 \$84 \$298 \$727 \$45 \$259 \$688 \$682 \$1,043 \$1,766 \$648 \$1,010 \$1,732 \$302 \$516 \$944 \$1,080 \$1,441 \$2,163 All costs are total costs (Direct manufacturing costs + Indirect costs); all costs are incremental to the baseline; all resultant engines are DOHC; note that costs are shown for 27 bar BMEP engines with V6 engines. In fact, the agencies do not believe that manufacturers will employ 27 bar BMEP technology on V6 engines to comply with the final standards, instead using the additional boost to allow for downsizing V6 engines to smaller 14 engines than would be used for 18 bar BMEP or 24 bar BMEP 14 engines and/or downsizing V8 engines to 14 engines. As a result, whenever a 27 bar BMEP engine is chosen by either agency's model, the engine configuration will be an 14 and will include cooled EGR, as discussed in section 3.4.1.8. 3.4.1.9 Cooled Exhaust-Gas Recirculation (EGR) While not considered in the technology packages used for assessing potential compliance pathways in the 2012-2016 light-duty rule, the agencies have considered an emerging technology referred to as cooled exhaust gas recirculation (cooled-EGR) as applied to downsized, turbocharged GDI engines. In the 2010 TAR, the agencies considered this technology as an advanced gasoline technology since it was considered an emerging and not yet available technology in the light-duty gasoline market. While a cooled or "boosted" EGR technology was discussed in the 2012-2016 light-duty rule record, the technology considered here is comparatively more advanced as described in the 2010 TAR. As such, the agencies have considered new costs and new effectiveness values for it. The effectiveness values used for vehicle packages with cooled EGR within this analysis reflect a conservative estimate of system performance at approximately 24-bar BMEP. Vehicle simulation modeling of technology packages using the more highly boosted and downsized cooled EGR engines (up 3-95 ------- Technologies Considered in the Agencies' Analysis to 27-bar BMEP, and utilizing EGR rates of 20-25%) with dual-stage turbocharging has been completed as part of EPA's contract with Ricardo Engineering as described in 3.3.1.2. For this FRM, consistent with the proposal, the agencies have updated the effectiveness of vehicle packages with cooled EGR using the new Ricardo vehicle simulation modeling runs. Cooled exhaust gas recirculation or Boosted EGR is a combustion concept that involves utilizing EGR as a charge dilutent for controlling combustion temperatures and cooling the EGR prior to its introduction to the combustion system. Higher exhaust gas residual levels at part load conditions reduce pumping losses for increased fuel economy. The additional charge dilution enabled by cooled EGR reduces the incidence of knocking combustion and obviates the need for fuel enrichment at high engine power. This allows for higher boost pressure and/or compression ratio and further reduction in engine displacement and both pumping and friction losses while maintaining performance. Engines of this type use GDI and both dual cam phasing and discrete variable valve lift. The EGR systems considered in this final rule, consistent with the proposal, would use a dual-loop system with both high and low pressure EGR loops and dual EGR coolers. The engines would also use single-stage, variable geometry turbocharging with higher intake boost pressure available across a broader range of engine operation than conventional turbocharged SI engines. Such a system is estimated to be capable of an additional 3 to 5 percent effectiveness relative to a turbocharged, downsized GDI engine without cooled-EGR.47'48 The agencies have also considered a more advanced version of such a cooled EGR system that employs very high combustion pressures by using dual stage turbocharging. This modeling work has been completed by Ricardo Engineering. The simulation modeling is similar to work that Ricardo conducted for EPA for its 2008 staff report on GHG effectiveness of light-duty vehicle technologies.49 The agencies have considered this more advanced cooled EGR approach for this final rule, consistent with the proposal. For the MYs 2012-2016 final rule and TAR, NHTSA and EPA assumed a 5 percent fuel consumption effectiveness for cooled EGR compared to a conventional downsized DI turbocharged engine.50 Based on the data from the Ricardo and Lotus reports, NHTSA and EPA estimate the incremental reduction in fuel consumption for EGR Boost to be 5 percent over a turbocharged and downsized DI engine. Thus, if cooled EGR is applied to 24-bar engine, adding the 19.3 percent from the turbocharging and downsizing to the 5 percent gain from cooled EGR results in total fuel consumption reduction of 22.1 percent. This is in agreement with the range suggested in the Lotus and Ricardo reports. In the 2010 TAR, the agencies estimated the DMC of the cooled EGR system at \$240 (2007\$, see 2010 TAR at page B-12)). This DMC becomes \$244 (2010\$) for this analysis. This DMC is considered applicable in the 2012MY. The agencies consider cooled EGR technology to be on the flat portion of the learning curve and have applied a medium complexity ICM of 1.39 through 2024 then 1.29 thereafter. The resultant costs are shown in Table 3-34. Table 3-34 Costs for Cooled EGR (2010\$) Cost type Engine type 2017 2018 2019 2020 2021 2022 2023 2024 2025 3-96 ------- Technologies Considered in the Agencies' Analysis DMC 1C TC All All All \$212 \$93 \$305 \$208 \$93 \$301 \$204 \$93 \$296 \$199 \$93 \$292 \$195 \$92 \$288 \$192 \$92 \$284 \$188 \$92 \$280 \$184 \$92 \$276 \$180 \$69 \$249 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. Note that, in the 2010 TAR, the agencies presented the cooled EGR system costs inclusive of turbo charging costs (see 2010 TAR, Table B2.2-1 at page B-12). For this analysis, the agencies are presenting the cooled EGR costs as a stand-alone technology that can be added to any turbo/downsized engine provided sufficient boost is provided and sufficient engine robustness is accounted for. As such, the cooled EGR system is considered applicable only the 24 bar BMEP and 27 bar BMEP engines. Further, the agencies believe that 24 bar BMEP engines are capable of maintaining NOx control without cooled EGR, so each agency's respective models may choose 24 bar BMEP engines with and/or without cooled EGR. However, as noted above, 27 bar BMEP engines are considered to require cooled EGR to maintain NOx emission control. As such, neither agency's model is allowed to choose 27 bar BMEP technology without also adding cooled EGR. 3.4.1.10 Diesel Engine Technology (DSL) Diesel engines have several characteristics that give them superior fuel efficiency compared to conventional gasoline, spark-ignited engines. Pumping losses are much lower due to lack of (or greatly reduced) throttling in a diesel engine. The diesel combustion cycle operates at a higher compression ratio than does a gasoline engine. As a result, turbocharged light-duty diesels typically achieve much higher torque levels at lower engine speeds than equivalent-displacement naturally-aspirated gasoline engines. Future high BMEP turbocharged and downsized engines, mentioned above, are projected to improve torque levels at lower engine speeds thus reducing the diesel advantage in this area. Diesels also operate with a very lean air/fuel mixture. These attributes - reduced pumping losses, higher compression ratio and lean/air fuel mixture — allow the engine to extract more energy from a given mass of fuel than a gasoline engine, and thus make it more efficient. Additionally, diesel fuel has higher energy content per gallon than does gasoline. While diesel fuel has a higher energy content than gasoline, it also contains more carbon per gallon than does gasoline: diesel produces 22.2 pounds of CC>2 per gallon when burned, while gasoline produces 19.4 pounds of CO2 per gallon. This higher carbon content slightly offsets the GHG emissions benefit of diesel fuel relative to gasoline, however, the disbenefit is more than compensated by the greater efficiency of the diesel engine. Since diesel engines are more fuel efficient than current naturally aspirated PFI gasoline engines, the agencies anticipate that manufacturers will evaluate and potentially invest in diesel engine production as a way to comply with more stringent CAFE standards. However, there are two primary reasons why manufacturers might not choose to invest significantly in diesel engine technologies as a way to comply with the CAFE and GHG standards for MYs 2017-2025. As discussed above, even though diesel has higher energy content than gasoline it also has a higher carbon density that results in higher amounts of CC>2 emitted per gallon, approximately 15 percent more than a gallon of gasoline. This is commonly referred to as the "carbon penalty" associated with using diesel fuel - a diesel vehicle yields greater fuel 3-97 ------- Technologies Considered in the Agencies' Analysis economy improvements compared to its CC>2 emissions reduction improvements, so a manufacturer that invests in diesel technology to meet CAFE standards may have more trouble meeting the GHG standards than if it used a different and more cost effective (from a GHG perspective) technology. And second, diesel engines also have emissions characteristics that present challenges to meeting federal Tier 2 NOx emissions standards. By way of comparison for readers familiar with the European on-road fleet, which contains many more diesel vehicles than the U.S. on-road fleet, U.S. Tier 2 emissions fleet average requirement of bin 5 require roughly 45 to 65 percent more NOX reduction compared to the Euro VI standards. Despite considerable advances by manufacturers in developing Tier 2-compliant diesel engines, it remains somewhat of a systems-engineering challenge to maintain the full fuel consumption advantage of the diesel engine while meeting Tier 2 emissions regulations because some of the emissions reduction strategies can increase fuel consumption (relative to a Tier 1 compliant diesel engine), depending on the combination of strategies employed. A combination of combustion improvements (that reduce NOx emissions leaving the engine) and aftertreatment (capturing and reducing NOx emissions via a NOx adsorption catalyst, or via selective catalytic reduction (SCR) using a reductant such as urea) that have left the engine before they leave the vehicle tailpipe) are being introduced on Tier 2 compliant light- duty diesel vehicles today. However, recently there have been a small number of announcements that diesel engines will be added to some passenger cars, in some cases a segment first for a manufacturer51, or that new passenger car diesel engines are being designed to meet all global emissions regulations.52 This suggests to the agencies that some manufacturers may be planning to use diesel engines in their plans to meet the tighter CAFE standards in the mid-term, which may be enabled by advances in diesel engine and emission control technology. Manufacturers that focus on diesel engines have also stated to the agencies their expectation that diesel engines will continue to be a viable technology for improving fuel economy and GHG emissions in the future. We spend time here discussing available emissions reduction technologies for diesel engines as part of this rulemaking because of the potential they have to impact fuel economy and GHG emissions for the vehicles that have them. With respect to combustion improvements, we note that several key advances in diesel engine combustion technology have made it possible to reduce emissions coming from the engine prior to aftertreatment, which reduces the need for aftertreatment. These technologies include improved fuel systems (higher injection pressure and multiple-injection capability), advanced controls and sensors to optimize combustion and emissions performance, higher EGR levels and EGR cooling to reduce NOx, and advanced turbocharging systems. These systems are available today and they do not adversely impact fuel efficiency. However, additional improvements in these technologies will be needed to reduce engine emissions further, should future emissions standards become more stringent. Further development may also be needed to reduce the fuel efficiency penalty associated with EGR. With respect to catalytic exhaust emission control systems, typical 3-way exhaust catalysts without NOx storage capability are not able to reduce NOx emissions from engines operated lean of stoichiometry (diesel or lean-burn gasoline). To reduce NOx, hydrocarbons, 3-98 ------- Technologies Considered in the Agencies' Analysis and particulate emissions, all diesels will require a catalyzed diesel particulate filter (CDPF) and sometimes a separate diesel oxidation catalyst (DOC), and either a lean NOx trap (LNT) ^ or the use of a selective catalytic reduction system, typically base-metal zeolite urea-SCR11. The increased cost of diesel emissions control technologies relative to powertrains with stoichiometric gasoline engines that are approaching comparable efficiency may also make diesels less attractive to manufacturers as a technology solution for more stringent CAFE and GHG standards. However, recognizing that some manufacturers may still employ diesel technology to meet the future standards, the agencies have included diesels in our analysis as follows: The agencies sought to ensure that diesel engines would have equivalent performance to comparable gasoline engine vehicles. For the Subcompact, Compact, and Midsize Passenger Car, Performance Subcompact Car, and Small Light Truck vehicle subclasses, the agencies assumed that an 14 gasoline base engine would be replaced by an in-line 4-cylinder diesel engine with displacement varying around 2.0 liters. For the Performance Compact, Performance Midsize, Large Passenger Car, Minivan, and Midsize Truck vehicle subclasses for the CAFE model, the agencies assumed that a V6 gasoline base engine would be replaced by an in-line 4-cylinder diesel engine with displacement varying around 2.8 liters. For the Large Truck and Performance Large Car vehicle subclasses for the CAFE model, the agencies assumed that a V8 gasoline base engine would be replaced with a V6 diesel engine with displacement varying around 4.0 liters to meet vehicle performance requirements. It was also assumed that diesel engines for all of these classes would utilize SCR aftertreatment systems given recent improvements in zeolite-based SCR systems and system efficiency. These assumptions impacted our estimates of the costs of implementing diesel engines as compared to the base gasoline engines. ^ A lean NOX trap operates by oxidizing NO to NO2 in the exhaust and storing NO2 on alkali sorbent material, most often BaO. When the control system determines (via mathematical model and typically a NOX sensor) that the trap is saturated with NOX, it switches the engine into a operating mode just rich of stoichiometry that allow NOx to be released from the alkali storage and temporarily allow three-way function of the catalyst similar to three-way catalysts used in stoichiometric gasoline applications. LNTs preferentially store sulfate compounds from the fuel, which reduces NOx storage capacity over time, thus the system must undergo periodic desulfurization by operating at a net-fuel-rich condition at high temperatures in order to retain NOX trapping efficiency. 11 An SCR aftertreatment system uses a reductant (typically, ammonia derived from urea) that is injected into the exhaust stream ahead of the SCR catalyst. Ammonia is a strong reductant even under net lean conditions. It combines with NOX in the SCR catalyst to form N2 and water. The hardware configuration for an SCR system is sometimes more complicated than that of an LNT, due to the onboard urea storage and delivery system (which requires a urea pump and injector to inject urea into the exhaust stream), which generally makes an SCR system cost more than an LNT system. While a rich engine-operating mode is not required for NOX reduction, the urea is typically injected at a rate of approximately 3 percent of the fuel consumed. The agencies understand that manufacturers designing SCR systems intend to align urea tank refills with standard maintenance practices such as oil changes as more diesel vehicles are introduced into the market. For diesel vehicles currently on the market, this is generally already the practice, and represents an ongoing maintenance cost for vehicles with this technology. 3-99 ------- Technologies Considered in the Agencies' Analysis Diesel engines are more costly than port-injected spark-ignition gasoline engines. These higher costs result from more costly components, more complex systems for emissions control, and other factors. The vehicle systems that are impacted include: • Fuel systems (higher pressures and more responsive injectors); • Controls and sensors to optimize combustion and emissions performance; • Engine design (higher cylinder pressures require a more robust engine, but higher torque output means diesel engines can have reduced displacement); • Turbocharger(s); • Aftertreatment systems, which tend to be more costly for diesels; In the MYs 2012-2016 final rule, the agencies estimated the DMC for converting a gasoline PFI engine with 3-way catalyst aftertreatment to a diesel engine with diesel aftertreatment at \$1,697 (2007\$), \$2,399 (2007\$), \$1,956 (2007\$) and \$2,676 (2007\$) for a small car, large car, medium/large MPV & small truck, and large truck, respectively (see final Joint TSD, Table 3-12 at page 3-44). All of these costs were for SCR-based diesel systems, with the exception of the small car, which was a LNT-based system. For this final rule, consistent with the proposal, we are using the same methodology as used in the MYs 2012- 2016 final rule, but have made four primary changes to the cost estimates as was also done in the proposal for this rule. First, the agencies have not estimated costs for a LNT-based system, and instead have estimated costs for all vehicle types assuming they will employ SCR-based systems. Second, the agencies assumed that manufacturers would meet a Tier 2 bin 2 average rather than a Tier 2 bin 5 average, assuming that more stringent levels of compliance will be required in the future. In order to estimate costs for Tier 2 bin 2 compliant vehicles, catalyst volume costs were estimated based on an assumed increase in volume of 20 percent. This was the estimated necessary increase needed to meet Tier 2, bin 2 emission level of 0.02 grams of NOx per mile. Increased catalyst volume resulted in a higher cost estimate for diesel aftertreatment than was estimated for the MYs 2012-2016 final rule. The third is to update all platinum group metal costs from the March 2009 values used in the 2012-2016 final rule to February 2011 values.111"1 The February 2011 values were used for purposes of the NPRM analysis, at which time they represented the most recent monthly average prices available at the time the agencies "locked-down" all cost estimates for the purposes of moving into the modeling phase of analysis."11 For the final rule analysis, the mm As reported by Johnson-Matthey, the March 2009 monthly average costs were \$1,085 per Troy ounce and \$1,169 per Troy ounce for platinum (Pt) and rhodium (Rh), respectively. As also reported by Johnson-Matthey, the February 2011 monthly average costs were \$1,829 per Troy ounce and \$2,476 per Troy ounce for Pt and Rh, respectively. See www.platinum.matthey.com. 1111 Note that there is no good way of determining what PGM prices to use when conducting cost analyses. Spot prices are inherently dangerous to use because spot prices, like stock prices on the stock market, can vary considerably from day to day. One could argue that an average price is best, but average prices can vary considerably depending on the length of time included in the average. And if too much time is included in the average, then average prices from a time prior to PGM use in diesel engines may be included which would lead some to conclude that we had cherry picked our values. Given no good option, it seems most transparent and 3-100 ------- Technologies Considered in the Agencies' Analysis agencies did not update the cost for platinum group metals. The fourth is to include an additional \$50 DMC for all costs to cover costs associated with improvements to fuel and urea controls. All of the diesel costs are considered applicable to MY 2012. The agencies consider diesel technology to be on the flat portion of the learning curve and have applied a medium complexity ICM of 1.39 through 2018, and then an ICM of 1.29 thereafter. The resultant costs are shown in Table 3-35. Table 3-35 Costs for Conversion to Advanced Diesel (2010\$) Cost type DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC Vehicle class Small car Standard car Large car Small MPV Large MPV Large truck Small car Standard car Large car Small MPV Large MPV Large truck Small car Standard car Large car Small MPV Large MPV Large truck 2017 \$2,059 \$2,059 \$2,522 \$2,064 \$2,082 \$2,886 \$905 \$905 \$1,109 \$907 \$915 \$1,268 \$2,965 \$2,965 \$3,631 \$2,971 \$2,996 \$4,154 2018 \$2,018 \$2,018 \$2,472 \$2,023 \$2,040 \$2,828 \$903 \$903 \$1,106 \$905 \$913 \$1,266 \$2,922 \$2,922 \$3,578 \$2,928 \$2,953 \$4,094 2019 \$1,978 \$1,978 \$2,422 \$1,982 \$1,999 \$2,772 \$675 \$675 \$827 \$677 \$683 \$946 \$2,653 \$2,653 \$3,249 \$2,659 \$2,682 \$3,718 2020 \$1,938 \$1,938 \$2,374 \$1,943 \$1,959 \$2,716 \$674 \$674 \$826 \$676 \$681 \$945 \$2,612 \$2,612 \$3,200 \$2,618 \$2,641 \$3,661 2021 \$1,900 \$1,900 \$2,326 \$1,904 \$1,920 \$2,662 \$673 \$673 \$824 \$674 \$680 \$943 \$2,572 \$2,572 \$3,151 \$2,578 \$2,600 \$3,605 2022 \$1,862 \$1,862 \$2,280 \$1,866 \$1,882 \$2,609 \$672 \$672 \$823 \$673 \$679 \$941 \$2,533 \$2,533 \$3,103 \$2,539 \$2,561 \$3,550 2023 \$1,824 \$1,824 \$2,234 \$1,828 \$1,844 \$2,556 \$671 \$671 \$821 \$672 \$678 \$940 \$2,495 \$2,495 \$3,056 \$2,501 \$2,522 \$3,496 2024 \$1,788 \$1,788 \$2,190 \$1,792 \$1,807 \$2,505 \$669 \$669 \$820 \$671 \$677 \$938 \$2,457 \$2,457 \$3,010 \$2,463 \$2,484 \$3,443 2025 \$1,752 \$1,752 \$2,146 \$1,756 \$1,771 \$2,455 \$668 \$668 \$819 \$670 \$676 \$937 \$2,420 \$2,420 \$2,964 \$2,426 \$2,446 \$3,392 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. For the MYs 2012-16 final rule and TAR, NHTSA and EPA estimated the fuel consumption reduction of a SCR-based diesel engine to be between 20 to 25 percent over a baseline gasoline engine. NHTSA and EPA have revisited these values and have now estimated, based on the Ricardo 2011 study, the effectiveness of a SCR-based diesel engine to be 28.4 to 30.5 percent. For purposes of CC>2 reduction, EPA estimates a 7 to 20 percent for light-duty diesels equipped with SCR. 3.4.2 Transmission Technologies NHTSA and EPA have also reviewed the transmission technology estimates used in the 2012-2016 final rule and the 2010 TAR. In doing so, NHTSA and EPA considered or reconsidered all available sources and updated the estimates as appropriate. The section below describes each of the transmission technologies considered for this rulemaking. As discussed above, for the final rule NHTSA has updated the effectiveness values for advanced transmissions when coupled to naturally-aspirated engines based on the ANL simulation least self-serving to simply choose a price and report its basis. In the end, the PGM costs represent 16-23 percent of the diesel DMC in this analysis. Further, diesels play very little to no role in enabling compliance with the final standards. 3-101 ------- Technologies Considered in the Agencies' Analysis modeling. These changes are documented in detail in NHTSA's RIA. These changes are not included in this joint TSD because they are specific to NHTSA's analysis only. 3.4.2.1 Improved Automatic Transmission Control (Aggressive Shift Logic and Early Torque Converter Lockup) Calibrating the transmission shift schedule to upshift earlier and quicker, and to lock- up or partially lock-up the torque converter under a broader range of operating conditions can reduce fuel consumption and CC>2 emissions. However, this operation can result in a perceptible degradation in noise, vibration, and harshness (NVH). The degree to which NVH can be degraded before it becomes noticeable to the driver is strongly influenced by characteristics of the vehicle, and although it is somewhat subjective, it always places a limit on how much fuel consumption can be improved by transmission control changes. Aggressive Shift Logic and Early Torque Converter Lockup are best optimized simultaneously when added to an automatic transmission due to the fact that adding both of them requires only minor modifications to the transmission mechanical components or calibration software. As a result, these two technologies are combined in the modeling when added to an automatic transmission. Since a dual clutch transmission (DCT) has no torque converter, the early torque converter lockup technology is not included when adding ASL to the DCT. 3.4.2.2 Aggressive Shift Logic During operation, a transmission's controller manages the operation of the transmission by scheduling the upshift or downshift, and, in automatic transmissions, locking or allowing the torque converter to slip based on a preprogrammed shift schedule. The shift schedule contains a number of lookup table functions, which define the shift points and torque converter lockup based on vehicle speed and throttle position, and other parameters such as temperature. Aggressive shift logic (ASL) can be employed in such a way as to maximize fuel efficiency by modifying the shift schedule to upshift earlier and inhibit downshifts under some conditions, which reduces engine pumping losses and engine friction. The application of this technology does require a manufacturer to confirm that drivability, durability, and NVH are not significantly degraded. For this final rule, consistent with the proposal, the agencies considered two levels of ASL. The first level is that discussed in the 2012-2016 final rule and the 2010 TAR. ASL- level 1 is an early upshift strategy whereby the transmission shifts to the next higher gear "earlier" (or at lower RPM during a gradual acceleration) than would occur in a traditional automatic transmission. This early upshift reduces fuel consumption by allowing the engine to operate at a lower RPM and higher load, which typically moves the engine into a more efficient operating region. ASL-level 2 is a shift optimization strategy whereby the engine and/or transmission controller(s) continuously evaluate all possible gear options that would provide the necessary tractive power (while limiting the adverse effects on driveline NVH) and select the gear that lets the engine run in the most efficient operating zone. Ricardo acknowledged in its report that the ASL-level 2 ("shift optimization") strategy currently causes significant implications 3-102 ------- Technologies Considered in the Agencies' Analysis for drivability and hence affects consumer acceptability. However, Ricardo recommended the inclusion of this technology for the 2020-2025 timeframe with the assumption that manufacturers will develop a means of yielding the fuel economy benefit without adversely affecting driver acceptability. The agencies believe these drivability challenges could include shift busyness - that is, a high level of shifting compared to current vehicles as perceived by the customers. The agencies note that in confidential discussions with two major transmission suppliers, the suppliers described transmission advances which reduce shifting time and provide smoother torque transitions than today's designs, making the shifting event less apparent to the driver, however these improvements will not influence the customer's perception of shift business related to the changes in engine speed. In addition, the agencies note that several auto companies and transmission firms have announced future introduction of transmissions into the U.S. market with even a higher number of gears than were included in the Ricardo simulation and in the agencies' feasibility assessment for this final rule, consistent with the proposal (which is 8 forward speeds). These announcements include both 9 and 10 speed transmissions which may present further challenges with shift busyness, given the availability of one or two additional gears. At the same time, the associated closer gear spacing will generally result in smaller engine speed changes during shifting that may be less noticeable to the driver. The agencies are including shift optimization in the analysis under the premise that manufacturers and suppliers are developing means to mitigate these drivability issues by MY 2017, as assumed in the 2011 Ricardo study (more information on Ricardo's treatment of the optimized shift strategy is described in Section 6.4 of the 2011 Ricardo report). If manufacturers are not able to solve these drivability issues, the assumed effectiveness could be lower and the cost could be higher or both. The agencies sought comment on the feasibility of ASL-level 2 and the likelihood that manufacturers will be able to overcome the drivability issues, however no comments were submitted on this issue. In MYs 2012-2016 final rule, the agencies estimated an effectiveness improvement of 1 to 2 percent for aggressive shift logic which was supported by the 2002 NAS and NESCCAF reports as well as confidential manufacturer data. The agencies updated the effectiveness of ASL-level 1 ranging from 1.9 to 2.7 based on 2010 Ricardo study. In CAFE model an incremental effectiveness ranging for both ASL and early torque converter lockup ranging from 2.3 to 3.1 percent is applied (Early torque converter has effectiveness of 0.5 percent). ASL-level 2 is new to this analysis which is based on the shift optimization algorithm in 2011 Ricardo study. The effectiveness for ASL-level 2 ranges from 5.1 to 7.0 percent improvement over transmission with unimproved shift logic or roughly 4 to 5 percent over a transmission that already incorporates aggressive shift logic. In the CAFE model, an incremental effectiveness ranging from 3.27 to 4.31 percent is applied. In the 2012-2016 rule, the agencies estimated the DMC at \$26 (2007\$) which was considered applicable to the 2015MY. This DMC becomes \$27 (2010\$) for this analysis. The agencies consider ASL-level 1 technology to be on the flat portion of the learning curve and have applied a medium complexity ICM of 1.39 through 2018 then 1.29 thereafter. For 3-103 ------- Technologies Considered in the Agencies' Analysis ASL-level 2, the agencies are estimating the DMC at an equivalent \$27 (2010\$) except that this cost is considered applicable to the 2017MY. Essentially this yields a nearly negligible incremental cost for ASL-level 2 over ASL-level 1. The agencies consider ASL-level 2 technology to be on the flat portion of the learning curve and have applied a medium complexity ICM of 1.39 through 2024 then 1.29 thereafter. The timing of the ASL-level 2 ICMs is different than that for the level 1 technology because the level 2 technology is newer and not yet being implemented in the fleet. The resultant costs are shown in Table 3-36. Note that both levels of ASL technology are incremental to the baseline system, so ASL-level 2 is not incremental to ASL-level 1. Table 3-36 Costs for Aggressive Shift Logic Levels 1 & 2 (2010\$) Cost type DMC DMC 1C 1C TC TC Technology ASL-level 1 ASL-level 2 ASL-level 1 ASL-level 2 ASL-level 1 ASL-level 2 Transmission type All All All All All All 2017 \$26 \$27 \$7 \$7 \$33 \$34 2018 \$26 \$27 \$7 \$7 \$32 \$33 2019 \$25 \$26 \$5 \$7 \$30 \$32 2020 \$24 \$25 \$5 \$7 \$30 \$32 2021 \$24 \$24 \$5 \$7 \$29 \$31 2022 \$24 \$24 \$5 \$7 \$29 \$30 2023 \$23 \$23 \$5 \$7 \$28 \$30 2024 \$23 \$23 \$5 \$7 \$28 \$29 2025 \$22 \$22 \$5 \$5 \$27 \$27 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. 3.4.2.3 Early Torque Converter Lockup A torque converter is a fluid coupling located between the engine and transmission in vehicles with automatic transmissions and continuously-variable transmissions (CVT). This fluid coupling allows for slip so the engine can run while the vehicle is idling in gear (as at a stop light), provides for smoothness of the powertrain, and also provides for torque multiplication during acceleration, and especially launch. During light acceleration and cruising, the inherent slip in a torque converter causes increased fuel consumption, so modern automatic transmissions utilize a clutch in the torque converter to lock it and prevent this slippage. Fuel consumption can be further reduced by locking up the torque converter at lower vehicle speeds, provided there is sufficient power to propel the vehicle, and noise and vibration are not excessive.00 If the torque converter cannot be fully locked up for maximum efficiency, a partial lockup strategy can be employed to reduce slippage. Early torque converter lockup is applicable to all vehicle types with automatic transmissions. Some torque converters will require upgraded clutch materials to withstand additional loading and the slipping conditions during partial lock-up. As with aggressive shift logic, confirmation of acceptable drivability, performance, durability and NVH characteristics is required to successfully implement this technology. 00 Although only modifications to the transmission calibration software are considered as part of this technology, very aggressive early torque converter lock up may require an adjustment to damper stiffness and hysteresis inside the torque converter. 3-104 ------- Technologies Considered in the Agencies' Analysis Regarding the effectiveness of Early Torque Converter Lockup, the 2012-2016 final rule, TAR, and the 2010 Ricardo study estimated an effectiveness improvement of 0.4 to 0.5 percent. In the 2012-2016 rule, the agencies estimated the DMC at \$24 (2007\$) which was considered applicable to the 2015MY. This DMC remains \$25 (2010\$) for this analysis.pp The agencies consider early torque converter lockup technology to be on the flat portion of the learning curve and have applied a low complexity ICM of 1.24 through 2018 then 1.19 thereafter. The resultant costs are shown in Table 3-37. Table 3-37 Costs for Early Torque Converter Lockup (2010\$) Cost type DMC 1C TC Transmission type Automatic Automatic Automatic 2017 \$24 \$6 \$30 2018 \$23 \$6 \$29 2019 \$23 \$5 \$27 2020 \$22 \$5 \$27 2021 \$22 \$5 \$27 2022 \$21 \$5 \$26 2023 \$21 \$5 \$26 2024 \$20 \$5 \$25 2025 \$20 \$5 \$25 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. 3.4.2.4 High Efficiency Gearbox For this rule, a high efficiency gearbox refers to some or all of a suite of incremental gearbox improvement technologies that should be available within the 2017 to 2025 timeframe. The majority of these improvements address mechanical friction within the gearbox. These improvements include but are not limited to: shifting clutch technology improvements (especially for smaller vehicle classes), improved kinematic design, dry sump lubrication systems, more efficient seals, bearings and clutches (reducing drag), component superfinishing and improved transmission lubricants. More detailed description can be found in the 2011 Ricardo report53. Note that the high efficiency gearbox technology is applicable to any type of transmission. EPA analyzed detailed transmission efficiency input data provided by Ricardo and implemented it directly into the lumped parameter model. Based on the LP effectiveness resulting from these inputs, EPA and NHTSA estimate that a high efficiency gearbox can provide a GHG or fuel consumption reduction in the range of 3.8 to 5.7 percent (3.8% for 4WD trucks with an unimproved rear axle) over a baseline automatic transmission in MY2017 and beyond. The agencies estimate the DMC of the high efficiency gearbox at \$200 (2009\$). We have based this on the DMC for engine friction reduction in a V8 engine which, as presented in Table 3-24 is \$197 (2010\$). In the proposal, we rounded this value up to \$200 (2009\$) pp As is true throughout this presentation of cost estimates, the agencies round costs to the nearest dollar. In the actual model input files, the cost in 2007\$ would have been \$23.68 and the cost in 2009\$ is \$24.42. So an impact of the dollar-year conversion is reflected in the analysis even when it does not appear so in this presentation. 3-105 ------- Technologies Considered in the Agencies' Analysis which becomes \$202 (2010\$) for the final analysis. This DMC is considered applicable for the 2017MY. The agencies consider high efficiency gearbox technology to be on the flat portion of the learning curve and have applied a low complexity ICM of 1.24 through 2024 then 1.19 thereafter. The resultant costs are shown in Table 3-38. Table 3-38 Costs for High Efficiency Gearbox (2010\$) Cost type DMC 1C TC Transmission type Automatic/Dual clutch Automatic/Dual clutch Automatic/Dual clutch 2017 \$202 \$49 \$251 2018 \$196 \$49 \$245 2019 \$190 \$49 \$239 2020 \$184 \$49 \$233 2021 \$179 \$49 \$227 2022 \$173 \$49 \$222 2023 \$170 \$48 \$218 2024 \$167 \$48 \$215 2025 \$163 \$39 \$202 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; all costs are incremental to the baseline. 3.4.2.5 Automatic 6-, 7- and 8-Speed Transmissions (NAUTO and 8SPD) Manufacturers can also choose to replace 4- and 5-speed transmission with 6-, 7-, or 8-speed automatic transmissions. Additional ratios allow for further optimization of engine operation over a wider range of conditions, but this is subject to diminishing returns as the number of speeds increases. As additional planetary gear sets are added (which may be necessary in some cases to achieve the higher number of ratios), additional weight and friction are introduced. Also, the additional shifting of such a transmission can be perceived as bothersome to some consumers, so manufacturers need to develop strategies for smooth shifts. Some manufacturers are replacing 4- and 5-speed automatics with 6-speed automatics, and 7- and 8-speed automatics have also entered production. While a six speed transmission application was most prevalent for the 2012-2016 final rule, eight speed transmissions are expected to be readily available and applied in the 2017 through 2025 timeframe. As discussed in the MY 2011 CAFE final rule, confidential manufacturer data projected that 6-speed transmissions could incrementally reduce fuel consumption by 0 to 5 percent from a baseline 4-speed automatic transmission, while an 8-speed transmission could incrementally reduce fuel consumption by up to 6 percent from a baseline 4-speed automatic transmission. GM has publicly claimed a fuel economy improvement of up to 4 percent for its new 6-speed automatic transmissions.54 The 2008 EPA Staff Technical Report found a 4.5 to 6.5 percent fuel consumption improvement for a 6-speed over a 4-speed automatic transmission.55 Based on this information, NHTSA estimated in the MY 2011 rule, that the conversion to a 6-,7- and 8-speed transmission (NAUTO) from a 4 or 5-speed automatic transmission with IATC would have an incremental fuel consumption benefit of 1.4 percent to 3.4 percent, for all vehicle classes. From a baseline 4 or 5 speed transmission without IATC, the incremental fuel consumption benefit would be approximately 3 to 6 percent, which is consistent with the EPA Staff Report estimate. In MYs 2012-2016 final rule, NHTSA and EPA reviewed these effectiveness estimates and concluded that they remain accurate. While the CAFE model follows the incremental approach discussed above, the GHG model estimates the packaged effectiveness of 4.5 to 6.5 percent 3-106 ------- Technologies Considered in the Agencies' Analysis In this FRM analysis, consistent with the proposal, the agencies divided the improvement for this technology into two steps, first from 4 or 5 speed transmission to 6 or 7 speed transmission (NAUTO), then from 6 or 7 speed transmission to 8 speed transmission (8SPD). The effectiveness estimates for NAUTO and 8SPD are based on 2011 Ricardo study. In this FRM analysis, consistent with the proposal, the effectiveness for a 6-speed transmission relative to a 4-speed base transmission ranges from 3.1 to 3.9 percent (2.1 percent for large truck with unimproved rear axle) including 7 percent of transmission gearbox efficiency improvement that the agencies assumed accompanying the new 6 speed transmission after MY 2010. NHTSA incorporated this effectiveness estimate into the CAFE model as incremental improvement over IATC ranging from 1.89 to 2.13 percent. In this FRM analysis, consistent with the proposal, the agencies assumed that 8-speed transmission will not start to phase in until MY2017. NHTSA applied 8-speed automatic transmission succeeding 6-speed automatic transmission to vehicles with towing requirement, such as Minivan, Midsize light truck and large light truck. All other vehicle subclasses use 8-speed DCT to succeed 6-speed DCT. The effectiveness for an 8-speed DCT relative to a 4-speed DCT transmission ranges from 11.1 to 13.1 percent for subcompact car, small car and small light truck. The effectiveness for an 8-speed automatic transmission relative to 4-speed automatic transmission ranges for large CUV and large truck ranges from 8.7 to 9.2 percent in the lumped parameter model. This translates into effectiveness in the range of 3.85 to 4.57 percent for an 8-speed DCT relative to a 6-speed DCT and 4.9 to 5.34 percent for 8-speed automatic transmission relative to 6-speed automatic transmission in CAFE model. In the 2010 TAR, the agencies estimated the DMC at -\$13 (2008\$) for a 6 speed automatic transmission relative to a 4 speed auto transmission, applicable in the 2017MY (see 2010 TAR, Table B2.1-1 at page B-10). For the 2012MY, that DMC was -\$15 (2008\$), although that value was not presented in the TAR. The latter DMC remains -\$15 (2010\$) for this analysis which is considered to be applicable in the 2012MY. The agencies consider 6 speed automatic transmission technology to be on the flat portion of the learning curve and have applied a low complexity ICM of 1.24 through 2018 then 1.19 thereafter. The resultant costs are shown in Table 3-39. New for the proposal was the cost of an 8 speed automatic transmission. For the cost of this technology, the agencies have relied on a tear-down study completed by FEV since publication of the TAR. 6 In that study, the 8 speed auto transmission was found to be \$62 (2007\$) more costly than the 6 speed auto transmission. This DMC becomes \$64 (2010\$) for this analysis. Adding the \$64 (2010\$) to the -\$15 (2010\$) DMC for a 6 speed relative to a 4 speed, the 8 speed auto transmission relative to a 4 speed auto transmission would be \$50 (2010\$). The agencies consider this DMC to be applicable to the 2012MY. The agencies consider the 8 speed auto transmission technology to be on the flat portion of the learning curve and have applied a medium complexity ICM of 1.39 through the 2018MY then 1.29 3-107 ------- Technologies Considered in the Agencies' Analysis thereafter.qq The resultant costs for both 6 speed and 8 speed auto transmissions are shown in Table 3-39. Table 3-39 Costs for 6 and 8 Speed Automatic Transmissions (2010\$) Cost type DMC DMC DMC 1C 1C 1C TC TC TC Transmission type 6spAT from 4spAT SspAT from 6spAT SspAT from 4spAT 6spAT from 4spAT SspAT from 6spAT SspAT from 4spAT 6spAT from 4spAT SspAT from 6spAT SspAT from 4spAT 2017 -\$13 \$56 \$43 \$4 \$25 \$19 -\$9 \$80 \$62 2018 -\$13 \$55 \$42 \$4 \$24 \$19 -\$9 \$79 \$61 2019 -\$12 \$54 \$41 \$3 \$18 \$14 -\$10 \$72 \$55 2020 -\$12 \$53 \$40 \$3 \$18 \$14 -\$9 \$71 \$54 2021 -\$12 \$51 \$40 \$3 \$18 \$14 -\$9 \$70 \$54 2022 -\$12 \$50 \$39 \$3 \$18 \$14 -\$9 \$69 \$53 2023 -\$11 \$49 \$38 \$3 \$18 \$14 -\$9 \$68 \$52 2024 -\$11 \$48 \$37 \$3 \$18 \$14 -\$8 \$67 \$51 2025 -\$11 \$47 \$37 \$3 \$18 \$14 -\$8 \$66 \$50 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; sp=speed; AT=automatic transmission Note that the cost for the 8 speed automatic transmission relative to the 6 speed automatic transmission is lower here than that used in the recent heavy-duty GHG rule. In that rule, we remained consistent with the proposal for that rule which carried an estimated DMC of \$210 (2008\$). That DMC was based on an estimate derived by NAS (see NAS 2010, Table 7-10). For this final rule, consistent with the proposal, we have chosen to use the more recent DMC shown in Table 3-39 which is based on a tear-down analysis done by FEV. 3.4.2.6 Dual Clutch Transmissions / Automated Manual Transmissions (DCTAM) An Automated Manual Transmission (AMT) is mechanically similar to a conventional manual transmission, but shifting and launch functions are automatically controlled by the electronics. There are two basic types of AMTs, single-clutch and dual-clutch (DCT). A single-clutch AMT is essentially a manual transmission with automated clutch and shifting. Because of shift quality issues with single-clutch designs, DCTs are far more common in the U.S. and are the basis of the estimates that follow. A DCT uses separate clutches (and separate gear shafts) for the even-numbered gears and odd-numbered gears. In this way, the next expected gear is pre-selected, which allows for faster and smoother shifting. For example, if the vehicle is accelerating in third gear, the shaft with gears one, three and five has gear three engaged and is transmitting power. The shaft with gears two, four, and six is idle, but has gear four engaged. When a shift is required, the controller disengages the odd- gear clutch while simultaneously engaging the even-gear clutch, thus making a smooth shift. If, on the other hand, the driver slows down instead of continuing to accelerate, the transmission will have to change to second gear on the idling shaft to anticipate a downshift. This shift can be made quickly on the idling shaft since there is no torque being transferred on it. qq This ICM would be applied to the 6 speed to 8 speed increment of \$64 (2010\$) applicable in 2012. The 4 speed to 6 speed increment would carry the low complexity ICM. 3-108 ------- Technologies Considered in the Agencies' Analysis In addition to single-clutch and dual-clutch AMTs, there are also wet clutch and dry clutch designs which are used for different types of vehicle applications. Wet clutch AMTs offer a higher torque capacity that comes from the use of a hydraulic system that cools the clutches. Wet clutch systems are less efficient than the dry clutch systems due to the losses associated with hydraulic pumping. Additionally, wet AMTs have a higher cost due to the additional hydraulic hardware required. Overall, DCTs likely offer the greatest potential for effectiveness improvements among the various transmission options presented in this report because they offer the inherently lower losses of a manual transmission with the efficiency and shift quality advantages of electronic controls. The lower losses stem from the elimination of the conventional lock-up torque converter, and a greatly reduced need for high pressure hydraulic circuits to hold clutches or bands to maintain gear ratios (in automatic transmissions) or hold pulleys in position to maintain gear ratio (in Continuously Variable Transmissions). However, the lack of a torque converter will affect how the vehicle launches from rest, so a DCT will most likely be paired with an engine that offers sufficient torque at low engine speeds to allow for adequate launch performance or provide lower launch gears to approximate the torque multiplication of the torque converter to provide equivalent performance. In MYs 2012-2016 final rule, EPA and NHTSA estimated a 5.5 to 9.5 percent improvement in fuel consumption over a baseline 4/5-speed automatic transmission for a wet clutch DCT, which was assumed for all but the smallest of vehicle subclasses, Subcompact and Compact cars and small LT. This results in an incremental effectiveness estimate of 2.7 to 4.1 percent over a 6-speed automatic transmission with IATC. For Subcompact and Compact Cars and small LT, which were assumed to use a dry clutch DCT, NHTSA estimated an 8 to 13 percent fuel consumption improvement over a baseline 4/5-speed automatic transmission, which equates to a 5.5 to 7.5 percent incremental improvement over the 6-speed transmission. Based on the 2011 Ricardo study, EPA and NHTSA have concluded that 8 to 13 percent effectiveness is appropriate for 6-speed DCTs and 11 to 16 percent is appropriate for 8-speed DCTs for this final rule, consistent with the proposal. These values include not only the DCT but also the increase in stepped gears and also a high efficiency gearbox (mentioned later). Independent of other technologies, this translates to an effectiveness for the DCT, alone, of 4 to 5% (for wet-clutch designs) and 5 to 6% (for dry-clutch designs) compared to a baseline automatic transmission of similar vintage and number of fixed gears. In this FRM analysis, consistent with the proposal, NHTSA applied an incremental effectiveness of 4 percent for a 6-speed dry DCT and 3.4 to 3.8 percent for a wet DCT compared to a 6-speed automatic transmission based on the lumped parameter model which includes the accompanied transmission efficiency improvement for MY 2010 and after transmissions. This translates to an effectiveness range of 7.4 to 8.6 percent compared to a 4 speed automatic transmission for dry clutch design and 7.4 to 7.9 percent for a wet clutch design. NHTSA did not apply DCTs to vehicles with towing requirements, such as Minivan, Midsize light truck and large pickup truck. EPA did not apply DCTs to vehicle types classified as towing as described in Chapter 1 of EPA's RIA. 3-109 ------- Technologies Considered in the Agencies' Analysis In the 2010 TAR, the agencies estimated the DMC at -\$234 (2008\$) for a 6 speed dry- clutch DCT and -\$165 for a 6 speed wet-clutch DCT with both DMCs applicable in the 2017MY (see 2010 TAR, Table B2.1-1 at page B-10) and both incremental to a 4 speed automatic transmission. In the 2010 TAR, we pointed to Chapter 3 of the 2012-2016 final joint TSD where we noted that the DCT costs of-\$147 (2007\$ and incremental to a 6-speed automatic transmission) were based on a FEV tear-down study that assumed 450,000 units of production. We went on to state that we did not consider there to be sufficient US capacity in the 2012-2016 timeframe to produce 450,000 units and for that reason we were adjusting the tear-down values accordingly. The TAR timeframe for consideration was 2017-2025, and in the TAR we argued that production capacity would exist and that the FEV tear-down results be valid without adjustment. As noted in the proposal to this rule, we continue to believe that to be the case. In the final joint TSD supporting the 2012-2016 rule we also noted that the negative tear-down estimates found by FEV were not surprising when considering the relative simplicity of a dual-clutch transmission compared to an automatic transmission. Again, we continue to consider this to be true. For this analysis, we consider the 2010 TAR DMCs to be applicable to the 2012MY, thus the DMCs become -\$238 (2010\$) and -\$168 (2010\$) for 6 speed dry- and wet-clutch DCTs, respectively, both applicable in the 2012MY and incremental to a 4 speed auto transmission. The agencies consider the 6 speed DCT technology to be on the flat portion of the learning curve and have applied a medium complexity ICM of 1.39 through 2018 then 1.29 thereafter. The resultant costs are shown in Table 3-40. New for this rulemaking is costing for an 8 speed DCT. For the cost of this technology, the agencies have relied on a tear-down study completed by FEV since publication of the TAR.57 In that study, the 8 speed DCT was found to be \$198 (2007\$) more costly than the 6 speed DCT. This DMC increment becomes \$206 (2010\$) for this analysis. Adding the \$206 (2010\$) to the -\$238 (2010\$) DMC and the -\$168 (2010\$) DMC for a 6 speed dry- and wet-clutch DCT, the 8 speed dry- and wet-clutch DCTs relative to a 4 speed auto transmission would be -\$32 (2010\$) and \$38 (2010\$), respectively. The agencies consider this DMC to be applicable to the 2012MY. The agencies consider the 8 speed DCT technology to be on the flat portion of the learning curve and have applied a medium complexity ICM of 1.39 through the 2024MY then 1.29 thereafter. The 8 speed DCT has a later switch to long term ICMs because it is a newer technology that is not currently implemented in the fleet. The resultant costs for both 6 speed and 8 speed DCTs are shown in Table 3-40. Table 3-40 Costs for 6 & 8 Speed Dual Clutch Transmissions (2010\$) Cost type DMC DMC DMC DMC 1C 1C 1C 1C Transmission type 6spDCT-dry 6sp DCT-wet 8sp DCT-dry 8sp DCT-wet 6spDCT-dry 6sp DCT-wet 8sp DCT-dry 8sp DCT-wet 2017 -\$207 -\$146 -\$28 \$33 \$91 \$64 \$12 \$14 2018 -\$203 -\$143 -\$27 \$32 \$91 \$64 \$12 \$14 2019 -\$199 -\$140 -\$27 \$32 \$68 \$48 \$12 \$14 2020 -\$195 -\$137 -\$26 \$31 \$68 \$48 \$12 \$14 2021 -\$191 -\$134 -\$26 \$30 \$68 \$48 \$12 \$14 2022 -\$187 -\$132 -\$25 \$30 \$67 \$48 \$12 \$14 2023 -\$183 -\$129 -\$25 \$29 \$67 \$47 \$12 \$14 2024 -\$179 -\$127 -\$24 \$29 \$67 \$47 \$12 \$14 2025 -\$176 -\$124 -\$24 \$28 \$67 \$47 \$9 \$11 3-110 ------- Technologies Considered in the Agencies' Analysis TC TC TC TC 6spDCT-dry 6sp DCT-wet 8sp DCT-dry 8sp DCT-wet -\$116 -\$82 -\$16 \$47 -\$112 -\$79 -\$15 \$47 -\$131 -\$92 -\$15 \$46 -\$127 -\$89 -\$14 \$45 -\$123 -\$87 -\$14 \$45 -\$119 -\$84 -\$13 \$44 -\$116 -\$82 -\$13 \$44 -\$112 -\$79 -\$12 \$43 -\$109 -\$77 -\$15 \$39 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; sp=speed; dry=dry clutch; wet=wet-clutch Note that all costs are relative to a 4 speed automatic transmission. 3.4.2.7 6-Speed Manual Transmissions (6MAN) Manual transmissions are entirely dependent upon driver input to shift gears: the driver selects when to perform the shift and which gear to select. This is the most efficient transfer of energy of all transmission layouts, because it has the lowest internal gear losses, with a minimal hydraulic system, and the driver provides the energy to actuate the clutch. From a systems viewpoint, however, vehicles with manual transmissions have the drawback that the driver may not always select the optimum gear ratio for fuel economy. Nonetheless, increasing the number of available ratios in a manual transmission can improve fuel economy by allowing the driver to select a ratio that optimizes engine operation more often. Typically, this is achieved through adding overdrive ratios to reduce engine speed at cruising velocities (which saves fuel through reduced engine pumping losses) and pushing the torque required of the engine towards the optimum level. However, if the gear ratio steps are not properly designed, this may require the driver to change gears more often in city driving, resulting in customer dissatisfaction. Additionally, if gear ratios are selected to achieve improved launch performance instead of to improve fuel economy, then no fuel saving effectiveness is realized. The 2012-2016 final rule estimated an effectiveness increase of 0.5 percent for replacing a 5-speed manual with a 6-speed manual transmission, which was derived from confidential manufacturer data Based on the updated LPM for this 2017-2025 rule, NHTSA has found that an effectiveness increase of 2.0 to 2.5 percent is possible when moving from a 5-speed to a 6-speed manual transmission with improved internals. NHTSA updated costs from the 2012-2016 final rule to reflect the ICM low complexity markup of 1.11 which resulted in an incremental compliance cost of \$250 as compared to \$338 for MY 2012. This represents a DMC of \$225 (2007\$) which becomes \$234 (2010\$) for this analysis, applicable in the 2012MY. NHTSA continues to consider a 6 speed manual transmission to be on the flat portion of the learning curve and has applied a low complexity ICM of 1.24 through 2018 then 1.19 thereafter. NHTSA's resultant costs for a 6 speed manual transmission are shown in Table 3-41. Table 3-41 Costs for 6 Speed Manual Transmission (2010\$) Cost type DMC 1C TC Transmission type 6sp manual 6sp manual 6sp manual 2017 \$204 \$57 \$260 2018 \$199 \$57 \$256 2019 \$196 \$45 \$240 2020 \$192 \$44 \$236 2021 \$188 \$44 \$232 2022 \$184 \$44 \$229 2023 \$181 \$44 \$225 2024 \$177 \$44 \$221 2025 \$173 \$44 \$218 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost; sp=speed; dry=dry clutch; wet=wet-clutch Note that all costs are relative to a 5 speed manual transmission. 3-111 ------- Technologies Considered in the Agencies' Analysis 3.4.3 Vehicle electrification and hybrid electric vehicle technologies For the costs presented in this electrification and hybrid vehicle section, we have estimated costs for vehicle classes since the technologies are closely linked to the size of the vehicle as opposed to the number of cylinders on the engine or its valvetrain configuration. The vehicle classes for which we have estimated costs are consistent with the six vehicle classes developed for the lumped parameter model. Each agency has used the vehicle class specific costs and mapped those into their respective model-specific vehicle classes or types as shown in Table 3-42. This table simply presents the mapping of lumped parameter model vehicle classes (or cost vehicle classes) into model-specific vehicle classes (or vehicle types in the case of EPA's OMEGA model, please refer to Chapter 1 of EPA's final RIA for more details) to help the reader understand how the vehicle classes used for costing relate to the vehicle classes used for modeling. Note that there have been changes in the EPA data since the proposal. EPA now characterizes cost vehicle classes more consistently with the way they are classified in the lumped parameter model to avoid any confusion that the proposed cost vehicle classes may have generated. EPA has also reconfigured its 19 vehicle types in an effort to more closely align the vehicle types with the actual vehicles contained in each. Both of these changes are detailed in Chapter 1 of EPA's final RIA. Table 3-42 Mapping of Vehicle Class into each Agency's Model-Specific Vehicle Classes or Types EPA Vehicle Class for Cost Purpose Subcompact/ Small Car Standard Car Large Car Small MPV Large MPV Truck Lumped Parameter Classification Small Car Standard Car Large Car Small MPV Large MPV Truck Example Fiesta Focus Yaris Fusion Taurus Camry Crown Victoria Mustang Escape Rav4 Tacoma Edge Explorer 4Runner Sienna F150 Tundra OMEGA Model Vehicle Type* 1 2,3,4 5,6 7,13 8, 9, 10, 14, 15 11, 12, 16, 17, 18, 19 NHTSA/CAFE Model Classification Subcompact Subcompact Perf PC Compact Compact Perf PC Mid-size PC Mid-size Perf PC Large PC Large Perf PC Small LT Midsize LT Minivan LT Large LT * OMEGA uses 19 vehicle types as shown here and described in detail in Chapter 1 of EPA's final RIA. 3-112 ------- Technologies Considered in the Agencies' Analysis 3.4.3.1 Electrical Power Steering (EPS) / Electrohydraulic Power Steering (EHPS) Electric power steering (EPS) and Electrohydraulic power steering (EHPS) provide a potential reduction in CC>2 emissions and fuel consumption over hydraulic power steering because of reduced overall accessory loads. This eliminates the parasitic losses associated with belt-driven power steering pumps which consistently draw load from the engine to pump hydraulic fluid through the steering actuation systems even when the wheels are not being turned. EPS is an enabler for all vehicle hybridization technologies since it provides power steering when the engine is off. EPS may be implemented on most vehicles with a standard 12V system. Some heavier vehicles may require a higher voltage system or EHPS which may add cost and complexity. The 2012-2016 final rule, EPA and NHTSA estimated a 1 to 2 percent effectiveness for light duty vehicles based on the 2002 NAS report, Sierra Research Report and confidential OEM data. The 2010 Ricardo study also confirmed this estimate. NHTSA and EPA reviewed these effectiveness estimates and found them to be accurate, thus they have been retained for this final rule, consistent with the proposal. For large pickup truck the agencies used EHPS due to the utility requirement of these vehicles. The effectiveness of EHPS is estimated to be 0.8 percent. In the MY 2012-2016 final rule, the agencies estimated the DMC at \$88 (2007\$). Converting to 2010\$, this DMC becomes \$92 for this analysis, consistent with the recent heavy-duty GHG rule, which is considered applicable in the 2015MY. The agencies use the same DMC for EPS as for EHPS. Technically, EHPS is less costly than EPS. However, we believe that EHPS is likely to be used, if at all, on the largest trucks and utility vehicles. As such, it would probably need to be heavier-duty than typical EPS systems and the agencies consider the net effect to place EHPS on par with EPS in terms of costs. The agencies consider EPS/EHPS technology to be on the flat portion of the learning curve and have applied a low complexity ICM of 1.24 through 2018 then 1.19 thereafter. The resultant costs are shown in Table 3-43. Table 3-43 Costs of Electrical/Electro-hydraulic Power Steering (2010\$) Cost type DMC 1C TC 2017 \$87 \$22 \$109 2018 \$86 \$22 \$108 2019 \$84 \$18 \$101 2020 \$82 \$18 \$100 2021 \$80 \$18 \$98 2022 \$79 \$18 \$96 2023 \$77 \$18 \$95 2024 \$76 \$18 \$93 2025 \$74 \$18 \$92 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3.4.3.2 Improved Accessories The accessories on an engine, including the alternator, coolant and oil pumps are traditionally mechanically-driven. A reduction in CC>2 emissions and fuel consumption can be realized by driving them electrically, and only when needed ("on-demand"). Electric water pumps and electric fans can provide better control of engine cooling. For example, coolant flow from an electric water pump can be reduced and the radiator fan 3-113 ------- Technologies Considered in the Agencies' Analysis can be shut off during engine warm-up or cold ambient temperature conditions which will reduce warm-up time, reduce warm-up fuel enrichment, and reduce parasitic losses. Indirect benefit may be obtained by reducing the flow from the water pump electrically during the engine warm-up period, allowing the engine to heat more rapidly and thereby reducing the fuel enrichment needed during cold starting of the engine. Further benefit may be obtained when electrification is combined with an improved, higher efficiency engine alternator. Intelligent cooling can more easily be applied to vehicles that do not typically carry heavy payloads, so larger vehicles with towing capacity present a challenge, as these vehicles have high cooling fan loads. Both agencies also included a higher efficiency alternator in this category to improve the cooling system. The agencies considered whether to include electric oil pump technology for the rulemaking. Because it is necessary to operate the oil pump any time the engine is running, electric oil pump technology has insignificant effect on efficiency. Therefore, the agencies decided to not include electric oil pump technology for this final rule, consistent with the proposal. In MYs 2012-2016 final rule, the agencies used the effectiveness value in the range of 1 to 2 percent based on technologies discussed above. NHTSA did not apply this technology to large pickup truck due to the utility requirement concern for this vehicle class. For this final rule, consistent with the proposal, the agencies considered two levels of improved accessories. For level one of this technology (IACC1) NHTSA now incorporates a high efficiency alternator (70 percent efficiency). The second level of improved accessories (IACC2) adds the higher efficiency alternator and incorporates a mild regenerative alternator strategy, as well as intelligent cooling. NHTSA and EPA jointly reviewed the estimates of 1 to 2 percent effectiveness estimates used in the 2012-2016 final rule and TAR for level IACC1. More precisely, the agencies used effectiveness value in 1.2 to 1.8 percent range varying based on different vehicle subclasses. The incremental effectiveness for this technology in relative to EPS in the CAFE model is 0.91 to 1.61 percent. The combined effectiveness for IACC1 and IACC2 ranges from 3.1 to 3.9 percent and NHTSA applied incremental effectiveness of IACC2 in relative to IACC1 ranging from 1.74 to 2.55 percent. In the 2012-2016 rule, the agencies estimated the DMC of IACC1 at \$71 (2007\$). Converting to 2010\$, this DMC becomes \$75 for this analysis, applicable in the 2015MY, and consistent with the heavy-duty GHG rule. The agencies consider IACC1 technology to be on the flat portion of the learning curve and have applied a low complexity ICM of 1.24 through 2018 then 1.19 thereafter. Cost is higher for IACC2 due to the inclusion of a higher efficiency alternator and a mild level of regeneration. The agencies estimate the DMC of the higher efficiency alternator and the regeneration strategy at \$45 (2010\$) incremental to IACC1, applicable in the 2015MY. Including the costs for IACC1 results in a DMC for IACC2 of \$120 (2010\$) relative to the baseline case and applicable in the 2015MY. The agencies consider the IACC2 technology to be on the flat portion of the learning curve. The agencies have applied a low 3-114 ------- Technologies Considered in the Agencies' Analysis complexity ICM of 1.24 through 2018 then 1.19 thereafter. The resultant costs are shown in Table 3-44. Table 3-44 Costs for Improved Accessory Technology - Levels 1 & 2 (2010\$) Cost type DMC DMC 1C 1C TC TC IACC Technology IACC1 IACC2 IACC1 IACC2 IACC1 IACC2 2017 \$71 \$114 \$18 \$29 \$89 \$143 2018 \$70 \$112 \$18 \$29 \$88 \$141 2019 \$68 \$110 \$14 \$23 \$82 \$133 2020 \$67 \$107 \$14 \$23 \$81 \$131 2021 \$65 \$105 \$14 \$23 \$80 \$128 2022 \$64 \$103 \$14 \$23 \$78 \$126 2023 \$63 \$101 \$14 \$23 \$77 \$124 2024 \$62 \$99 \$14 \$23 \$76 \$122 2025 \$60 \$97 \$14 \$23 \$75 \$120 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Note that both levels of IACC technology are incremental to the baseline case. 3.4.3.3 Air Conditioner Systems We have a detailed description of the A/C program in Chapter 5 of this joint TSD. The reader is directed to that chapter to learn the specifics of the program, the credits involved, and details behind the costs we have estimated. Table 3-45 is a copy of Table 5-18 showing the total costs for A/C controls used in this final rule. Table 3-45 Total Costs for A/C Control Used in This Final rule (2010\$) Car/ Truck Car Truck Fleet Cost type TC TC TC TC TC TC TC Rule Reference Control Both Reference Control Both Both 2017 \$76 \$25 \$101 \$58 \$2 \$60 \$86 2018 \$75 \$40 \$115 \$57 \$46 \$103 \$111 2019 \$70 \$57 \$127 \$54 \$73 \$127 \$127 2020 \$69 \$65 \$134 \$53 \$82 \$134 \$134 2021 \$68 \$79 \$147 \$52 \$95 \$147 \$147 2022 \$67 \$77 \$144 \$51 \$93 \$144 \$144 2023 \$66 \$72 \$138 \$50 \$88 \$138 \$138 2024 \$65 \$71 \$135 \$49 \$86 \$135 \$135 2025 \$64 \$69 \$133 \$49 \$84 \$133 \$133 TC=Total cost 3.4.3.4 Stop-start (12V Micro Hybrid) The stop-start technology we consider for this final rule, consistent with the proposal—also known as idle-stop or 12-volt micro-hybrid—is the most basic hybrid system that facilitates idle-stop capability. When vehicle comes to a stop, the system will automatically shut down the internal combustion engine and restarts the engine when vehicle starts to move again. This is especially beneficial to reduce emission and fuel consumption when vehicle spends significant amount of time stopping in inner city driving or a traffic jam. Along with other enablers, this system typically replaces the standard 12-volt starter with an improved unit capable of higher power and increased cycle life. These systems typically incorporate an improved battery to prevent voltage-droop on restart. Different from MY 2012-2016 rule, this technology is applied to all vehicle classes, including large pickup truck. 3-115 ------- Technologies Considered in the Agencies' Analysis In MYs 2012-2016 final rule, even though EPA did not use 12 volt stop-start technology, NHTSA and EPA jointly reviewed the assumption. The effectiveness NHTSA used in the CAFE model for MYs 2012-2016 final rule ranged from 2 to 4 percent, depending on whether the vehicle is equipped with a 4-, 6- or 8-cylinder engine, with the 4-cylinder engine having the lowest range and the 8-cylinder having the highest. In this FRM analysis, consistent with the proposal, when combining IACC1, IACC2 and 12V stop-start system, the estimated effectiveness based on 2010 Ricardo study ranges from 4.8 percent to 5.9 percent. The agencies applied this effectiveness in the FRM analysis, consistent with the proposal. For CAFE modeling, the incremental effectiveness for 12V stop-start relative to IACC2 is 1.68 to 2.2 percent. Importantly, the effectiveness values presented here represent two-cycle effectiveness. Because stop-start technology provides considerable off-cycle benefits, both agencies apply a credit value to the technology. Off-cycle credits are discussed in Chapter 5 of this Joint TSD. In the 2012-2016 rule, the agencies estimated the DMC at \$282 (2007\$) to \$350 (2007\$) for small cars through large trucks, respectively. Converting to 2010\$, these DMCs become \$295 (2010\$) through \$367 (2010\$) for this analysis which are considered applicable in the 2015MY. The agencies consider 12V stop-start technology to be on the steep portion of the learning curve in the 2012-2016 timeframe and flat thereafter and have applied a medium complexity ICM of 1.39 through 2018 then 1.29 thereafter. The resultant costs are shown in Table 3-46. Table 3-46 EPA and NHTSA Costs for 12V Micro Hybrid or 12V Stop-Start (2010\$) Cost type DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC Vehicle Class Small car Standard car Large car Small MPV Large MPV Truck Small car Standard car Large car Small MPV Large MPV Truck Small car Standard car Large car Small MPV Large MPV Truck 2017 \$287 \$287 \$325 \$1 1 C 325 \$325 \$356 \$114 \$114 \$129 \$129 \$129 \$142 \$401 \$401 \$454 \$454 \$454 \$498 2018 \$278 \$278 \$315 \$315 \$315 \$346 \$114 \$114 \$129 \$129 \$129 \$141 \$392 \$392 \$444 \$444 \$444 \$487 2019 \$270 \$270 \$306 \$306 \$306 \$335 \$85 \$85 \$96 \$96 \$96 \$105 \$O e A 354 \$354 \$402 \$402 \$402 \$441 2020 \$261 \$261 \$296 \$296 \$296 \$325 \$85 \$85 \$96 \$96 \$96 \$105 \$346 \$346 \$392 \$392 \$392 \$430 2021 \$254 \$254 \$288 \$288 \$288 \$315 \$84 \$84 \$96 \$96 \$96 \$105 \$338 \$338 \$383 \$383 \$383 \$420 2022 \$246 \$246 \$279 \$279 \$279 \$306 \$84 \$84 \$95 \$95 \$95 \$105 \$330 \$o o f\ 330 \$O T /I 374 \$1 *7 A 374 \$374 \$410 2023 \$239 \$239 \$271 \$271 \$271 \$297 \$84 \$84 \$95 \$95 \$95 \$104 \$322 \$322 \$366 \$366 \$366 \$401 2024 \$232 \$232 \$262 \$262 \$262 \$288 \$84 \$84 \$95 \$95 \$95 \$104 \$315 \$315 \$357 \$357 \$357 \$392 2025 \$225 \$225 \$255 \$255 \$255 \$279 \$83 \$83 \$94 \$94 \$94 \$104 \$308 \$308 \$349 \$349 \$349 \$OOO JoJ DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3-116 ------- Technologies Considered in the Agencies' Analysis 3.4.3.5 Mild Hybrid Mild hybrid systems, also called Higher Voltage Stop-Start and Belt Mounted Integrated Starter Generator (BISG) systems are similar to a micro-hybrid system, offering idle-stop functionality, except that they utilize larger electric machine and a higher capacity battery, typically 42 volts or above, thus enabling a limited level of regenerative braking unavailable for a MHEV. The larger electric machine and battery also enables a limited degree of power assist, which MHEV cannot provide. However, because of the limited torque capacity of the belt-driven design, these systems have a smaller electric machine, and thus less capability than crank-integrated or stronger hybrid systems. These systems replace the conventional alternator with a belt-driven starter/alternator and may add high voltage electrical accessories (which may include electric power steering and an auxiliary automatic transmission pump). The limited electrical requirements of these systems allow the use of lead-acid batteries or supercapacitors for energy storage, or the use of a small lithium-ion battery pack, as is modeled in this analysis. While the mild hybrid system was not applied in the NPRM analysis because the agencies did not have solid information regarding its likely architecture, effectiveness or cost, the agencies are including the technology in the final rule because we now have good information about it. Further, the agencies are making available credits for mild hybrid pickup trucks in an effort to encourage such technologies. Lastly, the simulation modeling and cost estimation results show that the mild hybrid system could be a cost effective technology. For the BISG technology the agencies sized the system using a 15 kW starter/generator and 0.25 kWh Li-ion battery pack, which is similar to General Motors' eAssist BISG, which is available in MY 2012 Buick LaCrosse, Buick Regal, and Chevrolet Malibu vehicles. The agencies made this size system available to all vehicle subclasses, believing that manufacturers might use a similar strategy to control component complexity across the subclasses. As mentioned above, estimates were developed by ANL using Autonomie full vehicle simulation software. The absolute effectiveness for the CAFE analysis ranged from 8.5 to 11.6 percent depending on vehicle subclass. The effectiveness values include technologies that would be expected to incorporated with BISG which are stop/start (MHEV) and improved accessories (IACC1 and IACC2), however the effectiveness values do not include electric power steering (EPS). The costs for the mild hybrid technology are all new for this final rule and were developed in a manner consistent with costs generated for strong hybrids. These costs are presented in sections 3.4.3.7 through 3.4.3.10 of this Joint TSD. The same cost and effectiveness results were applied by both NHTSA and EPA. 3.4.3.5.1 Integrated Motor Assist (IMA)/Crank Integrated Starter Generator (CISC) CO IMA is a system developed and marketed by Honda and is similar to CISG. They both utilize a thin axial electric motor bolted to the engine's crankshaft and connected to the transmission through a torque converter or clutch. The axial motor is motor/generator that typically operates above 100 volts (but lower than the stronger hybrid systems discussed below, which typically operate at around 300 volts) and can provide sufficient torque for launch as well as generate sufficient current to provide significant levels of brake energy 3-117 ------- Technologies Considered in the Agencies' Analysis recovery. The motor/generator also acts as the starter for the engine and can replace a typical accessory-driven alternator. Current EVIA/CISG systems typically do not launch the vehicle on electric power alone, although some commercially available systems can cruise on electric power and dual-clutch IMA and CISG could be applied to all classes of vehicles. This technology is not used as an enabling technology in this FRM analysis, consistent with the proposal, by either EPA or NHTSA due to our expectation that manufacturers will be moving to more cost effective technologies. EPA relied on a combination of certification data (comparing vehicles available with and without a hybrid system and backing out other components where appropriate) and manufacturer-supplied information to determine that the effectiveness of these systems in terms of CC>2 reduction is 30 percent for small cars, 25 percent for large cars, and 20 percent for minivans and small trucks similar to the range estimated by NHTSA for the respective vehicle classes. The effectiveness for small cars assumes engine downsizing to maintain approximately equivalent performance. The large car, minivan, and small truck effectiveness values assume less engine downsizing in order to improve vehicle performance and/or maintain towing and hauling performance. In the 2012-2016 final rule, the agencies estimated the DMC at \$1,973, \$2,497, \$2,508, \$2,366 and \$3,063 (all values in 2007\$) for a small car, large car, minivan, small truck and large truck, respectively. For this final rule, the DMCs are \$2,070, \$2,620, \$2,631 and \$3,214 (all values in 2010\$) for small car/standard car, large car, small MPV and large MPV/truck. All of these DMCs are considered applicable in the 2015MY. The agencies consider the IMA technology to be on the steep portion of the learning curve and have applied a highl complexity ICM of 1.56 through 2018 then 1.35 thereafter. The resultant costs are as shown in Table 3-47. As noted earlier, the IMA technology is not included as an enabling technology in this analysis, although it is included as a baseline technology because it exists in the baseline fleet. The agencies moved away from this technology and applied P2 hybrids instead because P2 is more cost effective than IMA. Table 3-47 Costs for IMA Hybrids (2010\$) Cost type DMC DMC DMC DMC 1C 1C 1C 1C TC TC TC TC Vehicle Class Small car/Standard car Large car Small MPV Large MPV/Truck Small car/Standard car Large car Small MPV Large MPV/Truck Small car/Standard car Large car Small MPV Large MPV/Truck 2017 \$2,008 \$2,541 \$2,552 \$3,118 \$1,162 \$1,471 \$1,478 \$1,805 \$3,170 \$4,013 \$4,029 \$4,923 2018 \$1,947 \$2,465 \$2,475 \$3,024 \$1,159 \$1,467 \$1,473 \$1,799 \$3,106 \$3,932 \$3,948 \$4,823 2019 \$1,889 \$2,391 \$2,401 \$2,933 \$709 \$898 \$901 \$1,101 \$2,598 \$3,289 \$3,302 \$4,034 2020 \$1,832 \$2,319 \$2,329 \$2,845 \$707 \$895 \$899 \$1,098 \$2,540 \$3,215 \$3,228 \$3,944 2021 \$1,777 \$2,250 \$2,259 \$2,760 \$706 \$893 \$897 \$1,096 \$2,483 \$3,143 \$3,156 \$3,856 2022 \$1,724 \$2,182 \$2,191 \$2,677 \$704 \$891 \$895 \$1,093 \$2,428 \$3,073 \$3,086 \$3,770 2023 \$1,672 \$2,117 \$2,126 \$2,597 \$702 \$889 \$893 \$1,090 \$2,375 \$3,006 \$3,018 \$3,687 2024 \$1,622 \$2,053 \$2,062 \$2,519 \$701 \$887 \$891 \$1,088 \$2,323 \$2,940 \$2,952 \$3,607 2025 \$1,573 \$1,992 \$2,000 \$2,443 \$699 \$885 \$889 \$1,086 \$2,273 \$2,877 \$2,889 \$3,529 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3-118 ------- Technologies Considered in the Agencies' Analysis 3.4.3.6 HEV, PHEV, EV and Fuel Cell Vehicle Technologies A hybrid vehicle is a vehicle that combines two or more sources of propulsion energy, where one uses a consumable fuel (like gasoline), and one is rechargeable (during operation, or by another energy source). Hybrid technology is well established in the U.S. market and more manufacturers are adding hybrid models to their lineups. Hybrids reduce fuel consumption through three major mechanisms: • The internal combustion engine can be optimized (through downsizing, modifying the operating cycle, or other control techniques) to operate at or near its most efficient point more of the time. Power loss from engine downsizing can be mitigated by employing power assist from the secondary power source. • Some of the energy normally lost as heat while braking can be captured and stored in the energy storage system for later use. • The engine is turned off when it is not needed, such as when the vehicle is coasting or when stopped. Hybrid vehicles utilize some combination of the three above mechanisms to reduce fuel consumption and CC>2 emissions. A fourth mechanism to reduce petroleum fuel consumption, available only to plug-in hybrids, is by substituting the petroleum fuel energy with energy from another source, such as the electric grid. The effectiveness of fuel consumption and CC>2 reduction depends on the utilization of the above mechanisms and how aggressively they are pursued. One area where this variation is particularly prevalent is in the choice of engine size and its effect on balancing fuel economy and performance. Some manufacturers choose not to downsize the engine when applying hybrid technologies. In these cases, performance is vastly improved, while fuel efficiency improves significantly less than if the engine was downsized to maintain the same performance as the conventional version. While this approach has been used in cars such as the Lexus 600h luxury vehicle, it is more likely to be used in the future for vehicles like trucks where towing and/or hauling are an integral part of their performance requirements. In these cases, if the engine is downsized, the battery can be quickly drained during a long hill climb with a heavy load, leaving only a downsized engine to carry the entire load. Because towing capability is currently a heavily- marketed truck attribute, manufacturers are hesitant to offer a truck with downsized engine which can lead to a significantly diminished towing performance when the battery state of charge level is low, and therefore engines are traditionally not downsized for these vehicles. Although hybrid vehicles using other energy storage concepts (flywheel, hydraulic) have been developed, the automotive systems in production for passenger cars and light trucks are all hybrid electric vehicles (HEV) that use battery storage and electric drive systems. This appears likely to be the case for the foreseeable future. HEVs are part of a continuum of vehicles using systems with differing levels of electric drive and electric energy storage. This range of vehicles includes relatively basic system without electric energy storage such as engine start/stop systems; HEV systems with varying degrees of electric storage and electric drive system capability including mild-hybrid electric vehicles (MHEV) 3-119 ------- Technologies Considered in the Agencies' Analysis with limited capability but lower cost; strong hybrid electric vehicles (SHEV) with full hybridization capability such as the P2 hybrid technology which the agencies evaluate as a compliance option in this FRM; plug-in hybrid electric vehicles (PHEV) with differing degrees of all electric range and battery electric vehicles (EV) that rely entirely on electric drive and battery electric energy storage. Different HEV, PHEV and EV concepts utilize these mechanisms differently, so they are treated separately for the purposes of this analysis. In many applications, particularly with PHEV and EV, the battery represents the most costly and system-limiting sub-component of the hybrid system. Currently, there are many battery chemistries being developed and refined for hybrid applications that are expected to enhance the performance of future hybrid vehicles. Section 3.4.3.6.4 contains a discussion of battery energy storage and the major hybrid concepts that were determined to be available during the MY 2017-2015 timeframe. Fuel cell vehicles are a separate category of electric vehicle that rely entirely on electric propulsion with electricity produced on-board the vehicle using a proton-exchange- membrane fuel cell (PEMFC) fueled with hydrogen. Fuel cell vehicles under development are typically configured as a hybrid with battery storage used to provide brake energy recovery and improved response to fast transients in vehicle energy demand. 3.4.3.6.1 Power-split hybrid Power-split hybrid (PSHEV) - a hybrid electric drive system that replaces the traditional transmission with a single planetary gear set and two motor/generators. The smaller motor/generator uses the engine to either charge the battery or to supply additional power to the drive motor. The second, more powerful motor/generator is permanently connected to the vehicle's final drive and always turns with the wheels. The planetary gear splits engine power between the first motor/generator and the drive motor to either charge the battery or supply power to the wheels. Power-split hybrids are not used as an enabling technology in this final rule, consistent with the proposal. In MYs 2012-2016 final rule, EPA and NHTSA used a combination of manufacturer- supplied information and a comparison of vehicles available with and without a hybrid system from EPA's fuel economy test data to determine that the effectiveness is 19 to 36 percent for the classes to which it is applied. The estimate would depend on whether engine downsizing is also assumed. In the CAFE incremental model, the range of effectiveness used was 23 to 33 percent as engine downsizing is not assumed (and accounted for elsewhere). For this analysis, in order to estimate baseline costs, the agencies are using power-split HEV costs generated by FEV as part of a tear-down study. In that study, FEV found the DMC of the entire power-split system (battery-pack and non-battery components) to be \$2,853 (2007\$), \$3,175 (2007\$), \$3,435 (2007\$), \$4,168 (2007\$) for vehicle sized, for example, like a Ford Fiesta, Ford Focus, Ford Fusion and Ford Flex, respectively. For this analysis, these values become \$2,967, \$3,302, \$3,572 and \$4,335, respectively, all in 2010 dollars. In the 2012-2016 final rule, the agencies estimated the DMC of a large truck power- split system at \$5,137 (2007\$) which becomes \$5,391 for this analysis (2010\$) and we are using this value for the large MPV vehicle class. All of these DMCs are considered 3-120 ------- Technologies Considered in the Agencies' Analysis applicable in the 2015MY. The agencies consider the power-split technology to be on the flat portion of the learning curve and have applied a highl complexity ICM of 1.56 through 2018 then 1.35 thereafter. The resultant costs are as shown in Table 3-48. As noted earlier, the Power-split technology is not included as an enabling technology in this analysis, although it is included as a baseline technology because it exists in the baseline fleet. Table 3-48 Costs for Power-Split Hybrids (2010\$) Cost type DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C TC TC TC TC TC Vehicle Class Small car Standard car Large car Small MPV Large MP Small car Standard car Large car Small MPV Large MP Small car Standard car Large car Small MPV Large MP 2017 \$2,820 \$3,139 \$3,396 \$4,120 \$5,125 \$1,663 \$1,851 \$2,002 \$2,429 \$3,021 \$4,483 \$4,990 \$5,398 \$6,549 \$8,146 2018 \$2,764 \$3,076 \$3,328 \$4,038 \$5,023 \$1,659 \$1,846 \$1,998 \$2,424 \$3,015 \$4,423 \$4,923 \$5,326 \$6,462 \$8,037 2019 \$2,709 \$3,015 \$3,261 \$3,957 \$4,922 \$1,017 \$1,131 \$1,224 \$1,485 \$1,847 \$3,725 \$4,146 \$4,485 \$5,442 \$6,769 2020 \$2,655 \$2,954 \$3,196 \$3,878 \$4,824 \$1,015 \$1,129 \$1,222 \$1,483 \$1,844 \$3,669 \$4,084 \$4,418 \$5,361 \$6,668 2021 \$2,602 \$2,895 \$3,132 \$3,801 \$4,727 \$1,013 \$1,128 \$1,220 \$1,480 \$1,841 \$3,615 \$4,023 \$4,352 \$5,281 \$6,568 2022 \$2,549 \$2,837 \$3,070 \$3,725 \$4,633 \$1,012 \$1,126 \$1,218 \$1,478 \$1,838 \$3,561 \$3,963 \$4,288 \$5,202 \$6,471 2023 \$2,498 \$2,781 \$3,008 \$3,650 \$4,540 \$1,010 \$1,124 \$1,216 \$1,475 \$1,835 \$3,508 \$3,905 \$4,224 \$5,125 \$6,375 2024 \$2,449 \$2,725 \$2,948 \$3,577 \$4,449 \$1,008 \$1,122 \$1,214 \$1,473 \$1,832 \$3,457 \$3,847 \$4,162 \$5,050 \$6,281 2025 \$2,400 \$2,671 \$2,889 \$3,505 \$4,360 \$1,007 \$1,120 \$1,212 \$1,471 \$1,829 \$3,406 \$3,791 \$4,101 \$4,976 \$6,190 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3.4.3.6.2 2-mode hybrid 2-mode hybrid (2MHEV) - is a hybrid electric drive system that uses an adaptation of a conventional stepped-ratio automatic transmission by replacing some of the transmission clutches with two electric motors that control the ratio of engine speed to vehicle speed, while clutches allow the motors to be bypassed. This improves both the transmission torque capacity for heavy-duty applications and reduces fuel consumption and CO2 emissions at highway speeds relative to other types of hybrid electric drive systems. 2-mode hybrids were not been considered in the proposal and the agencies sought comments on whether or not 2- mode hybrids should be considered for vehicles with towing requirements, such as pickup trucks. However, no comments were received on their applicability in the future and thus consistent with the proposal, 2-mode hybrids were not included in the final rule analysis. For MYs 2012-2016 final rule, the CAFE model considered a range of 23 to 33 percent with a midpoint of 28 percent, assuming no engine downsizing to preserve the utility nature of medium and large trucks (e.g., maintaining full towing capability even in situations with low battery charge) and EPA estimates CC>2 emissions reduction effectiveness to be 25 percent for large trucks (LDT3 and LDT4 categories) based on vehicle certification data. EPA estimates an effectiveness of 40 percent for smaller vehicles. The agencies have estimated the costs for 2-mode hybrids using costs used in the 2010 TAR. For this analysis, the 2-mode battery pack DMC is estimated at \$1,100 (2010\$) and the 3-121 ------- Technologies Considered in the Agencies' Analysis DMC of non-battery components is estimated at \$2,997 (2010\$). The battery pack DMC is considered to be applicable for the 2025MY while the non-battery pack DMC would be applicable for the 2012MY. The agencies consider the 2-mode battery packs to be on the steep portion of the learning curve during the 2017-2025 timeframe. The agencies have applied a highl complexity ICM of 1.56 through 2018 then 1.35 thereafter. For 2-mode non- battery components, the agencies consider them to be on the flat portion of the learning curve in the 2017-2025 timeframe and have applied a highl complexity ICM of 1.56 through 2018 then 1.35 thereafter. The resultant 2-mode hybrid costs are presented in Table 3-49. Table 3-49 Costs for 2-Mode Hybrids (2010\$) Cost type Vehicle Class 2017 2018 2019 2020 2021 2022 2023 2024 2025 Battery-pack DMC 1C TC Small MPV/Large MPV/Truck \$2,148 \$688 \$2,835 \$1,718 \$660 \$2,378 \$1,718 \$399 \$2,118 \$1,374 \$389 \$1,763 \$1,374 \$389 \$1,763 \$1,374 \$389 \$1,763 \$1,374 \$389 \$1,763 \$1,374 \$389 \$1,763 \$1,100 \$380 \$1,479 Non-battery pack components DMC 1C TC Small MPV/Large MPV/Truck \$2,600 \$1,664 \$4,264 \$2,548 \$1,660 \$4,208 \$2,497 \$1,019 \$3,517 \$2,447 \$1,018 \$3,465 \$2,398 \$1,016 \$3,415 \$2,350 \$1,015 \$3,365 \$2,303 \$1,013 \$3,317 \$2,257 \$1,012 \$3,269 \$2,212 \$1,010 \$3,222 Battery -pack and non-battery pack components TC Small MPV/Large MPV/Truck \$7,099 \$6,586 \$5,634 \$5,228 \$5,178 \$5,128 \$5,080 \$5,032 \$4,702 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3.4.3.6.3 P2 Hybrid A P2 hybrid is hybrid technology that uses a transmission-integrated electric motor placed between the engine and a gearbox or CVT and coupled to the engine crankshaft via a clutch. The engine and the drive motor are mechanically independent of each other, allowing the engine or motor to power the vehicle separately or combined. Disengaging the engine clutch allows all-electric operation and more efficient brake-energy recovery. The P2 HEV system is similar to the Honda IMA HEV architecture with the exception of the added clutch, and larger batteries and motors. Examples of this include the Hyundai Sonata HEV and Infmiti M35h. The agencies believe that the P2 is an example of a "strong" hybrid technology that is typical of what will be prevalent in the timeframe of this rule. The agencies could have equally chosen the power-split architecture as the representative HEV architecture. These two HEV's have similar average effectiveness values (combined city and highway fuel economy), though the P2 systems may have lower cost due to having only a single, smaller motor/generator. For purposes of this rulemaking analysis, the agencies are assuming that P2 hybrids will become the dominant technology in the MYs 2017-2025 timeframe, replacing costlier power-split or 2-mode architectures while providing substantially similar efficiency improvement. At the present time, P2 hybrids are relatively new to the market and the agencies have not attempted to quantify any measurable performance differential between these technologies. As mentioned, the 2011 Hyundai Sonata, 2011 Volkswagen Touareg 3-122 ------- Technologies Considered in the Agencies' Analysis Hybrid, the 2011 Porsche S Hybrid, and the 2012 Infiniti M35 Hybrid are examples of P2 hybrids currently in production and available to consumers. The agencies are aware of some articles in trade journals, newspapers and other reviews that some first generation P2 hybrid vehicles with automatic transmissions have trade-offs in NVH and drivability - though these reviews do not cover all of the P2 systems available today, and a number of reviews are very positive with respect to NVH and drivability. The agencies recognize that manufacturers will have several years to test, develop and improve P2 technology in the years before 2017. We expect that manufacturers will address any perceived integration issues in early production models. However, we believe it is important to continue to monitor development of P2 hybrids and market acceptance of this technology. We will continue to gather information on these issues and consider them as part of the mid-term evaluation. The agencies requested comment regarding the potential of P2 hybrids to overcome these issues or others and we specifically sought comment from automakers developing and considering P2 technology on whether they believe these to be significant impediments to deployment and how they may be addressed. There were no comments submitted. The effectiveness used for vehicle packages with the P2-hybrid configuration within this analysis reflects a conservative estimate of system performance. Vehicle simulation modeling of technology packages using the P-2 hybrid has recently been completed under a contract with Ricardo Engineering. The agencies have updated the effectiveness of hybrid electric vehicle packages using the new Ricardo vehicle simulation modeling runs for this analysis. Due to the lower cost and comparative effectiveness of P2 hybrid in relative to other strong hybrid technologies, such as power-split hybrid and 2-mode hybrid, the agencies assume P2 hybrid application for all vehicle sub-classes in this FRM analysis, consistent with the proposal, and increased HEV effectiveness by approximately 2% comparing to 2012-2016 light duty GHG/CAFE final rule based on published data for new HEVs that have entered into production, such as 2011 Hyundai Sonata hybrid, 2010 Hyundai Elantra LPI HEV (Korean market only), 2011 Infiniti G35 Hybrid and 2011 Volkswagen Touareg Hybrid). In addition, for the Large Car, Minivan and Small Truck subclasses, the agencies further increased HEV effectiveness by assuming that towing capacity could be reduced from their current rating^ to approximately 1,500 pounds for some vehicles in these subclasses without significantly impacting consumers' need for utility in these vehicles.ss The agencies believe that consumers for these vehicles who require higher towing capacity could acquire it by purchasing a vehicle with a more capable non-hybrid powertrain (as they do today).tt " Current small SUVs and Minivans have an approximate average towing capacity of 2000 pounds (without a towing package), but range from no towing capacity to 3500 pounds. ss We note that there are some gasoline vehicles in the large car/minivan/small truck segments sold today which do not have any towing rating. tt The agencies recognize that assuming that certain consumers will choose to purchase non-hybrid vehicles in order to obtain their desired towing capacity could lead to some increase in fuel consumption and CO2 emissions as compared to assuming that towing capacity is maintained for hybrid vehicles across the board. However, the 3-123 ------- Technologies Considered in the Agencies' Analysis Moreover, it is likely that some fraction of consumers who purchase the larger engine option do so for purposes of hauling and acceleration performance, not just maximum towing. A reduction in towing capacity allows greater engine downsizing, which increases estimated overall HEV system incremental effectiveness by 5 to 10 percent for Large Cars, Minivans, and Small Trucks, similar to the HEV effectiveness value assumed for Small Cars and Compact Cars."11 Based on the recent Ricardo study, the effectiveness for P2 hybrid used in this FRM, consistent with the proposal, is 46.2 percent for subcompact and compact passenger cars, 48.6 percent for midsize passenger car, 49.4 percent for large passenger car, 46.1 percent for small light truck, 45.7 percent for midsize SUV, truck and minivan and 45.1 percent for large pickup truck. The process for battery sizing for the P2 hybrids is explained in Section 3.4.3.8. The battery sizing is different for the 2008 and 2010 baseline vehicle fleets, because vehicle mass for each subclass is slightly different between the two baseline fleets, thus requiring a slightly different battery size to maintain equivalent performance. The battery sizes with no applied mass reduction are listed in Table 3-50. Table 3-50 NHTSA Battery Sizes for P2 Hybrid Applied in Volpe Model without Mass Reduction (kWh) Baseline Fleet 2008 2010 Subcompact PC/PerfPC Compact PC/ PerfPC 0.81 0.84 Midsize PC/PerfPC 1.00 1.02 Large PC/Perf PC 1.16 1.20 Midsize LT Minivan 1.28 1.27 Small LT 1.04 1.06 Large LT 1.49 1.56 The agencies have applied a high complexity ICM to both the battery and non-battery component costs for P2 hybrid. But for battery for P2 hybrid, the ICM switches from short term value of 1.56 to long term value of 1.35 at 2024 while for the non-battery component the switch happens at 2018. The costs for P2 hybrids without mass reduction as used in the Volpe model are listed in Table 3-51. The battery costs are calculated using the battery sizes for both the 2008 and 2010 baseline fleets. NHTSA accounts the cost impact from the interaction between mass reduction and sizing of the electrification system (battery and non-battery system) as a cost agencies think it likely that the net improvement in fuel consumption and CO2 emissions due to the increased numbers of hybrids available for consumers to choose will offset any potential increase in fuel consumption and CO2 emissions resulting from consumers selecting the higher-performance non-hybrid powertrain vehicles. uu The effectiveness of HEVs for heavier vehicles which require conventional towing capabilities is markedly less because the rated power of the 1C engine must be similar to its non-hybrid brethren. As such, there is less opportunity for downsizing with these vehicles. 3-124 ------- Technologies Considered in the Agencies' Analysis synergy as described in section 3.4.3.9. Estimated costs for P2 HEVs with mass reduction as used in the OMEGA model are presented in Sections 3.4.3.9 and 3.4.3.10 below. Table 3-51 NHTSA Costs for P2 Hybrid Applied in Volpe Model without Mass Reduction (2010\$) Tech. Battery Battery Battery Battery Battery Battery Non- battery Non- battery Non- battery Non- battery Non- battery Non- battery Battery Battery Cost Type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C NHTSA Vehicle Class Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Midsize LT Minivan Small LT Large LT Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Midsize LT Minivan Small LT Large LT Subcompact PC/PerfPC Compact PC/PerfPC Midsize Baseline Fleet 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2017 \$733 \$726 \$818 \$809 \$959 \$946 \$887 \$885 \$796 \$787 \$1,029 \$1,020 \$1,474 \$1,468 \$1,645 \$1,627 \$1,949 \$1,906 \$1,817 \$1,798 \$1,587 \$1,557 \$1,918 \$1,901 \$413 \$409 \$461 2018 \$711 \$704 \$793 \$784 \$931 \$918 \$860 \$858 \$773 \$763 \$998 \$989 \$1,445 \$1,438 \$1,612 \$1,595 \$1,910 \$1,868 \$1,780 \$1,762 \$1,555 \$1,526 \$1,879 \$1,863 \$411 \$408 \$459 2019 \$689 \$683 \$769 \$761 \$903 \$890 \$834 \$832 \$749 \$740 \$968 \$960 \$1,416 \$1,410 \$1,580 \$1,563 \$1,872 \$1,830 \$1,745 \$1,727 \$1,524 \$1,496 \$1,842 \$1,825 \$410 \$406 \$458 2020 \$669 \$662 \$746 \$738 \$876 \$864 \$809 \$807 \$727 \$718 \$939 \$931 \$1,388 \$1,381 \$1,549 \$1,531 \$1,834 \$1,794 \$1,710 \$1,693 \$1,493 \$1,466 \$1,805 \$1,789 \$409 \$405 \$456 2021 \$648 \$642 \$724 \$716 \$849 \$838 \$785 \$783 \$705 \$697 \$911 \$903 \$1,360 \$1,354 \$1,518 \$1,501 \$1,798 \$1,758 \$1,676 \$1,659 \$1,464 \$1,436 \$1,769 \$1,753 \$407 \$404 \$455 2022 \$629 \$623 \$702 \$694 \$824 \$813 \$761 \$760 \$684 \$676 \$884 \$876 \$1,333 \$1,327 \$1,487 \$1,471 \$1,762 \$1,723 \$1,642 \$1,626 \$1,434 \$1,408 \$1,733 \$1,718 \$406 \$402 \$453 2023 \$610 \$604 \$681 \$674 \$799 \$788 \$739 \$737 \$663 \$655 \$857 \$850 \$1,306 \$1,300 \$1,457 \$1,441 \$1,727 \$1,688 \$1,609 \$1,593 \$1,406 \$1,380 \$1,699 \$1,684 \$405 \$401 \$452 2024 \$592 \$586 \$661 \$653 \$775 \$765 \$716 \$715 \$643 \$636 \$831 \$824 \$1,280 \$1,274 \$1,428 \$1,413 \$1,692 \$1,655 \$1,577 \$1,561 \$1,378 \$1,352 \$1,665 \$1,650 \$404 \$400 \$451 2025 \$574 \$569 \$641 \$634 \$752 \$742 \$695 \$693 \$624 \$617 \$807 \$799 \$1,254 \$1,249 \$1,400 \$1,384 \$1,658 \$1,621 \$1,546 \$1,530 \$1,350 \$1,325 \$1,631 \$1,617 \$248 \$246 \$277 3-125 ------- Technologies Considered in the Agencies' Analysis Battery Battery Battery Battery Non- battery Non- battery Non- battery Non- battery Non- battery Non- battery Battery Battery Battery Battery Battery Battery Non- battery 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC PC/PerfPC Large PC/PerfPC Midsize LT Minivan Small LT Large LT Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Midsize LT Minivan Small LT Large LT Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Midsize LT Minivan Small LT Large LT Subcompact PC/PerfPC Compact PC/PerfPC 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 \$456 \$541 \$533 \$500 \$499 \$449 \$443 \$580 \$575 \$943 \$939 \$1,053 \$1,041 \$1,247 \$1,219 \$1,162 \$1,150 \$1,015 \$996 \$1,227 \$1,216 \$1,145 \$1,135 \$1,278 \$1,264 \$1,500 \$1,480 \$1,386 \$1,383 \$1,245 \$1,230 \$1,609 \$1,595 \$2,418 \$2,407 \$454 \$539 \$532 \$498 \$497 \$447 \$442 \$578 \$573 \$941 \$937 \$1,050 \$1,039 \$1,244 \$1,217 \$1,160 \$1,148 \$1,013 \$994 \$1,224 \$1,213 \$1,122 \$1,111 \$1,252 \$1,239 \$1,469 \$1,449 \$1,358 \$1,355 \$1,220 \$1,205 \$1,576 \$1,562 \$2,386 \$2,375 \$453 \$537 \$530 \$496 \$495 \$446 \$440 \$576 \$571 \$578 \$575 \$645 \$638 \$764 \$747 \$712 \$705 \$622 \$610 \$752 \$745 \$1,099 \$1,089 \$1,227 \$1,213 \$1,440 \$1,420 \$1,331 \$1,327 \$1,195 \$1,181 \$1,544 \$1,531 \$1,994 \$1,985 \$451 \$535 \$528 \$495 \$494 \$444 \$439 \$574 \$569 \$577 \$574 \$644 \$637 \$763 \$746 \$711 \$704 \$621 \$610 \$751 \$744 \$1,077 \$1,067 \$1,202 \$1,189 \$1,411 \$1,392 \$1,304 \$1,301 \$1,171 \$1,157 \$1,513 \$1,500 \$1,965 \$1,956 \$450 \$534 \$526 \$493 \$492 \$443 \$438 \$572 \$567 \$576 \$574 \$643 \$636 \$762 \$745 \$710 \$703 \$620 \$609 \$749 \$743 \$1,056 \$1,046 \$1,179 \$1,166 \$1,383 \$1,364 \$1,278 \$1,275 \$1,148 \$1,134 \$1,483 \$1,470 \$1,936 \$1,927 \$448 \$532 \$525 \$492 \$490 \$442 \$436 \$571 \$566 \$575 \$573 \$642 \$635 \$761 \$744 \$709 \$702 \$619 \$608 \$748 \$742 \$1,035 \$1,025 \$1,155 \$1,143 \$1,356 \$1,337 \$1,253 \$1,250 \$1,125 \$1,112 \$1,454 \$1,442 \$1,908 \$1,899 \$447 \$530 \$523 \$490 \$489 \$440 \$435 \$569 \$564 \$574 \$572 \$641 \$634 \$759 \$743 \$708 \$701 \$618 \$607 \$747 \$741 \$1,015 \$1,006 \$1,133 \$1,121 \$1,330 \$1,311 \$1,229 \$1,226 \$1,104 \$1,090 \$1,426 \$1,414 \$1,881 \$1,872 \$446 \$529 \$522 \$489 \$488 \$439 \$434 \$567 \$562 \$574 \$571 \$640 \$633 \$758 \$741 \$707 \$700 \$617 \$606 \$746 \$739 \$996 \$986 \$1,111 \$1,099 \$1,304 \$1,286 \$1,205 \$1,202 \$1,082 \$1,069 \$1,399 \$1,386 \$1,854 \$1,845 \$274 \$325 \$320 \$300 \$299 \$270 \$266 \$348 \$345 \$573 \$570 \$639 \$632 \$757 \$740 \$706 \$699 \$616 \$605 \$745 \$738 \$822 \$814 \$918 \$907 \$1,077 \$1,062 \$995 \$993 \$894 \$883 \$1,155 \$1,145 \$1,827 \$1,819 3-126 ------- Technologies Considered in the Agencies' Analysis Non- battery Non- battery Non- battery Non- battery Non- battery TC TC TC TC TC Midsize PC/PerfPC Large PC/PerfPC Midsize LT Minivan Small LT Large LT 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 \$2,698 \$2,668 \$3,196 \$3,125 \$2,979 \$2,949 \$2,602 \$2,554 \$3,144 \$3,116 \$2,663 \$2,633 \$3,155 \$3,085 \$2,940 \$2,910 \$2,568 \$2,520 \$3,103 \$3,076 \$2,225 \$2,201 \$2,636 \$2,577 \$2,457 \$2,432 \$2,146 \$2,106 \$2,593 \$2,570 \$2,192 \$2,168 \$2,597 \$2,540 \$2,421 \$2,396 \$2,115 \$2,075 \$2,555 \$2,533 \$2,160 \$2,137 \$2,559 \$2,503 \$2,386 \$2,361 \$2,084 \$2,045 \$2,518 \$2,496 \$2,129 \$2,106 \$2,522 \$2,466 \$2,351 \$2,327 \$2,054 \$2,015 \$2,482 \$2,460 \$2,098 \$2,075 \$2,486 \$2,431 \$2,317 \$2,294 \$2,024 \$1,986 \$2,446 \$2,424 \$2,068 \$2,046 \$2,450 \$2,396 \$2,284 \$2,261 \$1,995 \$1,958 \$2,411 \$2,389 \$2,039 \$2,016 \$2,415 \$2,362 \$2,252 \$2,229 \$1,966 \$1,930 \$2,376 \$2,355 3.4.3.6.4 Plug-In Hybrid Plug-In Hybrid Electric Vehicles (PHEVs) are very similar to Hybrid Electric Vehicles, but with three significant functional differences. The first is the addition of a means to charge the battery pack from an outside source of electricity (e.g., the electric grid). Second, a PHEV would have a larger battery pack with more energy storage, and a greater capability to be discharged. Finally, a PHEV would have a control system that allows the battery pack to be significantly depleted during normal operation. Table 3-52 below, illustrates how PHEVs compare functionally to both hybrid electric vehicles (HEV) and electric vehicles (EV). These characteristics can change significantly within each vehicle class/subclass, so this is simply meant as an illustration of the general characteristics. In reality, the design options are so varied that all these vehicles exist on a continuum with HEVs on one end and EVs on the other. Table 3-52 Conventional, HEVs, PHEVs, and EVs Compared Attribute Drive Power Engine Size Electric Range Battery Charging Increasing Electrification Conventional Engine Full Size None None HEV Blended Engine/Electric Full Size or Smaller None to Very Short On-Board PHEV Blended Engine/Electric Smaller or Much Smaller Short to Medium Grid/On-Board EV Electric No Engine Medium to Long Grid Only Deriving some of their propulsion energy from the electric grid provides several advantages for PHEVs. PHEVs offer a significant opportunity to replace petroleum used for transportation energy with domestically-produced electricity. The reduction in petroleum usage does, of course, depend on the amount of electric drive the vehicle is capable of under its duty cycle. PHEVs also provide electric utilities the possibility to increase electric generation during "off-peak" periods overnight when there is excess generation capacity and 3-127 ------- Technologies Considered in the Agencies' Analysis electricity prices are lower. Utilities like to increase this "base load" because it increases overall system efficiency and lowers average costs. PHEVs can lower localized emissions of criteria pollutants and air toxics especially in urban areas by operating on electric power. The emissions from the power generation occur outside the urban area at the power generation plant which provides health benefits for residents of the more densely populated urban areas by moving emissions of ozone precursors out of the urban air shed. Unlike most other alternative fuel technologies, PHEVs can initially use an existing infrastructure for refueling (charging and liquid refueling) so investments in infrastructure may be reduced. In analyzing the impacts of grid-connected vehicles like PHEVs and EVs, the emissions from the electrical generation can be accounted for if a full upstream and downstream analysis is desired. While this issue is being studied on an on-going basis, upstream CC>2 emissions are not unique to grid-connected technologies and so are not included in this analysis. The respective agencies' RIAs and NHTSA's EIS have more information on upstream emissions. PHEVs will be considerably more costly than conventional vehicles and some other advanced technologies due to the fact that PHEVs require both conventional internal combustion engine and electrical driving system and the larger expensive battery pack. To take full advantage of their capability, consumers would have to be willing to charge the vehicles during electricity off-peak hours during the night, and would need access to electric power where they park their vehicles. For many urban dwellers who may park on the street, or in private or public lots or garages, charging may not be practical. Charging may be possible at an owner's place of work, but that would increase grid loading during peak hours which would eliminate some of the benefits to utilities of off-peak charging versus on-peak. Oil savings will still be the same in this case assuming the vehicle can be charged fully. The effectiveness potential of PHEVs depends on many factors, the most important being the energy storage capacity designed into the battery pack. To estimate the fuel consumption and tailpipe CC>2 reduction potential of PHEVs, EPA has developed an in-house vehicle energy model (PEREGRIN) to estimate the fuel consumption/CO2 emissions reductions of PHEVs. This model is based on the PERE (Physical Emission Rate Estimator) physics-based model used as a fuel consumption input for EPA's MOVES mobile source emissions model. How EPA Estimates PHEV Effectiveness The PHEV small car, large car, minivan and small trucks were modeled using parameters from a midsize car similar to today's hybrids and scaled to each vehicle's weight. The large truck PHEV was modeled separately assuming no engine downsizing. PHEVs can have a wide variation in the All Electric Range (AER) that they offer. Some PHEVs are of the "blended" type where the engine is on during most of the vehicle operation, but the proportion of electric energy that is used to propel the vehicle is significantly higher than that used in a PSHEV or 2MHEV. Each PHEV was modeled with enough battery capacity for a 20-mile-equivalent AER and a power requirement to provide similar performance to a hybrid vehicle. 20 miles was selected because it offers a good compromise for vehicle performance, 3-128 ------- Technologies Considered in the Agencies' Analysis weight, battery packaging and cost. Given expected near-term battery capability, a 20 mile range represents the likely capability that will be seen in PHEVs in the near-to-mid term. To calculate the total energy use of a PHEV, the PHEV can be thought of as operating in two distinct modes, electric (EV) mode, and hybrid (HEV) mode. At the tailpipe, the CC>2 emissions during EV operation are zero. The EV mode fuel economy can then be combined with the HEV mode fuel economy using the Utility Factor calculation in SAE J1711 to determine a total MPG value for the vehicle. (See Table 3-53) Table 3-53 Sample Calculation of PHEV Gasoline-Equivalent CO2 Reduction EV energy comb (0.55 city / 0.45 hwy) EV range (from PEREGRIN) SAE Jl 711 utility factor HEV mode comb FE (0.55 city / 0.45 hwy) Total UF-adjusted FE (UF*FCEV + (1-UF)*FCHEV) Baseline FE Percent FE gain Percent CO2 reduction Midsize Car 0.252 kwh/mi 20 miles 0.30 49.1 mpg 70.1 mpg 29.3 mpg 139% -58% Large Truck 0.429 kwh/mi 20 miles 0.30 25. 6 mpg 36.6 mpg 19.2 mpg 90% -47% Calculating a total fuel consumption and tailpipe CO2 reduction based on model outputs and the Utility Factor calculations results in a 58 percent reduction for small cars, large cars, minivans, and small trucks. For large trucks, the result is a 47 percent reduction. The lower improvement is due to less engine downsizing in the large truck class. How NHTSA Estimates PHEV Effectiveness For purposes of CAFE analysis, we assume that all future PHEVs during the rulemaking timeframe will meet the range requirements to qualify as a dual fuel vehicle. When calculating the fuel economy of a dual-fuel PHEV, NHTSA uses a petroleum equivalency factor for electricity consumption as stated in 49 U.S.C. 32904 and 32905. When deciding PHEV and EV effectiveness, NHTSA referenced the fuel economy of 3 pairs of vehicles for which NHTSA has fuel economy data in the CAFE database. These three vehicles pairs are MiniE electric vehicle versus gasoline powered Mini with automatic transmission, Tesla Roadster electric vehicle versus gasoline powered rear-wheel-drive Lotus Elise Sedan with a 6-speed manual transmission, and Nissan Leaf electric vehicle versus gasoline powered Nissan Sentra with automatic transmission. The fuel economy and fuel consumption for the first two pairs are shown in Table 3-54. Nissan Leaf information is used but not shown in the table because it is confidential information. Because technologies are applied in the CAFE model in an incremental manner, the effectiveness for each technology is incremental to the previous technology. In the electrification decision tree of the CAFE model, the order of technology selection starts from gasoline only powertrain, then moves to strong hybrid, to plug-in hybrid electric vehicle, and finally to electric vehicle. So the incremental effectiveness for each step has to be defined. 3-129 ------- Technologies Considered in the Agencies' Analysis Table 3-54 EV Fuel Economy and Fuel Consumption 104 mile range (Mini website) MiniE (mpg) Mini Gas ATX (mpg) 227 mile range (EPA) Tesla Roadster Lotus Elise sedan M6 RWD Fuel economy (mpg) 342.4 38.6 346.8 30.6 Fuel consumption (gPm) 0.0029206 0.0259067 0.0028835 0.0326797 In order to calculate the effectiveness of PHEV for purposes of a CAFE standard, fuel economy for strong hybrid electric vehicle (SHEV) is calculated first using the incremental effectiveness of strong hybrid from LPM model which is around 46 percent. For example, the derived fuel economy for SHEV based on Mini Gas ATX is 71.7 mpg. Then the fuel economy from gasoline source for PHEV is assumed to be the same as SHEV fuel economy, i.e. 71.7 mpg in the case of Mini E. The petroleum equivalent fuel economy from the electricity source is set to be the same as the EV fuel economy, e.g., 342.4 mpg in the case of Mini E. The combined fuel economy for PHEV is calculated using the 50-50 weighting factor as follows. Consistent with 49 U.S.C. 32904 and 32905, NHTSA is using a 50-50 weighting factor in the calculation above for CAFE model analysis of PHEV through 2019. After 2019, NHTSA will use the utility factor method defined by SAE standard J1711 for calculating CAFE fuel economy of PHEV. NHTSA expects that a PHEV with a 30 mile charge depleting range may reasonably represent the PHEVs that manufacturers may produce in MYs 2017 to 2025. According to SAE standard J2841, a vehicle with 30 mile charge depleting range has a 0.668 city specific utility factor and a 0.337 highway specific utility factor, which together give a 0.52 combined utility factor (55% city/45% highway split). Therefore NHTSA selected a PHEV with a 30 mile range for the CAFE model analysis, and the selection of a PHEV with a 30 mile range maintains continuity between pre-2020 and post-2020 PHEV fuel economy calculations. NHTSA assumes a 0.50 utility factor for MY2020 and beyond. In the FRM analysis, consistent with the proposal, EPA models a 20-mile range and a 40-mile range PHEV. The incremental fuel consumption reduction for PHEV is then calculated in relative to strong HEV. Using the example of Mini E, the incremental fuel consumption reduction for PHEV relative to SHEV is 39.5 percent as shown below. 3-130 ------- Technologies Considered in the Agencies' Analysis Table 3-55 lists the incremental effectiveness calculation for two pairs of vehicles, MiniE and Tesla Roaster. Incremental fuel consumption calculation for PHEV based on Nissan Leaf is not shown in Table 3-55 due to confidentiality of the fuel economy rating. The derived incremental effectiveness for Nissan Leaf is 40.6%. The average incremental effectiveness of these three pairs of vehicles is 40.65 percent which is used in CAFE modeling. Table 3-55 Incremental Effectiveness Calculation for purposes of CAFE modeling Mini E Combined Fuel Economy [mpg] Gasoline Fuel Economy [mpg] Electric Petroleum Equivalent Fuel Economy [mpg] Combined Fuel Consumption[gpm] Gasoline Fuel Consumption [gpm] Incremental Combined Fuel Consumption [%] Gasoline Weighing Factor[%] Electricity Weighing Factor [%] Gasoline 38.6 SHEV2 71.7 71.7 0.0139414 0.0139414 PHEV1 118.6 71.7 342.4 0.0084310 0.0139414 39.5% 50% 50% EV1 342.4 0.0029206 65.4% 0% 100% Tesla Combined Fuel Economy [mpg] Gasoline Fuel Economy [mpg] Electric Petroleum Equivalent Fuel Economy [mpg] Combined Fuel Consumption[gpm] Gasoline Fuel Consumption [gpm] Incremental Combined Fuel Consumption [%] Gasoline Weighing Factor[%] Gasoline 30.6 SHEV2 56.7 56.7 0.017647 0.017647 PHEV1 97.4 56.7 346.8 0.0102653 0.0176471 41.8% 50% EV1 346.8 0.0028835 71.9% 0% 3-131 ------- Technologies Considered in the Agencies' Analysis Electricity Weighing Factor [%] 50% 100% Once the fuel economy of the PHEV is calculated, the effectiveness of PHEV incremental to EV can be calculated similarly using the formula below. The average effectiveness for the three pairs of vehicles of 68.54% is used in CAFE modeling. The cost of PHEV consists of three parts, the cost for battery, the cost for non-battery systems and the cost for charger and the labor to install it. The battery sizing is calculated as in Section 3.4.3.8 and listed in Table 3-56. Costs for PHEVs without mass reduction as used in the Volpe model are listed in Table 3-57 to Table 3-61. NHTSA accounts the cost impact from the interaction between mass reduction and sizing of the electrification system (battery and non-battery system) as a cost synergy as described in section 3.4.3.9. Sections 3.4.3.9 and 3.4.3.10 contain the cost for PHEVs with mass reduction as used in EPA's OMEGA model. PHEV20 and PHEV40 are sized by EPA with the methodologies discussed in section 3.4.3.8. Table 3-56 NHTSA Battery Sizes for PHEV30 Hybrid Applied in Volpe Model without Mass Reduction (kWh) Baseline Fleet 2008 2010 Subcompact PC/PerfPC Compact PC/ PerfPC 10.42 10.81 Midsize PC/PerfPC 12.82 13.13 Large PC/Perf PC 15.21 15.79 Midsize LT Minivan 17.09 16.94 Small LT 13.48 13.69 Large LT 19.73 20.27 The battery pack DMCs for PHEV20 and PHEV40 are calculated using ANL's BatPaC model. NHTSA modeled a PHEV 30 for this final rule, for which NHTSA averaged the costs of PHEV20s and PHEV40s. The agencies have applied a high complexity ICM to non-battery component cost for PHEV and PHEV charger, which switch from short term value of 1.56 to long term value of 1.35 at 2018. The agencies applied a higher ICM factor to the battery of PHEV due to the fact that it a more complex technology. The ICM for PHEV battery switches from short term value of 1.77 to long term value of 1.50 at 2024. Table 3-57 NHTSA Costs Applied in Volpe Model for PHEV30 with No Mass Reduction (2010\$) 3-132 ------- Technologies Considered in the Agencies' Analysis Tech. Battery Battery Battery Non- battery Non- battery Non- battery Charger Charger Labor Battery Battery Battery Non- battery Non- battery Non- battery Charger Charger Labor Cost Type DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C DMC DMC DMC 1C 1C NHTSA Vehicle Class Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All Baseline Fleet 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 2017 \$6,208 \$6,095 \$7,415 \$7,251 \$9,835 \$9,610 \$2,586 \$2,522 \$3,252 \$3,132 \$4,685 \$4,494 \$210 \$1,010 \$2,671 \$2,622 \$3,190 \$3,119 \$4,231 \$4,134 \$1,655 \$1,614 \$2,081 \$2,004 \$2,997 \$2,875 \$67 \$0 2018 \$8,259 \$8,097 \$5,932 \$5,801 \$7,868 \$7,688 \$2,535 \$2,472 \$3,187 \$3,070 \$4,591 \$4,405 \$168 \$1,010 \$2,579 \$2,532 \$3,081 \$3,012 \$4,086 \$3,993 \$1,651 \$1,610 \$2,077 \$2,000 \$2,991 \$2,869 \$65 \$0 2019 \$8,259 \$8,097 \$5,932 \$5,801 \$7,868 \$7,688 \$2,484 \$2,422 \$3,124 \$3,008 \$4,499 \$4,316 \$168 \$1,010 \$2,579 \$2,532 \$3,081 \$3,012 \$4,086 \$3,993 \$1,014 \$989 \$1,275 \$1,228 \$1,836 \$1,762 \$65 \$0 2020 \$6,607 \$6,477 \$4,746 \$4,640 \$6,294 \$6,150 \$2,434 \$2,374 \$3,061 \$2,948 \$4,409 \$4,230 \$134 \$1,010 \$2,506 \$2,460 \$2,993 \$2,927 \$3,970 \$3,879 \$1,012 \$987 \$1,273 \$1,226 \$1,834 \$1,759 \$62 \$0 2021 \$6,607 \$6,477 \$4,746 \$4,640 \$6,294 \$6,150 \$2,386 \$2,326 \$3,000 \$2,889 \$4,321 \$4,145 \$134 \$1,010 \$2,506 \$2,460 \$2,993 \$2,927 \$3,970 \$3,879 \$1,011 \$986 \$1,271 \$1,224 \$1,831 \$1,756 \$62 \$0 2022 \$6,607 \$6,477 \$4,746 \$4,640 \$6,294 \$6,150 \$2,338 \$2,280 \$2,940 \$2,831 \$4,235 \$4,063 \$134 \$1,010 \$2,506 \$2,460 \$2,993 \$2,927 \$3,970 \$3,879 \$1,009 \$984 \$1,269 \$1,222 \$1,828 \$1,754 \$62 \$0 2023 \$6,607 \$6,477 \$4,746 \$4,640 \$6,294 \$6,150 \$2,291 \$2,234 \$2,881 \$2,775 \$4,150 \$3,981 \$134 \$1,010 \$2,506 \$2,460 \$2,993 \$2,927 \$3,970 \$3,879 \$1,008 \$983 \$1,267 \$1,220 \$1,825 \$1,751 \$62 \$0 2024 \$6,607 \$6,477 \$4,746 \$4,640 \$6,294 \$6,150 \$2,245 \$2,190 \$2,824 \$2,719 \$4,067 \$3,902 \$134 \$1,010 \$2,506 \$2,460 \$2,993 \$2,927 \$3,970 \$3,879 \$1,006 \$981 \$1,265 \$1,219 \$1,823 \$1,749 \$62 \$0 2025 \$5,286 \$5,182 \$3,797 \$3,712 \$5,035 \$4,920 \$2,200 \$2,146 \$2,767 \$2,665 \$3,986 \$3,824 \$108 \$1,010 \$1,578 \$1,550 \$1,885 \$1,844 \$2,501 \$2,444 \$1,005 \$980 \$1,264 \$1,217 \$1,820 \$1,746 \$37 \$0 3-133 ------- Technologies Considered in the Agencies' Analysis Battery Battery Battery Non- battery Non- battery Non- battery Charger Charger Labor TC TC TC TC TC TC TC TC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 \$8,878 \$8,717 \$10,605 \$10,370 \$14,066 \$13,744 \$4,241 \$4,136 \$5,333 \$5,136 \$7,682 \$277 \$277 \$1,010 \$7,545 \$7,408 \$9,013 \$8,813 \$11,954 \$11,681 \$4,186 \$4,082 \$5,264 \$5,069 \$7,582 \$233 \$233 \$1,010 \$7,545 \$7,408 \$9,013 \$8,813 \$11,954 \$11,681 \$3,498 \$3,411 \$4,399 \$4,236 \$6,336 \$233 \$233 \$1,010 \$6,479 \$6,361 \$7,739 \$7,567 \$10,264 \$10,030 \$3,446 \$3,361 \$4,334 \$4,174 \$6,243 \$197 \$197 \$1,010 \$6,479 \$6,361 \$7,739 \$7,567 \$10,264 \$10,030 \$3,396 \$3,312 \$4,271 \$4,113 \$6,152 \$197 \$197 \$1,010 \$6,479 \$6,361 \$7,739 \$7,567 \$10,264 \$10,030 \$3,347 \$3,264 \$4,209 \$4,054 \$6,063 \$197 \$197 \$1,010 \$6,479 \$6,361 \$7,739 \$7,567 \$10,264 \$10,030 \$3,299 \$3,217 \$4,148 \$3,995 \$5,975 \$197 \$197 \$1,010 \$6,479 \$6,361 \$7,739 \$7,567 \$10,264 \$10,030 \$3,251 \$3,171 \$4,089 \$3,938 \$5,890 \$197 \$197 \$1,010 \$4,757 \$4,670 \$5,682 \$5,556 \$7,536 \$7,364 \$3,205 \$3,126 \$4,031 \$3,882 \$5,806 \$145 \$145 \$1,010 3.4.3.6.5 Electric vehicles Electric vehicles (EV) - are vehicles with all-electric drive and with vehicle systems powered by energy-optimized batteries charged primarily from grid electricity. While the 2016 FRM did not anticipate a significant penetration of EVs, in this analysis, EVs with several ranges have been included. The GHG effectiveness is unchanged from estimates used for 2016 model year vehicles in the 2012-2016 final rule which is 100 percent GHG reduction. Per 49 U.S.C. 32904, NHTSA uses the Petroleum Equivalency Factor (PEF) in calculating the effectiveness for EVs as stated in the section above for PHEV. The PEF is determined by the U.S. Department of Energy as specified in 10 CFR Part 474. The PEF accounts for U.S. average fossil-fuel electricity generation and transmission efficiencies, petroleum refining and distribution efficiency, the energy content of gasoline, and includes a 0.15 divisor to incentivize the use of electricity in vehicles. The current PEF for electricity is 82.049 kWh per gallon of gasoline. Once the fuel economy of the PHEV is calculated as shown in the previous section, the effectiveness of PHEV incremental to EV can be calculated similarly using the formula below. 3-134 ------- Technologies Considered in the Agencies' Analysis The average effectiveness for the three pairs of vehicles of 68.54% is used in CAFE modeling. Battery costs assume that battery packs for EV applications will be designed to last for the full useful life of the vehicle at a useable state of charge equivalent to 80% of the nominal battery pack capacity. NHTSA included two levels of EVs, a 75-mile range EV and a 150- mile range EV in this FRM analysis, consistent with the proposal. As this technology is entering the market, it is expected that the OEMs will try to keep the cost low at the beginning so that there will be more penetration. Due to the high cost of the battery packs at this early stage of EVs, OEM will try to limit the battery pack size to reduce cost. Also the early adopters for this technology are normally urban drivers and range anxiety is not believed to be a big concern to them. Therefore NHTSA applied a 75-mile range EV for early adoption of this technology in the market, up to 5% penetration. As the technology develops and as the market penetration increases beyond 5%, NHTSA expects that OEMs would provide longer driving range to help the consumers overcome range anxiety. NHTSA applied 150-mile EV for this broad market adoption of this technology. The cost of an EV consists of three parts, cost of battery pack, cost of non-battery systems, and cost of charger and charger installation labor. An algorithm was used to select battery sizes. The algorithm is described in Section 3.4.3.8 and the battery sizes applied in the Volpe model for each type of EV and vehicle subclass are listed in Table 3-63. Table 3-58 NHTSA Battery Sizes for EVs Applied in Volpe Model with No Mass Reduction (kWh) EV75 EV100 EV150 Baseline Fleet 2008 2010 2008 2010 2008 2010 Subcompact PC/PerfPC Compact PC/ PerfPC 22.79 23.65 30.39 31.54 45.58 47.31 Midsize PC/PerfPC 28.03 28.72 37.38 38.30 56.07 57.45 Large PC/PerfPC 33.28 34.54 44.37 46.05 66.55 69.08 Midsize LT Minivan n/a n/a n/a n/a n/a n/a Small LT 29.48 29.95 39.30 39.94 58.96 59.90 Large LT n/a n/a n/a n/a n/a n/a 3-135 ------- Technologies Considered in the Agencies' Analysis The agencies have applied a high complexity ICM to non-battery component cost for EVs and EV chargers, which switch from short term value of 1.56 to long term value of 1.35 at 2018. The agencies applied a higher ICM factor to the battery of EVs due to the fact that it a more complex technology. The ICM for EV battery switches from short term value of 1.77 to long term value of 1.50 at 2024. The agencies present costs of EVs in Sections 3.4.3.9 and 3.4.3.10. The costs of EVs without mass reduction as applied in Volpe model are listed in Table 3-58 to Table 3-60. NHTSA accounts the cost impact from the interaction between mass reduction and sizing of electrification system (battery and non-battery system) as cost synergy as described in section 3.4.3.9. Table 3-59 NHTSA Costs for EV75 Applied in Volpe Model with No Mass Reduction (2010\$) Tech. Battery Battery Battery Non- battery Non- battery Non- battery Charger Charger Labor Battery Battery Battery Cost Type DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C NHTSA Vehicle Class Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Baseline Fleet 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 2010 2008 2010 2008 2010 2008 2017 \$10,324 \$10,121 \$12,140 \$11,881 \$15,634 \$15,238 \$410 \$354 \$1,267 \$1,156 \$2,236 \$2,080 \$395 \$1,010 \$4,441 \$4,354 \$5,222 \$5,111 \$6,725 \$6,555 2018 \$8,259 \$8,097 \$9,712 \$9,505 \$12,507 \$12,190 \$398 \$343 \$1,229 \$1,122 \$2,169 \$2,018 \$316 \$1,010 \$4,289 \$4,205 \$5,044 \$4,936 \$1,717 \$6,331 2019 \$8,259 \$8,097 \$9,712 \$9,505 \$12,507 \$12,190 \$386 \$333 \$1,193 \$1,088 \$2,104 \$1,957 \$316 \$1,010 \$4,289 \$4,205 \$5,044 \$4,936 \$1,712 \$6,331 2020 \$6,607 \$6,477 \$7,769 \$7,604 \$10,006 \$9,752 \$375 \$323 \$1,157 \$1,055 \$2,041 \$1,899 \$253 \$1,010 \$4,167 \$4,086 \$4,901 \$4,796 \$1,708 \$6,151 2021 \$6,607 \$6,477 \$7,769 \$7,604 \$10,006 \$9,752 \$363 \$313 \$1,122 \$1,024 \$1,980 \$1,842 \$253 \$1,010 \$4,167 \$4,086 \$4,901 \$4,796 \$1,703 \$6,151 2022 \$6,607 \$6,477 \$7,769 \$7,604 \$10,006 \$9,752 \$352 \$304 \$1,088 \$993 \$1,920 \$1,786 \$253 \$1,010 \$4,167 \$4,086 \$4,901 \$4,796 \$1,699 \$6,151 2023 \$6,607 \$6,477 \$7,769 \$7,604 \$10,006 \$9,752 \$345 \$298 \$1,067 \$973 \$1,882 \$1,751 \$253 \$1,010 \$4,167 \$4,086 \$4,901 \$4,796 \$1,696 \$6,151 2024 \$6,607 \$6,477 \$7,769 \$7,604 \$10,006 \$9,752 \$338 \$292 \$1,045 \$954 \$1,844 \$1,716 \$253 \$1,010 \$4,167 \$4,086 \$4,901 \$4,796 \$1,693 \$6,151 2025 \$5,286 \$5,182 \$6,215 \$6,083 \$8,005 \$7,802 \$332 \$286 \$1,024 \$935 \$1,808 \$1,681 \$202 \$1,010 \$2,625 \$2,573 \$3,087 \$3,021 \$1,090 \$3,874 3-136 ------- Technologies Considered in the Agencies' Analysis Non- battery Non- battery Non- battery Charger Charger Labor Battery Battery Battery Non- battery Non- battery Non- battery Charger Charger Labor DMC DMC DMC 1C 1C TC TC TC TC TC TC 1C 1C Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 \$316 \$272 \$976 \$890 \$1,722 \$1,602 \$126 \$0 \$14,765 \$14,475 \$17,362 \$16,992 \$22,359 \$21,793 \$726 \$626 \$2,243 \$2,047 \$3,959 \$3,682 \$521 \$1,010 \$315 \$272 \$973 \$888 \$18 \$1,597 \$121 \$0 \$12,548 \$12,302 \$14,755 \$14,441 \$19,002 \$18,521 \$713 \$615 \$2,203 \$2,010 \$3,887 \$3,615 \$437 \$1,010 \$314 \$271 \$970 \$885 \$18 \$1,593 \$121 \$0 \$12,548 \$12,302 \$14,755 \$14,441 \$19,002 \$18,521 \$700 \$603 \$2,163 \$1,973 \$3,817 \$3,550 \$437 \$1,010 \$313 \$270 \$968 \$883 \$18 \$1,588 \$117 \$0 \$10,775 \$10,563 \$12,670 \$12,400 \$16,317 \$15,903 \$688 \$593 \$2,125 \$1,938 \$3,749 \$3,487 \$370 \$1,010 \$313 \$269 \$965 \$881 \$18 \$1,584 \$117 \$0 \$10,775 \$10,563 \$12,670 \$12,400 \$16,317 \$15,903 \$676 \$582 \$2,087 \$1,904 \$3,683 \$3,426 \$370 \$1,010 \$312 \$269 \$963 \$878 \$18 \$1,580 \$117 \$0 \$10,775 \$10,563 \$12,670 \$12,400 \$16,317 \$15,903 \$664 \$572 \$2,051 \$1,871 \$3,619 \$3,367 \$370 \$1,010 \$311 \$268 \$961 \$877 \$18 \$1,578 \$117 \$0 \$10,775 \$10,563 \$12,670 \$12,400 \$16,317 \$15,903 \$657 \$566 \$2,028 \$1,850 \$3,578 \$3,328 \$370 \$1,010 \$311 \$268 \$960 \$876 \$18 \$1,575 \$117 \$0 \$10,775 \$10,563 \$12,670 \$12,400 \$16,317 \$15,903 \$649 \$559 \$2,005 \$1,829 \$3,538 \$3,291 \$370 \$1,010 \$200 \$172 \$618 \$563 \$10 \$1,014 \$70 \$0 \$7,911 \$7,755 \$9,302 \$9,104 \$11,980 \$11,676 \$532 \$458 \$1,642 \$1,498 \$2,897 \$2,695 \$272 \$1,010 Table 3-60 NHTSA Costs for EV100 Applied in Volpe Model with No Mass Reduction (2010\$) Tech. Battery Battery Cost Type DMC DMC NHTSA Vehicle Class Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Baseline Fleet 2010 2008 2010 2008 2017 \$12,341 \$12,063 \$14,159 \$13,919 2018 \$9,873 \$9,651 \$11,327 \$11,135 2019 \$9,873 \$9,651 \$11,327 \$11,135 2020 \$7,898 \$7,720 \$9,062 \$8,908 2021 \$7,898 \$7,720 \$9,062 \$8,908 2022 \$7,898 \$7,720 \$9,062 \$8,908 2023 \$7,898 \$7,720 \$9,062 \$8,908 2024 \$7,898 \$7,720 \$9,062 \$8,908 2025 \$6,319 \$6,176 \$7,250 \$7,127 3-137 ------- Technologies Considered in the Agencies' Analysis Battery Non- battery Non- battery Non- battery Charger Charger Labor Battery Battery Battery Non- battery Non- battery Non- battery Charger Charger Labor Battery Battery Battery Non- battery DMC DMC DMC DMC DMC DMC 1C 1C 1C DMC DMC DMC 1C 1C TC TC TC TC Large PC/PerfPC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Subcompact PC/PerfPC Compact 2010 2008 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 2010 2008 2010 2008 2010 2008 2010 \$17,482 \$17,025 \$410 \$354 \$1,267 \$1,156 \$2,236 \$2,080 \$395 \$1,010 \$5,309 \$5,189 \$6,091 \$5,988 \$7,520 \$7,324 \$316 \$272 \$976 \$890 \$1,722 \$1,602 \$126 \$0 \$17,650 \$17,253 \$20,250 \$19,907 \$25,002 \$24,349 \$726 \$13,985 \$13,620 \$398 \$343 \$1,229 \$1,122 \$2,169 \$2,018 \$316 \$1,010 \$5,127 \$5,012 \$5,883 \$5,783 \$7,263 \$7,073 \$315 \$272 \$973 \$888 \$1,717 \$1,597 \$121 \$0 \$15,000 \$14,662 \$17,210 \$16,918 \$21,248 \$20,693 \$713 \$13,985 \$13,620 \$386 \$333 \$1,193 \$1,088 \$2,104 \$1,957 \$316 \$1,010 \$5,127 \$5,012 \$5,883 \$5,783 \$7,263 \$7,073 \$314 \$271 \$970 \$885 \$1,712 \$1,593 \$121 \$0 \$15,000 \$14,662 \$17,210 \$16,918 \$21,248 \$20,693 \$700 \$11,188 \$10,896 \$375 \$323 \$1,157 \$1,055 \$2,041 \$1,899 \$253 \$1,010 \$4,982 \$4,870 \$5,716 \$5,619 \$7,057 \$6,873 \$313 \$270 \$968 \$883 \$1,708 \$1,588 \$117 \$0 \$12,880 \$12,590 \$14,778 \$14,527 \$18,245 \$17,769 \$688 \$11,188 \$10,896 \$363 \$313 \$1,122 \$1,024 \$1,980 \$1,842 \$253 \$1,010 \$4,982 \$4,870 \$5,716 \$5,619 \$7,057 \$6,873 \$313 \$269 \$965 \$881 \$1,703 \$1,584 \$117 \$0 \$12,880 \$12,590 \$14,778 \$14,527 \$18,245 \$17,769 \$676 \$11,188 \$10,896 \$352 \$304 \$1,088 \$993 \$1,920 \$1,786 \$253 \$1,010 \$4,982 \$4,870 \$5,716 \$5,619 \$7,057 \$6,873 \$312 \$269 \$963 \$878 \$1,699 \$1,580 \$117 \$0 \$12,880 \$12,590 \$14,778 \$14,527 \$18,245 \$17,769 \$664 \$11,188 \$10,896 \$345 \$298 \$1,067 \$973 \$1,882 \$1,751 \$253 \$1,010 \$4,982 \$4,870 \$5,716 \$5,619 \$7,057 \$6,873 \$311 \$268 \$961 \$877 \$1,696 \$1,578 \$117 \$0 \$12,880 \$12,590 \$14,778 \$14,527 \$18,245 \$17,769 \$657 \$11,188 \$10,896 \$338 \$292 \$1,045 \$954 \$1,844 \$1,716 \$253 \$1,010 \$4,982 \$4,870 \$5,716 \$5,619 \$7,057 \$6,873 \$311 \$268 \$960 \$876 \$1,693 \$1,575 \$117 \$0 \$12,880 \$12,590 \$14,778 \$14,527 \$18,245 \$17,769 \$649 \$8,951 \$8,717 \$332 \$286 \$1,024 \$935 \$1,808 \$1,681 \$202 \$1,010 \$3,138 \$3,067 \$3,600 \$3,539 \$4,445 \$4,329 \$200 \$172 \$618 \$563 \$1,090 \$1,014 \$70 \$0 \$9,457 \$9,244 \$10,850 \$10,666 \$13,396 \$13,046 \$532 3-138 ------- Technologies Considered in the Agencies' Analysis Non- battery Non- battery Charger Charger Labor TC TC 1C 1C PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All 2008 2010 2008 2010 2008 2008/2010 2008/2010 \$626 \$2,243 \$2,047 \$3,959 \$3,682 \$521 \$1,010 \$615 \$2,203 \$2,010 \$3,887 \$3,615 \$437 \$1,010 \$603 \$2,163 \$1,973 \$3,817 \$3,550 \$437 \$1,010 \$593 \$2,125 \$1,938 \$3,749 \$3,487 \$370 \$1,010 \$582 \$2,087 \$1,904 \$3,683 \$3,426 \$370 \$1,010 \$572 \$2,051 \$1,871 \$3,619 \$3,367 \$370 \$1,010 \$566 \$2,028 \$1,850 \$3,578 \$3,328 \$370 \$1,010 \$559 \$2,005 \$1,829 \$3,538 \$3,291 \$370 \$1,010 \$458 \$1,642 \$1,498 \$2,897 \$2,695 \$272 \$1,010 Table 3-61 NHTSA Costs for EV150 Applied in Volpe Model with No Mass Reduction (2010\$) Tech. Battery Battery Battery Non- battery Non- battery Non- battery Charger Charger Labor Battery Battery Battery Cost Type DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C NHTSA Vehicle Class Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large Baseline Fleet 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 2010 2008 2010 2008 2010 2017 \$16,369 \$15,939 \$19,585 \$19,240 \$22,552 \$21,936 \$410 \$355 \$1,267 \$1,157 \$2,236 \$2,082 \$395 \$1,010 \$7,042 \$6,857 \$8,425 \$8,277 \$9,702 2018 \$13,095 \$12,751 \$15,668 \$15,392 \$18,042 \$17,549 \$398 \$344 \$1,229 \$1,123 \$2,169 \$2,019 \$316 \$1,010 \$6,801 \$6,622 \$8,137 \$7,993 \$9,370 2019 \$13,095 \$12,751 \$15,668 \$15,392 \$18,042 \$17,549 \$386 \$334 \$1,193 \$1,089 \$2,104 \$1,959 \$316 \$1,010 \$6,801 \$6,622 \$8,137 \$7,993 \$9,370 2020 \$10,476 \$10,201 \$12,534 \$12,313 \$14,433 \$14,039 \$375 \$324 \$1,157 \$1,056 \$2,041 \$1,900 \$253 \$1,010 \$6,608 \$6,434 \$7,906 \$7,767 \$9,104 2021 \$10,476 \$10,201 \$12,534 \$12,313 \$14,433 \$14,039 \$363 \$314 \$1,122 \$1,025 \$1,980 \$1,843 \$253 \$1,010 \$6,608 \$6,434 \$7,906 \$7,767 \$9,104 2022 \$10,476 \$10,201 \$12,534 \$12,313 \$14,433 \$14,039 \$352 \$305 \$1,088 \$994 \$1,920 \$1,788 \$253 \$1,010 \$6,608 \$6,434 \$7,906 \$7,767 \$9,104 2023 \$10,476 \$10,201 \$12,534 \$12,313 \$14,433 \$14,039 \$345 \$299 \$1,067 \$974 \$1,882 \$1,752 \$253 \$1,010 \$6,608 \$6,434 \$7,906 \$7,767 \$9,104 2024 \$10,476 \$10,201 \$12,534 \$12,313 \$14,433 \$14,039 \$338 \$293 \$1,045 \$954 \$1,844 \$1,717 \$253 \$1,010 \$6,608 \$6,434 \$7,906 \$7,767 \$9,104 2025 \$8,381 \$8,161 \$10,028 \$9,851 \$11,547 \$11,231 \$332 \$287 \$1,024 \$935 \$1,808 \$1,682 \$202 \$1,010 \$4,162 \$4,053 \$4,980 \$4,892 \$5,734 3-139 ------- Technologies Considered in the Agencies' Analysis Non- battery Non- battery Non- battery Charger Charger Labor Battery Battery Battery Non- battery Non- battery Non- battery Charger Charger Labor DMC DMC DMC 1C 1C TC TC TC TC TC TC 1C 1C PC/PerfPC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC Subcompact PC/PerfPC Compact PC/PerfPC Midsize PC/PerfPC Large PC/PerfPC All All 2008 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2010 2008 2008/2010 2008/2010 \$9,437 \$316 \$273 \$976 \$891 \$1,722 \$1,603 \$126 \$0 \$23,411 \$22,796 \$28,010 \$27,517 \$32,254 \$31,372 \$726 \$628 \$2,243 \$2,048 \$3,959 \$3,685 \$521 \$1,010 \$9,114 \$315 \$273 \$973 \$889 \$1,717 \$1,598 \$121 \$0 \$19,896 \$19,373 \$23,805 \$23,385 \$27,411 \$26,662 \$713 \$617 \$2,203 \$2,011 \$3,887 \$3,618 \$437 \$1,010 \$9,114 \$314 \$272 \$970 \$886 \$1,712 \$1,594 \$121 \$0 \$19,896 \$19,373 \$23,805 \$23,385 \$27,411 \$26,662 \$700 \$606 \$2,163 \$1,975 \$3,817 \$3,552 \$437 \$1,010 \$8,855 \$313 \$271 \$968 \$884 \$1,708 \$1,590 \$117 \$0 \$17,084 \$16,635 \$20,441 \$20,080 \$23,537 \$22,894 \$688 \$595 \$2,125 \$1,940 \$3,749 \$3,489 \$370 \$1,010 \$8,855 \$313 \$270 \$965 \$881 \$1,703 \$1,585 \$117 \$0 \$17,084 \$16,635 \$20,441 \$20,080 \$23,537 \$22,894 \$676 \$585 \$2,087 \$1,906 \$3,683 \$3,428 \$370 \$1,010 \$8,855 \$312 \$270 \$963 \$879 \$1,699 \$1,581 \$117 \$0 \$17,084 \$16,635 \$20,441 \$20,080 \$23,537 \$22,894 \$664 \$574 \$2,051 \$1,873 \$3,619 \$3,369 \$370 \$1,010 \$8,855 \$311 \$269 \$961 \$878 \$1,696 \$1,579 \$117 \$0 \$17,084 \$16,635 \$20,441 \$20,080 \$23,537 \$22,894 \$657 \$568 \$2,028 \$1,852 \$3,578 \$3,330 \$370 \$1,010 \$8,855 \$311 \$269 \$960 \$876 \$1,693 \$1,576 \$117 \$0 \$17,084 \$16,635 \$20,441 \$20,080 \$23,537 \$22,894 \$649 \$561 \$2,005 \$1,831 \$3,538 \$3,293 \$370 \$1,010 \$5,578 \$200 \$173 \$618 \$564 \$1,090 \$1,014 \$70 \$0 \$12,543 \$12,214 \$15,007 \$14,743 \$17,281 \$16,809 \$532 \$460 \$1,642 \$1,499 \$2,897 \$2,697 \$272 \$1,010 3.4.3.6.6 Fuel cell electric vehicles Fuel cell electric vehicles (FCEVs) - utilize a full electric drive platform but consume electricity generated by an on-board fuel cell and hydrogen fuel. Fuel cells are electro- chemical devices that directly convert reactants (hydrogen and oxygen via air) into electricity, with the potential of achieving more than twice the efficiency of conventional internal combustion engines. High pressure gaseous hydrogen storage tanks are used by most automakers for FCEVs that are currently under development. The high pressure tanks are similar to those used for compressed gas storage in more than 10 million CNG vehicles 3-140 ------- Technologies Considered in the Agencies' Analysis worldwide, except that they are designed to operate at a higher pressure (350 bar or 700 bar vs. 250 bar for CNG). Due to the uncertainty of the future availability for this technology, FCEVs were not included in any OMEGA or Volpe model runs. 3.4.3.7 Batteries for Mild HEV, HEV, PHEV and EV Applications The design of battery secondary cells can vary considerably between Stop/Start, Mild HEV (ISO), HEV, PHEV and EV applications. MHEV batteries: Due to their lower voltage (12-42 VDC) and reduced power and energy requirements, MHEV systems may continue to use lead-acid batteries even long term (2017 model year and later). MHEV battery designs differ from those of current starved- electrolyte (typical maintenance free batteries) or flooded-electrolyte (the older style lead-acid batteries requiring water "top-off) batteries used for starting, lighting and ignition (SLI) in automotive applications. Standard SLI batteries are primarily designed to provide high- current for engine start-up and then recharge immediately after startup via the vehicle's charging system. Deeply discharging a standard SLI battery will greatly shorten its life. MHEV applications are expected to use: • Extended-cycle-life flooded (ELF) lead-acid batteries • Absorptive glass matt, valve-regulated lead-acid (AGM/VRLA) batteries -or - • Asymmetric lead-acid battery/capacitor hybrids (e.g., flooded ultrabatteries) MHEV systems using electrolytic double-layer capacitors are also under development and may provide improved performance and reduced cost in the post-2017 timeframe. Mild HEV and HEV batteries: Mild HEV and HEV applications operate in a narrow, short-cycling, charge-sustaining state of charge (SOC). Energy capacity in Mild HEV and HEV applications is somewhat limited by the ability of the battery and power electronics to accept charge and by space and weight constraints within the vehicle design. Mild HEV and HEV battery designs tend to be optimized for high power density rather than high energy density, with thinner cathode and anode layers and more numerous current collectors and separators (Figure 3-20). EV batteries: EV batteries tend to be optimized for high energy density and are considerably larger and heavier than HEV batteries in order to provide sufficient energy capacity. EV battery cells tend to have thicker cathode and anode layers and fewer collectors and separators than HEV cells. This reduces the specific cost on a per-kWh basis for EV battery cells relative to HEV battery cells. PHEV batteries: PHEV battery designs are intermediate between power-optimized HEV and energy-optimized EV battery cell designs. PHEV batteries must provide both charge depleting operation similar to an EV and charge sustaining operation similar to an HEV. Unlike HEV applications, charge-sustaining operation with PHEVs occurs at a relatively low battery state of charge (SOC) which can pose a significant challenge with 3-141 ------- Technologies Considered in the Agencies' Analysis respect to attaining acceptable battery cycle life. In the case of the GM Volt, this limits charge depleting operation to a minimum SOC of approximately 30 percent.59 An alternative approach for PHEV applications that has potential to allow extension of charge depletion to a lower battery SOC is using energy-optimized lithium-ion batteries for charge depleting operation in combination with the use of supercapacitors or power-optimized batteries for charge sustaining operation.60 Figure 3-20: Schematic representation of power and energy optimized prismatic-layered battery cells Collector (-) Cathode (-) Separator Anode (+) Collector (+) HEV Power-optimized Battery Cell EV Energy-optimized Battery Cell Power-split hybrid vehicles from Toyota, Ford and Nissan, integrated motor assist hybrid vehicles from Honda and the GM 2-mode hybrid vehicles currently use nickel-metal hydride (NiMH) batteries. Lithium-ion (Li-ion) batteries offer the potential to approximately double both the energy and power density relative to current NiMH batteries, enabling much more electrification of automotive drive applications such as PHEVs and EVs. Li-ion batteries for high-volume automotive applications differ substantially from those used in consumer electronics applications with respect to cathode chemistry, construction and cell size. Li-ion battery designs currently in production by CPI (LG-Chem) for the GM Volt PHEV and by AESC and GS-Yuasa (respectively) for the Nissan Leaf and Mitsubishi iMiEV use large-format, layered-prismatic cells assembled into battery modules. The modules are then combined into battery packs. Two families of cathode chemistries are used in large-format, automotive Li-ion batteries currently in production - LiMn2O4-spinel (CPI, GS-Yuasa, AESC) and LiFePO4 (A123 Systems). Current production batteries typically use graphite anodes. Automotive Li- 3-142 ------- Technologies Considered in the Agencies' Analysis ion batteries using lithium nickel manganese cobalt (NMC) oxide cathodes with graphite anodes are in advanced stages of development for PHEV and EV applications. The agencies expect large-format Li-ion batteries to completely replace NiMH batteries for post-2017 HEV applications. We also expect that stacked and/or folded prismatic Li-ion cell designs will continue to be used for PHEV and EV applications and that NMC/graphite Li-ion batteries will be a mature technology for 2017-2025 light-duty vehicle applications. 3.4.3.8 HEV, PHEV and EV System Sizing Methodology Battery packs are (and will continue to be) one of the most expensive components for EVs, PHEVs and HEVs. To obtain reasonable cost estimates for electrified vehicles, it was therefore important to establish a reliable approach for determining battery attributes for each vehicle and class. Both battery energy content ("size") and power rating are key inputs used to establish costs per ANL's battery costing model. For EVs and PHEVs in particular, battery size and weight are closely related, and so battery weight must be known as well. The following section details the steps taken to size a battery for a) EVs and PHEVs (at various all-electric ranges), b) a more simplified separate approach for MHEVs and HEVs. 3.4.3.8.1 Battery Pack Sizing for EVs and PHEVs Calculation of required battery pack energy requirements for EVs and PHEVs is not straightforward. Because vehicle energy consumption is strongly dependent on weight, and battery packs are very heavy, the weight of the battery pack itself can change the energy required to move the vehicle. As vehicle energy consumption increases, the battery size must increase for a given range (in the case of EVs and PHEVs) - as a result, vehicle weight increases, and per-mile energy consumption increases as well, increasing the battery size, and soon. EPA built spreadsheets to estimate the required battery size for each vehicle and class. Listed below are the steps EPA has taken in these spreadsheets to estimate not only battery size, but associated weight for EVs and PHEVs of varying ranges and designs. 1. Establish baseline FE/energy consumption 2. Assume nominal weight of electrified vehicle (based on weight reduction target) 3. Calculate vehicle energy demand at this target weight 4. Calculate required battery energy 5. Calculate actual battery and vehicle weight 6. Do vehicle weight and battery size match estimated values? 3-143 ------- Technologies Considered in the Agencies' Analysis Steps 2-6 were iterated until each assumed weight reduction target (and nominal vehicle weight) reconciled with required battery size and the calculated weight of each vehicle. Baseline vehicle energy consumption is estimated based on a fitted trendline for FE vs. inertia weight, or ETW (from FE Trends data for 2008 MY vehicles, table M-80) and converting to Wh/mi. This is shown in Figure 3-21. 2008 Fuel Economy vs. Inertia Weight (source: Fuel Economy Trends Report, Table M-80) si £ O 20 y - O.OOOOOlSOx2 - 0.02194637X + 85.81284974 Rz= 0.99565332 Inertia wt (Ibs) Figure 3-21: Average fuel economy based on inertia weight (ETW) from FE Trends data Then, fuel economy was converted into energy consumption (assuming 33700 Wh energy in 1 gallon of gasoline) and used to populate a range of test weights between 2000 and 6000 Ibs. A linear trendline was used to fit this curve and then applied to estimate generic energy consumption for baseline vehicles of a given ETW (shown below in Figure 3-22). 3-144 ------- Technologies Considered in the Agencies' Analysis 2008 Energy Consumption vs. Inertia Weight IS Inertia wt (Ibs) Figure 3-22: Equivalent energy consumption (in Wh/mi) for baseline vehicles To calculate battery pack size, the electrified vehicle weight must first be known; to calculate vehicle weight, the battery pack size must first be known. This circular reference required an iterative solution. EPA assumed a target vehicle glider (a rolling chassis with no powertrain) weight reduction and applied that to the baseline curb weight. The resulting nominal vehicle weight was then used to calculate the vehicle energy demand. To calculate the energy demand (efficiency) of an electric vehicle in Wh/mi, the following information was needed: • Baseline energy consumption / mpg • Efficiency (r|) improvement of electric vehicle • Change in road loads In Table 3-62 below, the following definitions apply: • Brake eff (brake efficiency) - the % amount of chemical fuel energy converted to energy at the engine crankshaft (or, for batteries, the amount of stored electrical energy converted to shaft energy entering the transmission) • D/L eff (driveline efficiency) - the % of the brake energy entering the transmission delivered through the driveline to the wheels • Wheel eff (wheel efficiency) - the product of brake and driveline efficiency • Cycle eff (cycle efficiency) - the % of energy delivered to the wheels used to overcome road loads and power the vehicle (it does not include energy lost as braking heat) • Vehicle efficiency - the product of wheel and cycle efficiency 3-145 ------- Technologies Considered in the Agencies' Analysis • Road loads - the amount of resistant energy the vehicle must overcome during a city/highway test. Composed of vehicle weight (inertia), aerodynamic drag and rolling resistance Table 3-62: EV100 efficiency and energy demand calculations, 20% applied weight reduction Class Baseline gas ICE Small car Std car Large car Small MPV Large MPV Truck Brake eff 24% 85% 85% 85% 85% 85% 85% D/L eff 81% 93% 93% 93% 93% 93% 93% Wheel eff 20% 79% 79% 79% 79% 79% 79% Cycle eff 77% 97% 97% 97% 97% 97% 97% Vehicle eff 15% 77% 77% 77% 77% 77% 77% Road loads 100% 88% 88% 88% 89% 89% 88% Energy reduction 83% 83% 83% 83% 83% 83% Energy eff increase 478% 478% 478% 475% 475% 482% IW- based, base ICE nominal mpgge 37 30 25 29 23 20 Base fuel energy req'd Wh/mi 912 1122 1332 1180 1497 1727 FTP fuel energy req'd Wh/mi 158 194 230 205 260 297 Onroad fuel energy req'd Wh/mi 225 277 329 293 372 424 The energy efficiency of a baseline vehicle (around 15%), as indicated in the table above, was estimated using efficiency terms derived from EPA's lumped parameter model (engine/battery brake efficiency, driveline efficiency, cycle efficiency and road load ratio to baseline). To calculate the energy consumption of an EV (or PHEV in charge-depleting mode), the following assumptions were made: • "Brake" efficiency (for an EV, the efficiency of converting battery energy to tractive energy at the transmission input shaft) was estimated at 85% - assuming, roughly a 95% efficiency for the battery, motor, and power electronics, respectively. • The driveline efficiency (including the transmission) was comparable to the value calculated by the lumped parameter model for an advanced 6-speed dual-clutch transmission at 93%. • The cycle efficiency assumes regenerative braking where 97% recoverable braking energy is recaptured. As a result, most of the energy delivered to the wheels is used to overcome road loads. • The road loads were based on the weight reduction of the vehicle. In the case of a 100 mile EV with a 20% weight reduction, road loads (as calculated by the LP model) are reduced to 88-89% of the baseline vehiclew. ' Included in this example road load calculation is a 10% reduction in rolling resistance and aerodynamic drag. 3-146 ------- Technologies Considered in the Agencies' Analysis The energy consumption of the EV includes ratio of the roadloads of the EV to the baseline vehicle, and the ratio of the efficiency of the EV compared to the baseline vehicle. It is expressed mathematically as shown below in Equation 3-1: EV energy consumption: Equation 3-1: EV energy consumption In Table 3-63, the baseline energy required (in Wh/mi) is in the column labeled "Base fuel energy reqd". The energy required for each vehicle class EV over the FTP is in the column "FTP fuel energy reqd Wh/mi" and incorporates the equation above. This energy rate refers to the laboratory or unadjusted test cycle value, as opposed to a real-world "onroad" value. EPA assumes a 30% fuel economy shortfall, based loosely on the 5-cycle Fuel Economy Labeling Rule (year) which is directionally correct for electrified vehicles. This corresponds to an increase in fuel consumption of 43%. Applying this 43% increase gives the onroad energy consumption values for EVs as shown in the far right column of the previous table. From this value, one can determine an appropriate battery pack size for the vehicle. The required battery energy for EVs equals the onroad energy consumption, multiplied by the desired range, divided by the useful state-of-charge window of the battery. It is calculated as follows in Equation 3-2 Equation 3-2: Required battery pack energy (size) for EVs Assumed usable SOC (battery state-of-charge) windows were 80% for EVs (10-90%) and 70% for PHEVs (15%-85%). The battery pack sizes are listed in orange in Table 3-63 for the 100-mile EV case and show both the onroad energy consumption ("EV adj Wh/mi" column) and the nominal battery energy content or "battery pack size". 3-147 ------- Technologies Considered in the Agencies' Analysis Table 3-63: Battery pack sizes for EV100 based on inertia weight, 20% applied weight reduction Class Baseline curb wt (Ib) Inertia wt (Ib) EV unadj (Wh/mi) EVadj (Wh/mi) 100 mile batt pack size (kWh) 2008 Baseline Small car Std car Large car Small MPV Large MPV Truck 2633 3306 3897 3474 4351 5108 2933 3606 4197 3774 4651 5408 158 194 230 205 260 297 225 277 329 293 372 424 28.2 34.7 41.1 36.7 46.5 53.0 20 10 Baseline Small car Stdcar Large car Small MPV Large MPV Truck 2753 3387 4035 3528 4313 5346 3053 3687 4335 3828 4613 5646 164 200 241 209 257 307 234 286 344 298 367 439 29.2 35.7 43.0 37.3 45.8 54.8 EPA used the following formula to determine weight of an EV (Equation 3-3): Equation 3-3: EV weight calculation Any weight reduction technology was applied only to the glider (baseline vehicle absent powertrain) as defined in Equation 3-4. Equation 3-4: Weight reduction of the glider In the case of a PHEV, it was assumed that the base ICE powertrain remains so it is not deducted; the proper equation for PHEVs is shown in Equation 3-5. Equation 3-5: Weight calculation for PHEV class: Listed in Table 3-64 are the assumed baseline ICE-powertrain weights, by vehicle 3-148 ------- Technologies Considered in the Agencies' Analysis Table 3-64: Baseline ICE-powertrain weight assumptions, by class Class Small car Std car Large car Small MPV Large MPV Truck Engine 250 300 375 300 400 550 Trans (diffnot included) 125 150 175 150 200 200 Fuel sys (50% fill) 50 60 70 60 80 100 Engine mounts/ NVH treatments 25 25 25 25 25 25 Exhaust 20 25 30 25 30 40 12V batt 25 30 35 30 40 50 Total ICE powertrain weight 495 590 710 590 775 965 EPA then estimated the weight of the electric drive subsystem using the energy content of the battery pack as an input. EPA scaled the weight by applying a specific energy for the electric drive subsystem-including the battery pack, drive motor, wiring, power electronics, etc.-of 120 Wh/kg (or 18.33 Ib/kWh). This specific energy value is based on adding components to an assumed battery pack specific energy of 150 Wh/kgww. Then, the gearbox (the only subsystem excluded from the electric drive scaling) was added to the weight of the electric drive subsystem; this total was included into the electric vehicle weight calculation as Weiectric drive™ • A summary table of electric drive weights for 100-mile EVs is shown as Table 3-65. ww 150 Wh/kg is a conservative estimate for year 2017 and beyond: outputs from ANL's battery cost model, which shows specific energy values of 160-180 Wh/kg for a similar timeframe. xx Applies only to the EV. Because the baseline ICE powertrain weight (which includes gearbox weight) was not deducted from the PHEV, it is not added back in for the PHEV. 3-149 ------- Technologies Considered in the Agencies' Analysis Table 3-65: Total electric drive weights for 100-mile EVs Class Batt pack size (kWh) 2020 electric content (Ibs) Gearbox (power-split or other ) 2020 EV powertrain total 2008 Baseline Small car Std car Large car Small MPV Large MPV Truck 28.2 34.7 41.1 36.7 46.5 53.0 517 635 754 672 853 972 50 60 70 60 80 100 567 695 824 732 933 1072 20 10 Baseline Small car Std car Large car Small MPV Large MPV Truck 29.2 35.7 43.0 37.3 45.8 54.8 536 655 788 683 840 1005 50 60 70 60 80 100 586 715 858 743 920 1105 The difference between the actual weight and the predicted or nominal weight should be zero. However, if not then a revised weight reduction was used for another iteration of steps 2-6 until the two vehicle weights match. Spreadsheet tools such as "solver" in MS Excel were used for automating this iterative process. Table 3-66 shows example results for 100-mile range EVs; in this case a 20% applied glider weight reduction for a variety of vehicle classes. Table 3-66: Sample calculation sheet for 100-mile EVs for the 2008 Baseline Class Small car Std car Large car Small MPV Large MPV Truck Base curb wt (Ib) 2633 3306 3897 3474 4351 5108 Base power/wt ratio 0.0486 0.0575 0.0872 0.0463 0.0565 0.0617 Powertrain weight (Ib) 495 590 710 590 775 965 Base glider wt (Ib) 2138 2716 3187 2884 3576 4143 WR of glider 428 543 637 577 715 829 NewEV wt (nominal Ib) 2205 2763 3260 2897 3636 4279 Energy cons adjusted (Wh/mi) 225 277 329 293 372 424 Batt pack size (kWh) 28.2 34.7 41.1 36.7 46.5 53.0 Electric drive wt (Ib) 567 695 824 732 933 1072 New EV weight (Ib) 2277 2868 3374 3039 3794 4387 Error 0 0 0 0 0 0 %WR from curb 13.5% 13.2% 13.4% 12.5% 12.8% 14.1% %RL vs base 88% 88% 88% 89% 89% 88% Table 3-67 shows the effect on net electric vehicle weight reduction after 20% glider weight reduction was applied to EVs and PHEVs. As battery pack size increases for larger- range EVs and PHEVs, the overall realized vehicle weight reduction decreases (because it requires more energy to carry the extra battery weight). In this example, EVs with a 150 mile range require almost 20% weight reduction to the glider to make up for the additional weight of the electric drive and battery pack compared to a conventional ICE-based powertrain. 3-150 ------- Technologies Considered in the Agencies' Analysis Table 3-67: Actual weight reduction percentages for EVs and PHEVs with 20% weight reduction applied to glider 75 Mile EV Actual %WR vs. base vehicle 100 Mile EV Actual %WR vs. base vehicle 150 Mile EV Actual %WR vs. base vehicle 20 Mile PHEV Actual %WR vs. base vehicle 40 Mile PHEV Actual %WR vs. base vehicle 2008 Baseline Small car Standard car Large car Small MPV Large MPV Truck 19% 18% 19% 18% 18% 19% 14% 13% 13% 13% 13% 14% 2% 2% 2% 1% 1% 3% 12% 12% 12% 12% 12% 11% 7% 7% 7% 7% 7% 6% 20 10 Baseline Small car Standard car Large car Small MPV Large MPV Truck 18% 18% 18% 18% 18% 19% 13% 13% 13% 12% 13% 14% 1% 1% 1% 1% 1% 3% 12% 12% 12% 12% 12% 11% 7% 7% 7% 8% 7% 6% Because there is no "all-electric range" requirement for HEVs, battery pack sizes were relatively consistent for a given weight class. Furthermore, because battery pack sizes are at least an order of magnitude smaller for HEVs than for all-electric vehicles, the sensitivity of HEV vehicle weight (and hence energy consumption) to battery pack size is rather insignificant. For these reasons, a more direct approach (rather than an iterative process) works for battery sizing of HEVs. • FIEV batteries were scaled similar to the 2010 Fusion Hybrid based on nominal battery energy per Ib ETW (equivalent test weight), at 0.37 Wh/lb. • A higher usable SOC window of 40% (compared to 30% for Fusion Hybrid) reduced the required Li-Ion battery size to 75% of the Fusion Hybrid's NiMH battery. This resulted in a 0.28 Wh/lb ETW ratio. • In comparing anecdotal data for HEVs, the agencies assumed a slight weight increase of 4-5% for HEVs compared to baseline non-hybridized vehicles. The added weight of the Li-ion pack, motor and other electric hardware were offset partially by the reduced size of the base engine. 3.4.3.9 HEV, PHEV and EV battery pack design and cost analysis using the ANL BatPaC model The U.S. Department of Energy (DOE) has established long term industry goals and targets for advanced battery systems as it does for many energy efficient technologies. Argonne National Laboratory (ANL) was funded by DOE to provide an independent assessment of Li-ion battery costs because of their expertise in the field as one of the primary 3-151 ------- Technologies Considered in the Agencies' Analysis DOE National Laboratories responsible for basic and applied battery energy storage technologies for future HEV, PHEV and EV applications. A basic description of the ANL Li- ion battery cost model and initial modeling results for PHEV applications were published in a peer-reviewed technical paper presented at EVS-2461. ANL has extended modeling inputs and pack design criteria within the battery cost model to include analysis of manufacturing costs for EVs and HEVs as well has PHEVs.62 In early 2011, ANL issued a draft report detailing the methodology, inputs and outputs of their Battery Performance and Cost (BatPaC) model.63 A complete independent peer-review of the BatPaC model and its inputs and results for HEV, PHEV and EV applications has been completed64. ANL recently provided the agencies with an updated report documenting the BatPaC model that fully addresses the issues raised within the peer review.65 Based on the feedback from peer-reviewers, ANL updated the model in the following areas. 1. Battery pack cost is adjusted upward. This adjustment is based on the feedback from several peer-reviewers, and changes are related to limiting electrode thickness to 100 microns, changing allocation of overhead cost to more closely represent a Tier 1 auto supplier, increasing cost of tabs, changing capital cost of material preparation, etc; 2. Battery management system cost is increased to represent the complete monitoring and control needs for proper battery operation and safety as shown in Table 5.3 in the report; 3. Battery automatic and manual disconnect unit cost is added based on safety considerations as shown in Table 5.3 in the report; 4. Liquid thermal management system is added. ANL stated in the report that the liquid-cooled closure design it uses in the model would not have sufficient surface area and cell spacing to be cooled by air effectively as shown in Table 5.3 in the report. Subsequently, the agencies requested that an option be added to select between liquid or air thermal management and that adequate surface area and cell spacing be determined accordingly. Also, the agencies requested a feature to allow battery packs to be configured as subpacks in parallel or modules in parallel, as additional options for staying within voltage and cell size limits for large packs. ANL added these features in a version of the model distributed March 1, 2012. This version of the model is used for the battery cost estimates in the final rule. This model and the peer review report are available in the public dockets for this rulemaking.64'66 NHTSA and EPA decided to use the ANL BatPaC model for estimating large-format lithium-ion batteries for this final rule, consistent with the proposal, for the following reasons. First, the ANL model has been described and presented in the public domain and does not rely upon confidential business information (which would therefore not be reviewable by the public). The model was developed by scientists at ANL who have significant experience in 3-152 ------- Technologies Considered in the Agencies' Analysis this area. The model uses a bill of materials methodology which the agencies believe is the preferred method for developing cost estimates. The ANL model appropriately considers the vehicle applications power and energy requirements, which are two of the fundamental parameters when designing a lithium-ion battery for an HEV, PHEV, or EV. The ANL model can estimate high volume production costs, which the agencies believe is appropriate for the 2025 time frame. Finally, the ANL model's cost estimates, while generally lower than the estimates we received from the OEMs, is consistent with some of the supplier cost estimates the agencies received from large-format lithium-ion battery pack manufacturers. A portion of those data was received from on-site visits to vehicle manufacturers and battery suppliers done by the EPA in 2008. The ANL battery cost model is based on a bill of materials approach in addition to specific design criteria for the intended application of a battery pack. The costs include materials, manufacturing processes, the cost of capital equipment, plant area, and labor for each manufacturing step as well as the design criteria include a vehicle application's power and energy storage capacity requirements, the battery's cathode and anode chemistry, and the number of cells per module and modules per battery pack. The model assumes use of a laminated multi-layer prismatic cell and battery modules consisting of double-seamed rigid containers. The model also assumes that the battery modules are liquid-cooled. The model takes into consideration the cost of capital equipment, plant area and labor for each step in the manufacturing process for battery packs and places relevant limits on electrode coating thicknesses and other processes limited by existing and near-term manufacturing processes. The ANL model also takes into consideration annual pack production volume and economies of scale for high-volume production. Basic user inputs to BatPaC include performance goals (power and energy capacity), choice of battery chemistry (of five predefined chemistries), the vehicle type for which the battery is intended (HEV, PHEV, or EV), the desired number of cells and modules, and the volume of production. BatPaC then designs the cells, modules, and battery pack, and provides an itemized cost breakdown at the specified production volume. BatPaC provides default values for engineering properties and material costs that allow the model to operate without requiring the user to supply detailed technical or experimental data. In general, the default properties and costs represent what the model authors consider to be reasonable values representing the state of the art expected to be available to large battery manufacturers in the year 2020. Users are encouraged to change these defaults as necessary to represent their own expectations or their own proprietary data. In using BatPaC, it is extremely important that the user monitor certain properties of the cells, modules, and packs that it generates, to ensure that they stay within practical design guidelines, adjusting related inputs if necessary. In particular, pack voltage and individual cell capacity should be limited to appropriate ranges for the application. These design guidelines are not rigidly defined but approximate ranges are beginning to emerge in the industry. Also inherent in BatPaC are certain modeling assumptions that are still open to some uncertainty or debate in the industry. For some, such as the available portion of total battery 3-153 ------- Technologies Considered in the Agencies' Analysis energy (aka "SOC window") for a PHEV/EV/HEV, the user can easily modify a single parameter to represent a value other than the default. For others, such as specific unit costs for thermal management or battery monitoring components, changes can often be made by replacing the relevant components of the model outputs. The cost outputs used by the agencies to determine 2025 HEV, PHEV and EV battery costs were based on the following inputs and assumptions. EPA selected basic user inputs as follows. For performance goals, EPA used the power and energy requirements derived from the scaling analysis described in the previous section. Specifically, these covered each of the six classes of vehicles (Small Car, Standard Car, Large Car, Small MPV, Large MPV and Truck) under each of the five weight reduction scenarios (0%, 2%, 7.5%, 10%, and 20%). The chosen battery chemistries were NMC441-G (for EVs and PHEV40) and LMO-G (for P2 HEVs and PHEV20). Vehicle types were EV75, EV100, EV150 (using the BatPaC "EV" setting); PHEV20 and PHEV40 (using the "PHEV" setting), and P2 HEV (using the "HEV-HP" setting). All modules were composed of 32 cells, with each pack having a varying number of modules. Cost outputs were generated for annual production volumes of 50K, 125K, 250K, and 450K packs. The cost outputs for the 450K production volume are used in the FRM analysis, consistent with the proposal, as being applicable in MY 2017 (HEVs) and MY 2025 (EVs and PHEVs). For engineering properties and material costs, and for other parameters not identified below, EPA used the defaults provided in the model. For design guidelines regarding pack voltage and cell capacity, EPA chose guidelines based on knowledge of current practices and developing trends of battery manufacturers and OEMs, supplemented by discussions with the BatPaC authors. Specifically: (1) allowable pack voltage was targeted to approximately 120V for HEVs and approximately 350-400V for EVs and PHEVs (with some EV150 packs for larger vehicles allowed to about 460-600V); (2) allowable cell capacity was limited to less than approximately 80 A-hr. EPA made several modeling assumptions that differed from the default model: (1) The SOC window for HEVs was increased to 40% rather than the default 25%. (2) HEV packs were modeled as air cooled instead of liquid cooled (except for Truck and MPV with Towing, which are modeled as liquid-cooled). EPA replaced the model's projected costs for air cooling components (blower motor, ducting, and temperature feedback) with costs derived from FEV's teardown studies, which may be more representative of volume production than the default values provided in the model. Additionally, EPA did not include warranty costs computed by BatPaC in the total battery cost because these are accounted for elsewhere by means of indirect cost multipliers (ICMs). Table 3-68 Summary of Inputs and Assumptions Used with BatPaC Category of BatPaC Default or Suggested 3-154 Agency Inputs for FRM Analysis ------- Technologies Considered in the Agencies' Analysis input/Assumptions Annual production volume Battery chemistry Allowable pack voltage Allowable cell capacity Cells per module SOC window for HEVs Thermal management Values n/a n/a forHEV: 1 60-260 V for PHEV, EV: 290-360 V < 60 A-hr 16-32 25% Liquid 450,000 forHEV, PHEV20: LMO-G for PHEV40, EV: NMC441-G forHEV:- 120V for PHEV, EV: ~ 360-600 V < 80 A-hr 32 40% Air, for small/medium HEVs Liquid for all others The cost projections produced by BatPaC are sensitive to the inputs and assumptions the user provides. Significant uncertainty remains regarding which will best represent manufacturer practice in the year 2020. The battery pack cost projection from BatPaC model ranges from \$161/kWh for EV150 truck to \$296/kWh for PHEV40 large car with NMC as chemistry and to \$373/kWh for PHEV20 small car as shown in Table 3-69 to Table 3-74. The agencies note that costs used in the analysis are lower than the costs generally reported in stakeholder meetings, which ranged from \$300/kW-hour to \$400/kW-hour range for 2020 and \$250 to \$300/kW-hour range for 2025. A comparison of BatPaC modeling results to the costs used in the 2012-2016 final rule and to cost estimates compiled by EPA from battery suppliers and auto OEMs is shown in Figure 3-24. The agencies also reviewed publically available PHEV and EV battery cost literature including reports from Anderman67, Frost & Sullivan68, TIAX69, Boston Consulting Group70, and NRC71. Due to the uncertainties inherent in estimating battery costs through the MY 2025 model year, a sensitivity analysis will be provided in each agency's RIA using a range of costs estimated by DOE technical experts to represent a reasonable outer bounds to the results from the BatPaC model. In a recent report to NHTSA and EPA, DOE and ANL suggested the following range for the sensitivity study with 95% confidence interval after analyzing the confidence bound using the BatPaC model. The agencies describe their respective sensitivities surrounding BatPaC costs in their respective RIAs (see Chapter 3.11 of EPA's final RIA and Chapter X of NHTSA's FRIA). Suggested confidence bounds as percentage of the calculated point estimate for a graphite based Li-ion battery using the default inputs in BatPaC Battery type HEV PHEV, EV PHEV, EV Cathodes LMO, LFP, NCA, NMC NMC, NCA LMO, LFP Confidence Interval lower -10% -10% -20% upper 10% 20% 35% 3-155 ------- Technologies Considered in the Agencies' Analysis Figure 3-23 Table from ANL Recommendation 72 Estimated Battery Pack Costs \$1,200.00 \$1,000.00 3" \$800.00 — \$600.00 D \$400.00 w \$200.00 Total Range of OEM Supplier Estimates (>5sources) collected by EPA from 2008 -2010 | Range from the majority of OEM stakeholder meetings in June-August 2010 D2012-2016 EPACost Estimate OBatPac Model (PHEV20] 4-BatPac Model (EV75] * BatPac Model (PHEV40] ABatPac Model (EV150) D BatPac Model (EV100) I ± + 2008 2010 2012 2014 2016 2018 2020 Calendar Year 2022 2024 2026 2028 2030 Figure 3-24 Comparison of direct manufacturing costs per unit of energy storage (S/kW-hr) between the estimates used by EPA in the 2012-2016 GHG final rule, the BatPaC model results for PHEV20, PHEV40, EV75, EV100 and EV150 packages compared to estimates from OEM battery suppliers (2009 dollars, markups not included). Multiple points shown for the BatPaC model results for PHEV 20, PHEV40, EV75, EV100 and EV150 reflect the range of energy-specific costs for EPA's subcompact through large- car package categories (see Table 3-70 through Table 3-74for details). A range of OEM estimated battery costs from stakeholder meetings is also shown for comparison (red bars) which may or may not reflect additional cost markups. While it is expected that other Li-ion battery chemistries with higher energy density, higher power density and lower cost will likely be available in the 2017-2025 timeframe, the specific chemistries used for the cost analysis were chosen due to their known characteristics and to be consistent with both public available information on current and near term HEV, PHEV and EV product offerings from Hyundai, GM and Nissan as well as confidential business information on future products currently under development.73'74'75'76 The cost outputs from the BatPaC model used by the agencies in this analysis are shown in Table 3-69 through Table 3-74 for different levels of applied weight reduction technology. We differentiate between "applied" weight reduction and "net" weight reduction in this analysis because to achieve the same amount of mass reduction, more mass reduction technologies might need to be applied to vehicles with electrification than with traditional powertrains because of the added weight of the electrification systems (i.e., the battery, electric motors, etc.). This also makes it clear that we have estimated vehicle level battery pack costs—and motor and other electrified vehicle specific costs—based on the net weight reduction of the 3-156 ------- Technologies Considered in the Agencies' Analysis vehicle. For example, a typical EV150 battery back and associated motors and other EV- specific equipment increases vehicle weight roughly 18 percent. As a result, an EV150 that applied 20 percent mass reduction technology (see section 3.4.5.5 for a full discussion of mass reduction technologies and costs) would have a net weight reduction of only 2 percent. In such a case, the agencies would estimate mass reduction costs associated with a 20 percent applied mass reduction, and EV150 costs associated with only a 2 percent net mass reduction (lower net mass reduction results in higher battery pack and motor costs). Similarly, HEV battery packs increase vehicle weight by roughly 5 or 6 percent. Therefore, for an HEV with 20 percent applied mass reduction technology—and costs associated with 20 percent applied mass reduction—would have HEV costs associated with a 15 percent net mass reduction. Furthermore, such an HEV would have an effectiveness level improvement associated with a 15 percent net mass reduction rather than a 20 percent net reduction. Table 3-69 Direct Manufacturing Costs for P2 HEV battery packs at different levels of applied vehicle weight reduction (2010 dollars, markups not included) P2 HEV (LMO) @ 450K/yr volume 0% weight reduction Pack \$/kWh 2% weight reduction Pack \$/kWh 7.5% weight reduction Pack \$/kWh 10% weight reduction Pack \$/kWh 20% weight reduction Pack \$/kWh 2008 Baseline Small Car Standard Car Large Car Small MPV Large MPV Truck \$726 \$801 \$938 \$779 \$876 \$1,010 \$896 \$804 \$809 \$747 \$682 \$676 \$722 \$796 \$929 \$775 \$870 \$1,003 \$909 \$815 \$817 \$758 \$691 \$685 \$712 \$783 \$909 \$762 \$853 \$983 \$950 \$849 \$848 \$790 \$718 \$711 \$708 \$777 \$900 \$757 \$846 \$974 \$970 \$866 \$862 \$806 \$731 \$724 \$700 \$765 \$882 \$746 \$830 \$957 \$1,008 \$901 \$894 \$839 \$760 \$747 20 10 Baseline Small Car Standard Car Large Car Small MPV Large MPV Truck \$732 \$809 \$950 \$788 \$878 \$1,019 \$904 \$813 \$819 \$756 \$683 \$682 \$729 \$805 \$943 \$784 \$872 \$1,012 \$918 \$824 \$830 \$767 \$692 \$691 \$718 \$791 \$920 \$771 \$855 \$992 \$958 \$858 \$858 \$800 \$720 \$718 \$714 \$785 \$911 \$765 \$847 \$983 \$978 \$875 \$873 \$816 \$733 \$731 \$705 \$773 \$893 \$754 \$832 \$967 \$1,017 \$909 \$904 \$848 \$762 \$754 Table 3-70 Direct Manufacturing Costs for PHEV20 battery packs at different levels of applied vehicle weight reduction (2010 dollars, markups not included) PHEV20 (LMO) @ 450K/yr volume 0% weight reduction Pack \$/kWh 2% weight reduction Pack \$/kWh 7.5% weight reduction Pack \$/kWh 10% weight reduction Pack \$/kWh 20% weight reduction Pack \$/kWh 2008 Baseline Small Car Standard Car \$2,531 \$2,962 \$364 \$347 \$2,517 \$2,938 \$364 \$348 \$2,469 \$2,835 \$370 \$345 \$2,447 \$2,808 \$371 \$346 \$2,431 \$2,784 \$373 \$347 3-157 ------- Technologies Considered in the Agencies' Analysis Large Car Small MPV Large MPV Truck \$3,734 \$2,835 \$3,424 \$3,874 \$368 \$316 \$300 \$295 \$3,696 \$2,813 \$3,393 \$3,834 \$369 \$317 \$301 \$295 \$3,592 \$2,754 \$3,309 \$3,732 \$369 \$319 \$302 \$295 \$3,546 \$2,730 \$3,274 \$3,681 \$368 \$320 \$303 \$297 \$3,510 \$2,703 \$3,244 \$3,671 \$369 \$323 \$303 \$296 20 10 Baseline Small Car Standard Car Large Car Small MPV Large MPV Truck \$2,572 \$3,019 \$3,813 \$2,933 \$3,434 \$3,922 \$370 \$353 \$376 \$326 \$301 \$298 \$2,554 \$2,992 \$3,773 \$2,911 \$3,403 \$3,881 \$370 \$354 \$376 \$328 \$302 \$298 \$2,507 \$2,927 \$3,668 \$2,811 \$3,319 \$3,778 \$376 \$357 \$376 \$326 \$303 \$299 \$2,487 \$2,858 \$3,621 \$2,783 \$3,282 \$3,732 \$377 \$352 \$376 \$326 \$303 \$301 \$2,468 \$2,829 \$3,575 \$2,754 \$3,253 \$3,706 \$379 \$353 \$376 \$329 \$304 \$298 Table 3-71 Direct Manufacturing Costs for PHEV40 battery pack at weight reduction (2010 dollars, markups not different levels of applied vehicle included) PHEV40 (NMC) @ 450K/yr volume 0% weight reduction Pack \$/kWh 2% weight reduction Pack \$/kWh 7.5% weight reduction Pack \$/kWh 10% weight reduction Pack \$/kWh 20% weight reduction Pack \$/kWh 2008 Baseline Small Car Standard Car Large Car Small MPV Large MPV Truck \$3,644 \$4,390 \$6,006 \$4,247 \$5,269 \$6,122 \$262 \$257 \$296 \$236 \$231 \$233 \$3,619 \$4,343 \$5,921 \$4,207 \$5,212 \$6,050 \$262 \$257 \$295 \$237 \$231 \$233 \$3,542 \$4,228 \$5,671 \$4,101 \$5,065 \$5,900 \$264 \$258 \$291 \$238 \$231 \$232 \$3,542 \$4,228 \$5,671 \$4,100 \$5,065 \$5,900 \$264 \$258 \$291 \$237 \$231 \$232 \$3,542 \$4,228 \$5,671 \$4,100 \$5,065 \$5,900 \$264 \$258 \$291 \$237 \$231 \$232 20 10 Baseline Small Car Standard Car Large Car Small MPV Large MPV Truck \$3,722 \$4,494 \$6,158 \$4,351 \$5,286 \$6,215 \$268 \$263 \$304 \$242 \$232 \$236 \$3,690 \$4,447 \$6,073 \$4,309 \$5,228 \$6,142 \$267 \$263 \$303 \$243 \$232 \$236 \$3,606 \$4,324 \$5,850 \$4,198 \$5,080 \$5,980 \$269 \$263 \$300 \$243 \$232 \$235 \$3,606 \$4,324 \$5,850 \$4,198 \$5,080 \$5,980 \$269 \$263 \$300 \$243 \$232 \$235 \$3,606 \$4,324 \$5,850 \$4,198 \$5,080 \$5,980 \$269 \$263 \$300 \$243 \$232 \$235 Table 3-72 Direct Manufacturing Costs for EV75 battery packs at different levels of applied vehicle weight reduction (2010 dollars, markups not included) EV75 (NMC) @ 450K/yr volume 0% weight reduction Pack \$/kWh 2% weight reduction Pack \$/kWh 7.5% weight reduction Pack \$/kWh 10% weight reduction Pack \$/kWh 20% weight reduction Pack \$/kWh 3-158 ------- Technologies Considered in the Agencies' Analysis 2008 Baseline Small Car Standard Car Large Car Small MPV Large MPV \$5,115 \$6,021 \$7,724 \$5,995 \$7,310 \$224 \$215 \$232 \$203 \$195 \$5,098 \$5,965 \$7,635 \$5,952 \$7,237 \$225 \$215 \$232 \$204 \$196 \$4,996 \$5,818 \$7,397 \$5,843 \$7,045 \$228 \$216 \$231 \$206 \$196 \$4,962 \$5,755 \$7,295 \$5,800 \$6,963 \$229 \$216 \$231 \$207 \$196 \$4,768 \$5,509 \$6,907 \$5,625 \$6,610 \$233 \$219 \$231 \$211 \$197 Truck 8,332 \$193 \$8,242 \$193 \$8,005 \$193 \$7,883 \$194 \$7,474 \$194 20 10 Baseline Small Car Standard Car Large Car Small MPV Large MPV Truck \$5,232 \$6,152 \$7,923 \$6,070 \$7,312 \$8,472 \$221 \$214 \$229 \$203 \$197 \$191 \$5,195 \$6,092 \$7,832 \$6,016 \$7,238 \$8,380 \$222 \$214 \$229 \$203 \$198 \$191 \$5,106 \$5,940 \$7,586 \$5,904 \$7,046 \$8,141 \$225 \$215 \$229 \$205 \$198 \$191 \$5,071 \$5,874 \$7,479 \$5,860 \$6,962 \$8,036 \$226 \$215 \$228 \$206 \$198 \$191 \$4,912 \$5,624 \$7,092 \$5,684 \$6,605 \$7,629 \$231 \$218 \$228 \$210 \$198 \$191 Table 3-73 Direct Manufacturing Costs for EV100 battery packs at different levels of applied vehicle weight reduction (2010 dollars, markups not included) EV100 (NMC) @ 450K/yr volume 0% weight reduction Pack \$/kWh 2% weight reduction Pack \$/kWh 7. 5% weight reduction Pack \$/kWh 10% weight reduction Pack \$/kWh 20% weight reduction Pack \$/kWh 2008 Baseline Small Car Standard Car Large Car Small MPV Large MPV Truck \$6,105 \$7,054 \$8,630 \$7,293 \$8,641 \$9,962 \$201 \$189 \$195 \$186 \$173 \$173 \$6,083 \$7,001 \$8,535 \$7,237 \$8,571 \$9,879 \$201 \$189 \$195 \$186 \$174 \$174 \$5,950 \$6,826 \$8,283 \$7,096 \$8,392 \$9,676 \$204 \$190 \$194 \$188 \$175 \$175 \$5,906 \$6,770 \$8,175 \$7,039 \$8,321 \$9,554 \$205 \$191 \$194 \$189 \$176 \$176 \$5,817 \$6,662 \$7,999 \$6,953 \$8,215 \$9,392 \$206 \$192 \$194 \$190 \$177 \$177 20 10 Baseline Small Car Standard Car Large Car Small MPV Large MPV Truck \$6,255 \$7,173 \$8,863 \$7,375 \$8,586 \$10,158 \$198 \$187 \$192 \$185 \$174 \$172 \$6,209 \$7,118 \$8,765 \$7,318 \$8,516 \$10,075 \$199 \$188 \$192 \$185 \$174 \$172 \$6,094 \$6,980 \$8,504 \$7,174 \$8,338 \$9,865 \$201 \$190 \$192 \$187 \$176 \$174 \$6,048 \$6,884 \$8,393 \$7,117 \$8,268 \$9,782 \$202 \$189 \$192 \$188 \$176 \$174 \$5,956 \$6,802 \$8,251 \$7,031 \$8,128 \$9,615 \$204 \$190 \$192 \$189 \$177 \$175 3-159 ------- Technologies Considered in the Agencies' Analysis Table 3-74 Direct Manufacturing Costs for EV150 battery packs at different levels of applied vehicle weight reduction (2010 dollars, markups not included) EV150 (NMQ @ 450K/yr volume 0% weight reduction Pack \$/kWh 2% weight reduction Pack \$/kWh 7.5% weight reduction Pack \$/kWh 10% weight reduction Pack \$/kWh 20% weight reduction Pack \$/kWh 2008 Baseline Small Car Standard Car Large Car Small MPV Large MPV Truck \$8,080 \$9,753 \$11,120 \$10,109 \$12,114 \$13,878 \$177 \$174 \$167 \$171 \$162 \$161 \$8,048 \$9,714 \$11,073 \$10,109 \$12,112 \$13,818 \$178 \$174 \$167 \$171 \$162 \$161 \$8,048 \$9,714 \$11,073 \$10,109 \$12,112 \$13,759 \$178 \$174 \$167 \$171 \$162 \$161 \$8,048 \$9,714 \$11,073 \$10,109 \$12,112 \$13,759 \$178 \$174 \$167 \$171 \$162 \$161 \$8,048 \$9,714 \$11,073 \$10,109 \$12,112 \$13,759 \$178 \$174 \$167 \$171 \$162 \$161 20 10 Baseline Small Car Standard Car Large Car Small MPV Large MPV Truck \$8,298 \$9,928 \$11,432 \$10,228 \$12,032 \$14,166 \$175 \$173 \$166 \$171 \$162 \$160 \$8,265 \$9,888 \$11,384 \$10,228 \$11,981 \$14,045 \$176 \$173 \$166 \$171 \$163 \$160 \$8,265 \$9,888 \$11,384 \$10,228 \$11,981 \$14,044 \$176 \$173 \$166 \$171 \$163 \$160 \$8,265 \$9,888 \$11,384 \$10,228 \$11,981 \$14,044 \$176 \$173 \$166 \$171 \$163 \$160 \$8,265 \$9,888 \$11,384 \$10,228 \$11,981 \$14,044 \$176 \$173 \$166 \$171 \$163 \$160 Specifically for modeling purposes, both agencies wanted HEV/PHEV/EV battery pack costs based on net weight reduction rather than applied weight reduction as shown in Table 3-69 through Table 3-74 above. The agencies did this by first determining the average weight differences (applied weight reduction vs net weight reduction) for each of the 6 major vehicle classes (small car, standard car, large car, small MPV, large MPV & truck) and each of the electrification types (P2 HEV, PHEV & EV). Due to the weight increases of adding the electrification system and battery pack and the weight decreases by applying smaller or no conventional internal combustion engine, the net mass reduction for HEV, PHEV and EV varies for different electrification packages and vehicle classes. For example, for a 20-mile small car PHEV, a 5% mass reduction of the glider is offset by the additional weight of the electrification system. Said another way, a 5% mass reduction needs to be applied to the glider to achieve a net 0% overall vehicle mass reduction for a PHEV20 small car. Those weight reduction differences are shown in Table 3-75. Table 3-75 EPA and NHTSA Weight Reduction Offset Associated with Electrification Technologies Vehicle Class P2HEV PHEV20 PHEV40 EV75 EV100 EV150 2008 Baseline Small car Standard car Large car Small MPV Large MPV 5% 5% 5% 5% 5% 7% 7% 8% 7% 7% 13% 12% 14% 12% 12% 0% 0% -1% 1% 0% 6% 6% 5% 7% 6% 18% 18% 17% 19% 18% 3-160 ------- Technologies Considered in the Agencies' Analysis Truck 4% 7% 12% 1% 7% 19% 20 10 Baseline Small car Standard car Large car Small MPV Large MPV Truck 5% 5% 5% 5% 5% 4% 7% 7% 8% 7% 7% 7% 12% 12% 13% 12% 12% 12% 0% 1% 0% 1% 1% 0% 6% 7% 6% 7% 7% 6% 19% 19% 17% 19% 19% 19% Notes: For example, PHEV40-specific technologies add 12-14% to vehicle weight so that a 20% applied weight reduction would result in a 6-8% net weight reduction. While an EV75 can actually reduce vehicle weight by 1-2% (i.e., battery packs and motors weigh less than the removed internal combustion engine and transmission), the agencies used a value of 0% where negative entries are shown. The agencies then generated linear regressions of battery pack costs against percentage net weight reduction using the costs shown in Table 3-69 through Table 3-74 and the weight reduction offsets shown in Table 3-75. These results are shown in Table 3-76. Table 3-76 EPA and NHTSA Linear Regressions of Battery Pack Direct Manufacturing Costs vs Net Weight Reduction (2010\$) Vehicle Class P2HEV PHEV20 PHEV40 EV75 EV100 EV150 2008 Baseline Small car Standard car Large car Small MPV Large MPV Truck -\$181x+\$726 -\$240x+\$801 -\$369x+\$937 -\$224x+\$779 -\$303x+\$876 -\$367x+\$l,010 -\$861x+\$2,533 -\$l,543x+\$2,962 -\$l,881x+\$3,734 -\$l,073x+\$2,835 -\$l,517x+\$3,646 -\$2,195x+\$4,389 -\$4,700x+\$6,010 -\$l,957x+\$4,247 -\$l,859x+\$5,131 -\$2,754x+\$6,023 -\$4,356x+\$7,725 -\$2,061x+\$5,997 -\$2,168x+\$6,115 -\$2,958x+\$7,056 -\$4,647x+\$8,630 -\$2,649x+\$7,293 -\$2,045x+\$8,080 -\$2,552x+\$9,753 -\$2,840x+\$l 1,120 -\$19x+\$10,109 20 10 Baseline Small car Standard car Large car Small MPV Large MPV Truck -\$188x+\$733 -\$248x+\$810 -\$387x+\$950 -\$233x+\$789 -\$305x+\$878 -\$364x+\$l,019 -\$866x+\$2,572 -\$l,573x+\$3,024 -\$l,957x+\$3,813 -\$l,516x+\$2,934 -\$l,612x+\$3,722 -\$2,291x+\$4,494 -\$4,217x+\$6,158 -\$2,022x+\$4,350 -\$l,717x+\$5,233 -\$2,887x+\$6,154 -\$4,543x+\$7,925 -\$2,155x+\$6,067 -\$2,209x+\$6,256 -\$2,883x+\$7,178 -\$4,744x+\$8,862 -\$2,706x+\$7,375 -\$2,700x+\$8,298 -\$3,242x+\$9,928 -\$4,250x+\$l 1,432 -\$21x+\$10,228 Notes: "x" in the equations represents the net weight reduction as a percentage, so a small car P2 HEV battery pack (2008 baseline) with a 20% applied weight reduction and, therefore, a 15% net weight reduction would cost (-\$181)x(15%)+\$726=\$698. The agencies did not regress PHEV or EV costs for the large MPV and truck vehicle classes since we do not believe these vehicle classes would use the technologies. For P2 HEV battery packs, the direct manufacturing costs shown in Table 3-76 are considered applicable to the 2017MY. The agencies consider the P2 battery packs technology to be on the flat portion of the learning curve during the 2017-2025 timeframe. The agencies have applied a highl complexity ICM of 1.56 through 2024 then 1.35 thereafter. For PHEV 3-161 ------- Technologies Considered in the Agencies' Analysis and EV battery packs, the direct manufacturing costs shown in Table 3-76 are considered applicable to the 2025MY. For the PHEV and EV battery packs, the agencies have applied the learning curve discussed in Section 3.2.3. The agencies have applied a high2 complexity ICM of 1.77 through 2024 then 1.50 thereafter. The resultant costs for P2 HEV, PHEV20, PHEV40, EV75, EV100 and EV150 battery packs for the 2008 and 2010 baselines are shown in Table 3-77 through Table 3-87, respectively. Table 3-77 Costs for P2 HEV Battery Packs for the 2008 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Large MPV Large MPV Large MPV Truck Truck Truck Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Large MPV Large MPV Large MPV Truck Truck Truck Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% Net WR 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 6% 11% 16% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 6% 11% 16% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 2017 \$717 \$707 \$698 \$789 \$777 \$765 \$919 \$900 \$882 \$768 \$757 \$745 \$861 \$846 \$830 \$988 \$970 \$951 \$404 \$399 \$394 \$444 \$438 \$431 \$518 \$507 \$497 \$433 \$426 \$420 \$485 \$477 \$468 \$557 \$546 \$536 \$1,120 \$1,106 \$1,092 \$1,233 \$1,214 \$1,196 \$1,436 \$1,407 \$1,379 \$1,201 \$1,183 2018 \$695 \$686 \$677 \$765 \$753 \$742 \$891 \$873 \$855 \$745 \$734 \$723 \$835 \$820 \$805 \$958 \$941 \$923 \$402 \$397 \$392 \$443 \$436 \$429 \$516 \$506 \$495 \$431 \$425 \$419 \$483 \$475 \$466 \$555 \$545 \$534 \$1,097 \$1,084 \$1,070 \$1,208 \$1,190 \$1,171 \$1,407 \$1,379 \$1,350 \$1,176 \$1,159 2019 \$674 \$666 \$657 \$742 \$731 \$719 \$864 \$847 \$830 \$722 \$712 \$701 \$810 \$796 \$781 \$930 \$912 \$895 \$401 \$396 \$391 \$441 \$435 \$428 \$514 \$504 \$494 \$430 \$424 \$417 \$482 \$473 \$465 \$553 \$543 \$533 \$1,075 \$1,062 \$1,048 \$1,183 \$1,165 \$1,147 \$1,378 \$1,351 \$1,323 \$1,152 \$1,135 2020 \$654 \$646 \$637 \$720 \$709 \$698 \$838 \$821 \$805 \$701 \$690 \$680 \$786 \$772 \$758 \$902 \$885 \$868 \$400 \$395 \$390 \$440 \$433 \$427 \$512 \$502 \$492 \$428 \$422 \$416 \$480 \$472 \$463 \$551 \$541 \$531 \$1,054 \$1,040 \$1,027 \$1,160 \$1,142 \$1,125 \$1,351 \$1,324 \$1,297 \$1,129 \$1,113 2021 \$634 \$626 \$618 \$698 \$688 \$677 \$813 \$797 \$781 \$680 \$670 \$660 \$762 \$749 \$735 \$875 \$858 \$842 \$399 \$393 \$388 \$439 \$432 \$425 \$511 \$501 \$490 \$427 \$421 \$415 \$479 \$470 \$462 \$549 \$539 \$529 \$1,033 \$1,020 \$1,007 \$1,137 \$1,119 \$1,102 \$1,324 \$1,297 \$1,271 \$1,107 \$1,091 2022 \$615 \$608 \$600 \$677 \$667 \$657 \$789 \$773 \$757 \$659 \$650 \$640 \$739 \$726 \$713 \$848 \$833 \$817 \$397 \$392 \$387 \$437 \$431 \$424 \$509 \$499 \$489 \$426 \$419 \$413 \$477 \$469 \$460 \$548 \$538 \$527 \$1,013 \$1,000 \$987 \$1,114 \$1,098 \$1,081 \$1,298 \$1,272 \$1,246 \$1,085 \$1,069 2023 \$597 \$589 \$582 \$657 \$647 \$637 \$765 \$750 \$734 \$640 \$630 \$621 \$717 \$704 \$692 \$823 \$808 \$792 \$396 \$391 \$386 \$436 \$429 \$423 \$508 \$498 \$487 \$424 \$418 \$412 \$476 \$467 \$459 \$546 \$536 \$526 \$993 \$980 \$968 \$1,093 \$1,076 \$1,060 \$1,273 \$1,247 \$1,222 \$1,064 \$1,048 2024 \$579 \$572 \$564 \$637 \$628 \$618 \$742 \$727 \$712 \$620 \$611 \$602 \$695 \$683 \$671 \$798 \$783 \$769 \$395 \$390 \$385 \$435 \$428 \$421 \$506 \$496 \$486 \$423 \$417 \$411 \$474 \$466 \$458 \$545 \$534 \$524 \$974 \$962 \$949 \$1,072 \$1,056 \$1,039 \$1,248 \$1,223 \$1,198 \$1,044 \$1,028 2025 \$562 \$554 \$547 \$618 \$609 \$599 \$720 \$705 \$691 \$602 \$593 \$584 \$675 \$663 \$651 \$774 \$760 \$746 \$243 \$239 \$236 \$267 \$263 \$259 \$311 \$305 \$298 \$260 \$256 \$252 \$291 \$286 \$281 \$334 \$328 \$322 \$804 \$794 \$784 \$885 \$872 \$858 \$1,031 \$1,010 \$989 \$862 \$849 3-162 ------- Technologies Considered in the Agencies' Analysis TC TC TC TC TC TC TC Small MPV Large MPV Large MPV Large MPV Truck Truck Truck 20% 10% 15% 20% 10% 15% 20% 15% 5% 10% 15% 6% 11% 16% \$1,165 \$1,346 \$1,322 \$1,298 \$1,545 \$1,516 \$1,487 \$1,142 \$1,318 \$1,295 \$1,272 \$1,513 \$1,485 \$1,457 \$1,119 \$1,292 \$1,269 \$1,246 \$1,483 \$1,455 \$1,428 \$1,096 \$1,266 \$1,243 \$1,221 \$1,453 \$1,426 \$1,399 \$1,074 \$1,241 \$1,219 \$1,197 \$1,424 \$1,398 \$1,371 \$1,053 \$1,216 \$1,195 \$1,174 \$1,396 \$1,370 \$1,344 \$1,033 \$1,193 \$1,172 \$1,151 \$1,369 \$1,344 \$1,318 \$1,013 \$1,170 \$1,149 \$1,129 \$1,343 \$1,318 \$1,293 \$836 \$966 \$949 \$932 \$1,109 \$1,088 \$1,068 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-78 Costs for P2 HEV Battery Packs for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Large MPV Large MPV Large MPV Truck Truck Truck Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Large MPV Large MPV Large MPV Truck Truck Truck Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% Net WR 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 6% 11% 16% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 6% 11% 16% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 2017 \$723 \$714 \$704 \$797 \$785 \$772 \$931 \$911 \$892 \$777 \$765 \$754 \$863 \$847 \$832 \$997 \$979 \$961 \$408 \$402 \$397 \$449 \$442 \$435 \$524 \$514 \$503 \$438 \$431 \$425 \$486 \$478 \$469 \$562 \$552 \$541 \$1,131 \$1,116 \$1,101 \$1,246 \$1,227 \$1,208 \$1,455 \$1,425 \$1,394 \$1,215 2018 \$701 \$692 \$683 \$773 \$761 \$749 \$903 \$884 \$865 \$754 \$742 \$731 \$837 \$822 \$807 \$967 \$949 \$932 \$406 \$401 \$396 \$448 \$441 \$434 \$523 \$512 \$501 \$436 \$430 \$423 \$485 \$476 \$467 \$560 \$550 \$540 \$1,108 \$1,093 \$1,079 \$1,221 \$1,202 \$1,183 \$1,425 \$1,396 \$1,366 \$1,190 2019 \$680 \$672 \$663 \$750 \$738 \$727 \$876 \$857 \$839 \$731 \$720 \$709 \$812 \$797 \$783 \$938 \$921 \$904 \$405 \$400 \$394 \$446 \$439 \$432 \$521 \$510 \$499 \$435 \$428 \$422 \$483 \$474 \$466 \$558 \$548 \$538 \$1,085 \$1,071 \$1,057 \$1,196 \$1,178 \$1,159 \$1,396 \$1,367 \$1,338 \$1,166 2020 \$660 \$651 \$643 \$727 \$716 \$705 \$849 \$832 \$814 \$709 \$698 \$688 \$787 \$773 \$759 \$910 \$893 \$877 \$403 \$398 \$393 \$445 \$438 \$431 \$519 \$508 \$498 \$433 \$427 \$420 \$481 \$473 \$464 \$556 \$546 \$536 \$1,063 \$1,050 \$1,036 \$1,172 \$1,154 \$1,136 \$1,369 \$1,340 \$1,312 \$1,142 2021 \$640 \$632 \$624 \$706 \$695 \$684 \$824 \$807 \$790 \$688 \$677 \$667 \$764 \$750 \$737 \$883 \$867 \$850 \$402 \$397 \$392 \$443 \$436 \$430 \$518 \$507 \$496 \$432 \$426 \$419 \$480 \$471 \$463 \$555 \$544 \$534 \$1,042 \$1,029 \$1,015 \$1,149 \$1,131 \$1,113 \$1,341 \$1,313 \$1,286 \$1,120 2022 \$621 \$613 \$605 \$685 \$674 \$663 \$799 \$782 \$766 \$667 \$657 \$647 \$741 \$728 \$715 \$856 \$841 \$825 \$401 \$396 \$391 \$442 \$435 \$428 \$516 \$505 \$494 \$431 \$424 \$418 \$478 \$470 \$461 \$553 \$543 \$533 \$1,022 \$1,009 \$995 \$1,126 \$1,109 \$1,091 \$1,315 \$1,288 \$1,260 \$1,098 2023 \$602 \$595 \$587 \$664 \$654 \$643 \$775 \$759 \$743 \$647 \$637 \$628 \$719 \$706 \$693 \$831 \$815 \$800 \$400 \$395 \$389 \$441 \$434 \$427 \$514 \$504 \$493 \$429 \$423 \$417 \$477 \$468 \$460 \$551 \$541 \$531 \$1,002 \$989 \$976 \$1,105 \$1,087 \$1,070 \$1,290 \$1,263 \$1,236 \$1,077 2024 \$584 \$577 \$569 \$644 \$634 \$624 \$752 \$736 \$721 \$628 \$618 \$609 \$697 \$685 \$672 \$806 \$791 \$776 \$399 \$393 \$388 \$439 \$433 \$426 \$513 \$502 \$492 \$428 \$422 \$415 \$475 \$467 \$459 \$550 \$540 \$529 \$983 \$970 \$957 \$1,083 \$1,067 \$1,050 \$1,265 \$1,238 \$1,212 \$1,056 2025 \$567 \$559 \$552 \$625 \$615 \$605 \$729 \$714 \$699 \$609 \$600 \$591 \$676 \$664 \$652 \$781 \$767 \$753 \$245 \$242 \$238 \$270 \$266 \$261 \$315 \$308 \$302 \$263 \$259 \$255 \$292 \$287 \$282 \$338 \$331 \$325 \$812 \$801 \$790 \$895 \$881 \$867 \$1,044 \$1,023 \$1,001 \$872 3-163 ------- Technologies Considered in the Agencies' Analysis TC TC TC TC TC TC TC TC Small MPV Small MPV Large MPV Large MPV Large MPV Truck Truck Truck 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 5% 10% 15% 6% 11% 16% \$1,196 \$1,178 \$1,349 \$1,325 \$1,301 \$1,559 \$1,531 \$1,502 \$1,172 \$1,154 \$1,321 \$1,298 \$1,275 \$1,527 \$1,499 \$1,471 \$1,148 \$1,131 \$1,295 \$1,272 \$1,249 \$1,496 \$1,469 \$1,442 \$1,125 \$1,108 \$1,269 \$1,246 \$1,224 \$1,466 \$1,440 \$1,413 \$1,103 \$1,086 \$1,243 \$1,222 \$1,200 \$1,437 \$1,411 \$1,385 \$1,081 \$1,065 \$1,219 \$1,198 \$1,176 \$1,409 \$1,383 \$1,358 \$1,060 \$1,044 \$1,195 \$1,174 \$1,153 \$1,382 \$1,356 \$1,331 \$1,040 \$1,024 \$1,172 \$1,152 \$1,131 \$1,355 \$1,330 \$1,306 \$859 \$846 \$968 \$951 \$934 \$1,119 \$1,099 \$1,078 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-79 Costs for PHEV20 Battery Packs for the 2008 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% NetWR 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 8% 13% 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 8% 13% 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 2017 \$4,896 \$4,812 \$4,728 \$5,696 \$5,545 \$5,394 \$7,219 \$7,035 \$6,851 \$5,474 \$5,370 \$5,265 \$2,106 \$2,070 \$2,034 \$2,450 \$2,385 \$2,321 \$3,105 \$3,026 \$2,947 \$2,355 \$2,310 \$2,265 \$7,003 \$6,883 \$6,762 \$8,146 \$7,930 \$7,715 \$10,324 \$10,061 \$9,799 \$7,829 2018 \$3,917 \$3,850 \$3,783 \$4,557 \$4,436 \$4,315 \$5,775 \$5,628 \$5,481 \$4,379 \$4,296 \$4,212 \$2,034 \$1,999 \$1,964 \$2,366 \$2,304 \$2,241 \$2,999 \$2,923 \$2,846 \$2,274 \$2,231 \$2,187 \$5,951 \$5,849 \$5,747 \$6,923 \$6,740 \$6,556 \$8,774 \$8,551 \$8,327 \$6,654 2019 \$3,917 \$3,850 \$3,783 \$4,557 \$4,436 \$4,315 \$5,775 \$5,628 \$5,481 \$4,379 \$4,296 \$4,212 \$2,034 \$1,999 \$1,964 \$2,366 \$2,304 \$2,241 \$2,999 \$2,923 \$2,846 \$2,274 \$2,231 \$2,187 \$5,951 \$5,849 \$5,747 \$6,923 \$6,740 \$6,556 \$8,774 \$8,551 \$8,327 \$6,654 2020 \$3,134 \$3,080 \$3,026 \$3,645 \$3,549 \$3,452 \$4,620 \$4,502 \$4,385 \$3,504 \$3,436 \$3,369 \$1,977 \$1,943 \$1,909 \$2,299 \$2,238 \$2,178 \$2,914 \$2,840 \$2,766 \$2,210 \$2,168 \$2,125 \$5,110 \$5,023 \$4,935 \$5,944 \$5,787 \$5,630 \$7,534 \$7,342 \$7,151 \$5,713 2021 \$3,134 \$3,080 \$3,026 \$3,645 \$3,549 \$3,452 \$4,620 \$4,502 \$4,385 \$3,504 \$3,436 \$3,369 \$1,977 \$1,943 \$1,909 \$2,299 \$2,238 \$2,178 \$2,914 \$2,840 \$2,766 \$2,210 \$2,168 \$2,125 \$5,110 \$5,023 \$4,935 \$5,944 \$5,787 \$5,630 \$7,534 \$7,342 \$7,151 \$5,713 2022 \$3,134 \$3,080 \$3,026 \$3,645 \$3,549 \$3,452 \$4,620 \$4,502 \$4,385 \$3,504 \$3,436 \$3,369 \$1,977 \$1,943 \$1,909 \$2,299 \$2,238 \$2,178 \$2,914 \$2,840 \$2,766 \$2,210 \$2,168 \$2,125 \$5,110 \$5,023 \$4,935 \$5,944 \$5,787 \$5,630 \$7,534 \$7,342 \$7,151 \$5,713 2023 \$3,134 \$3,080 \$3,026 \$3,645 \$3,549 \$3,452 \$4,620 \$4,502 \$4,385 \$3,504 \$3,436 \$3,369 \$1,977 \$1,943 \$1,909 \$2,299 \$2,238 \$2,178 \$2,914 \$2,840 \$2,766 \$2,210 \$2,168 \$2,125 \$5,110 \$5,023 \$4,935 \$5,944 \$5,787 \$5,630 \$7,534 \$7,342 \$7,151 \$5,713 2024 \$3,134 \$3,080 \$3,026 \$3,645 \$3,549 \$3,452 \$4,620 \$4,502 \$4,385 \$3,504 \$3,436 \$3,369 \$1,977 \$1,943 \$1,909 \$2,299 \$2,238 \$2,178 \$2,914 \$2,840 \$2,766 \$2,210 \$2,168 \$2,125 \$5,110 \$5,023 \$4,935 \$5,944 \$5,787 \$5,630 \$7,534 \$7,342 \$7,151 \$5,713 2025 \$2,507 \$2,464 \$2,421 \$2,916 \$2,839 \$2,762 \$3,696 \$3,602 \$3,508 \$2,803 \$2,749 \$2,696 \$1,245 \$1,224 \$1,202 \$1,448 \$1,410 \$1,372 \$1,835 \$1,789 \$1,742 \$1,392 \$1,365 \$1,339 \$3,752 \$3,688 \$3,623 \$4,364 \$4,249 \$4,133 \$5,531 \$5,391 \$5,250 \$4,195 3-164 ------- Technologies Considered in the Agencies' Analysis TC TC Small MPV Small MPV 15% 20% 8% 13% \$7,679 \$7,530 \$6,526 \$6,399 \$6,526 \$6,399 \$5,604 \$5,495 \$5,604 \$5,495 \$5,604 \$5,495 \$5,604 \$5,495 \$5,604 \$5,495 \$4,114 \$4,034 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-80 Costs for PHEV20 Battery Packs for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% NetWR 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 8% 13% 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 8% 13% 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 8% 13% 2017 \$4,973 \$4,888 \$4,804 \$5,815 \$5,661 \$5,508 \$7,371 \$7,180 \$6,989 \$5,643 \$5,494 \$5,346 \$2,139 \$2,103 \$2,066 \$2,501 \$2,435 \$2,369 \$3,171 \$3,089 \$3,007 \$2,427 \$2,364 \$2,300 \$7,112 \$6,991 \$6,870 \$8,316 \$8,097 \$7,877 \$10,542 \$10,269 \$9,996 \$8,070 \$7,858 \$7,646 2018 \$3,978 \$3,910 \$3,843 \$4,652 \$4,529 \$4,406 \$5,897 \$5,744 \$5,591 \$4,514 \$4,396 \$4,277 \$2,066 \$2,031 \$1,996 \$2,416 \$2,352 \$2,288 \$3,063 \$2,983 \$2,904 \$2,344 \$2,283 \$2,221 \$6,044 \$5,941 \$5,838 \$7,068 \$6,881 \$6,694 \$8,960 \$8,727 \$8,495 \$6,858 \$6,678 \$6,498 2019 \$3,978 \$3,910 \$3,843 \$4,652 \$4,529 \$4,406 \$5,897 \$5,744 \$5,591 \$4,514 \$4,396 \$4,277 \$2,066 \$2,031 \$1,996 \$2,416 \$2,352 \$2,288 \$3,063 \$2,983 \$2,904 \$2,344 \$2,283 \$2,221 \$6,044 \$5,941 \$5,838 \$7,068 \$6,881 \$6,694 \$8,960 \$8,727 \$8,495 \$6,858 \$6,678 \$6,498 2020 \$3,182 \$3,128 \$3,074 \$3,722 \$3,623 \$3,525 \$4,718 \$4,595 \$4,473 \$3,611 \$3,516 \$3,422 \$2,007 \$1,973 \$1,939 \$2,347 \$2,285 \$2,223 \$2,976 \$2,899 \$2,821 \$2,278 \$2,218 \$2,158 \$5,190 \$5,102 \$5,013 \$6,069 \$5,908 \$5,748 \$7,693 \$7,494 \$7,294 \$5,889 \$5,734 \$5,580 2021 \$3,182 \$3,128 \$3,074 \$3,722 \$3,623 \$3,525 \$4,718 \$4,595 \$4,473 \$3,611 \$3,516 \$3,422 \$2,007 \$1,973 \$1,939 \$2,347 \$2,285 \$2,223 \$2,976 \$2,899 \$2,821 \$2,278 \$2,218 \$2,158 \$5,190 \$5,102 \$5,013 \$6,069 \$5,908 \$5,748 \$7,693 \$7,494 \$7,294 \$5,889 \$5,734 \$5,580 2022 \$3,182 \$3,128 \$3,074 \$3,722 \$3,623 \$3,525 \$4,718 \$4,595 \$4,473 \$3,611 \$3,516 \$3,422 \$2,007 \$1,973 \$1,939 \$2,347 \$2,285 \$2,223 \$2,976 \$2,899 \$2,821 \$2,278 \$2,218 \$2,158 \$5,190 \$5,102 \$5,013 \$6,069 \$5,908 \$5,748 \$7,693 \$7,494 \$7,294 \$5,889 \$5,734 \$5,580 2023 \$3,182 \$3,128 \$3,074 \$3,722 \$3,623 \$3,525 \$4,718 \$4,595 \$4,473 \$3,611 \$3,516 \$3,422 \$2,007 \$1,973 \$1,939 \$2,347 \$2,285 \$2,223 \$2,976 \$2,899 \$2,821 \$2,278 \$2,218 \$2,158 \$5,190 \$5,102 \$5,013 \$6,069 \$5,908 \$5,748 \$7,693 \$7,494 \$7,294 \$5,889 \$5,734 \$5,580 2024 \$3,182 \$3,128 \$3,074 \$3,722 \$3,623 \$3,525 \$4,718 \$4,595 \$4,473 \$3,611 \$3,516 \$3,422 \$2,007 \$1,973 \$1,939 \$2,347 \$2,285 \$2,223 \$2,976 \$2,899 \$2,821 \$2,278 \$2,218 \$2,158 \$5,190 \$5,102 \$5,013 \$6,069 \$5,908 \$5,748 \$7,693 \$7,494 \$7,294 \$5,889 \$5,734 \$5,580 2025 \$2,546 \$2,503 \$2,459 \$2,977 \$2,899 \$2,820 \$3,774 \$3,676 \$3,578 \$2,889 \$2,813 \$2,737 \$1,264 \$1,243 \$1,221 \$1,479 \$1,439 \$1,400 \$1,874 \$1,826 \$1,777 \$1,435 \$1,397 \$1,359 \$3,810 \$3,746 \$3,681 \$4,456 \$4,338 \$4,220 \$5,648 \$5,502 \$5,356 \$4,324 \$4,210 \$4,097 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3-165 ------- Technologies Considered in the Agencies' Analysis Table 3-81 Costs for PHEV40 Battery Packs for the 2008 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC Vehicle class Small car Small car Standard car Standard car Large car Large car Small MPV Small MPV Small car Small car Standard car Standard car Large car Large car Small MPV Small MPV Small car Small car Standard car Standard car Large car Large car Small MPV Small MPV Applied WR 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% Net WR 2% 7% 3% 8% 1% 6% 3% 8% 2% 7% 3% 8% 1% 6% 3% 8% 2% 7% 3% 8% 1% 6% 3% 8% 2017 \$7,063 \$6,915 \$8,443 \$8,229 \$11,646 \$11,187 \$8,179 \$7,988 \$3,038 \$2,975 \$3,632 \$3,540 \$5,010 \$4,813 \$3,519 \$3,436 \$10,101 \$9,889 \$12,075 \$11,769 \$16,656 \$16,000 \$11,698 \$11,425 2018 \$5,650 \$5,532 \$6,754 \$6,583 \$9,317 \$8,950 \$6,544 \$6,391 \$2,934 \$2,873 \$3,508 \$3,419 \$4,838 \$4,648 \$3,398 \$3,319 \$8,584 \$8,404 \$10,262 \$10,002 \$14,155 \$13,597 \$9,942 \$9,709 2019 \$5,650 \$5,532 \$6,754 \$6,583 \$9,317 \$8,950 \$6,544 \$6,391 \$2,934 \$2,873 \$3,508 \$3,419 \$4,838 \$4,648 \$3,398 \$3,319 \$8,584 \$8,404 \$10,262 \$10,002 \$14,155 \$13,597 \$9,942 \$9,709 2020 \$4,520 \$4,425 \$5,404 \$5,266 \$7,453 \$7,160 \$5,235 \$5,113 \$2,851 \$2,791 \$3,408 \$3,322 \$4,701 \$4,516 \$3,302 \$3,225 \$7,371 \$7,217 \$8,812 \$8,588 \$12,155 \$11,676 \$8,537 \$8,337 2021 \$4,520 \$4,425 \$5,404 \$5,266 \$7,453 \$7,160 \$5,235 \$5,113 \$2,851 \$2,791 \$3,408 \$3,322 \$4,701 \$4,516 \$3,302 \$3,225 \$7,371 \$7,217 \$8,812 \$8,588 \$12,155 \$11,676 \$8,537 \$8,337 2022 \$4,520 \$4,425 \$5,404 \$5,266 \$7,453 \$7,160 \$5,235 \$5,113 \$2,851 \$2,791 \$3,408 \$3,322 \$4,701 \$4,516 \$3,302 \$3,225 \$7,371 \$7,217 \$8,812 \$8,588 \$12,155 \$11,676 \$8,537 \$8,337 2023 \$4,520 \$4,425 \$5,404 \$5,266 \$7,453 \$7,160 \$5,235 \$5,113 \$2,851 \$2,791 \$3,408 \$3,322 \$4,701 \$4,516 \$3,302 \$3,225 \$7,371 \$7,217 \$8,812 \$8,588 \$12,155 \$11,676 \$8,537 \$8,337 2024 \$4,520 \$4,425 \$5,404 \$5,266 \$7,453 \$7,160 \$5,235 \$5,113 \$2,851 \$2,791 \$3,408 \$3,322 \$4,701 \$4,516 \$3,302 \$3,225 \$7,371 \$7,217 \$8,812 \$8,588 \$12,155 \$11,676 \$8,537 \$8,337 2025 \$3,616 \$3,540 \$4,323 \$4,213 \$5,963 \$5,728 \$4,188 \$4,090 \$1,796 \$1,758 \$2,147 \$2,092 \$2,961 \$2,844 \$2,080 \$2,031 \$5,412 \$5,298 \$6,470 \$6,305 \$8,924 \$8,572 \$6,268 \$6,121 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-82 Costs for PHEV40 Battery Packs for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC Vehicle class Small car Small car Standard car Standard car Large car Applied WR 15% 20% 15% 20% 15% Net WR 3% 8% 3% 8% 2% 2017 \$7,175 \$7,018 \$8,642 \$8,418 \$11,862 2018 \$5,740 \$5,614 \$6,914 \$6,735 \$9,490 2019 \$5,740 \$5,614 \$6,914 \$6,735 \$9,490 2020 \$4,592 \$4,491 \$5,531 \$5,388 \$7,592 2021 \$4,592 \$4,491 \$5,531 \$5,388 \$7,592 2022 \$4,592 \$4,491 \$5,531 \$5,388 \$7,592 2023 \$4,592 \$4,491 \$5,531 \$5,388 \$7,592 2024 \$4,592 \$4,491 \$5,531 \$5,388 \$7,592 2025 \$3,674 \$3,593 \$4,425 \$4,310 \$6,073 3-166 ------- Technologies Considered in the Agencies' Analysis DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC Large car Small MPV Small MPV Small car Small car Standard car Standard car Large car Large car Small MPV Small MPV Small car Small car Standard car Standard car Large car Large car Small MPV Small MPV 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 7% 3% 8% 3% 8% 3% 8% 2% 7% 3% 8% 3% 8% 3% 8% 2% 7% 3% 8% \$11,450 \$8,378 \$8,180 \$3,087 \$3,019 \$3,718 \$3,622 \$5,103 \$4,926 \$3,604 \$3,519 \$10,262 \$10,037 \$12,360 \$12,040 \$16,965 \$16,376 \$11,982 \$11,700 \$9,160 \$6,702 \$6,544 \$2,981 \$2,916 \$3,591 \$3,498 \$4,928 \$4,757 \$3,481 \$3,399 \$8,721 \$8,530 \$10,504 \$10,232 \$14,418 \$13,918 \$10,183 \$9,943 \$9,160 \$6,702 \$6,544 \$2,981 \$2,916 \$3,591 \$3,498 \$4,928 \$4,757 \$3,481 \$3,399 \$8,721 \$8,530 \$10,504 \$10,232 \$14,418 \$13,918 \$10,183 \$9,943 \$7,328 \$5,362 \$5,235 \$2,896 \$2,833 \$3,489 \$3,398 \$4,789 \$4,622 \$3,382 \$3,302 \$7,488 \$7,324 \$9,020 \$8,786 \$12,380 \$11,951 \$8,744 \$8,538 \$7,328 \$5,362 \$5,235 \$2,896 \$2,833 \$3,489 \$3,398 \$4,789 \$4,622 \$3,382 \$3,302 \$7,488 \$7,324 \$9,020 \$8,786 \$12,380 \$11,951 \$8,744 \$8,538 \$7,328 \$5,362 \$5,235 \$2,896 \$2,833 \$3,489 \$3,398 \$4,789 \$4,622 \$3,382 \$3,302 \$7,488 \$7,324 \$9,020 \$8,786 \$12,380 \$11,951 \$8,744 \$8,538 \$7,328 \$5,362 \$5,235 \$2,896 \$2,833 \$3,489 \$3,398 \$4,789 \$4,622 \$3,382 \$3,302 \$7,488 \$7,324 \$9,020 \$8,786 \$12,380 \$11,951 \$8,744 \$8,538 \$7,328 \$5,362 \$5,235 \$2,896 \$2,833 \$3,489 \$3,398 \$4,789 \$4,622 \$3,382 \$3,302 \$7,488 \$7,324 \$9,020 \$8,786 \$12,380 \$11,951 \$8,744 \$8,538 \$5,863 \$4,290 \$4,188 \$1,824 \$1,784 \$2,197 \$2,141 \$3,016 \$2,911 \$2,130 \$2,080 \$5,498 \$5,377 \$6,622 \$6,451 \$9,090 \$8,774 \$6,420 \$6,268 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-83 Costs for EV75 Battery Packs for the 2008 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% Net WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 9% 14% 19% 10% 15% 2017 \$9,658 \$9,476 \$9,294 \$11,226 \$10,957 \$10,688 \$14,236 \$13,811 \$13,385 \$11,350 \$11,149 \$10,947 \$4,155 \$4,076 2018 \$7,726 \$7,581 \$7,436 \$8,980 \$8,765 \$8,550 \$11,389 \$11,049 \$10,708 \$9,080 \$8,919 \$8,758 \$4,012 \$3,937 2019 \$7,726 \$7,581 \$7,436 \$8,980 \$8,765 \$8,550 \$11,389 \$11,049 \$10,708 \$9,080 \$8,919 \$8,758 \$4,012 \$3,937 2020 \$6,181 \$6,065 \$5,948 \$7,184 \$7,012 \$6,840 \$9,111 \$8,839 \$8,567 \$7,264 \$7,135 \$7,006 \$3,899 \$3,825 2021 \$6,181 \$6,065 \$5,948 \$7,184 \$7,012 \$6,840 \$9,111 \$8,839 \$8,567 \$7,264 \$7,135 \$7,006 \$3,899 \$3,825 2022 \$6,181 \$6,065 \$5,948 \$7,184 \$7,012 \$6,840 \$9,111 \$8,839 \$8,567 \$7,264 \$7,135 \$7,006 \$3,899 \$3,825 2023 \$6,181 \$6,065 \$5,948 \$7,184 \$7,012 \$6,840 \$9,111 \$8,839 \$8,567 \$7,264 \$7,135 \$7,006 \$3,899 \$3,825 2024 \$6,181 \$6,065 \$5,948 \$7,184 \$7,012 \$6,840 \$9,111 \$8,839 \$8,567 \$7,264 \$7,135 \$7,006 \$3,899 \$3,825 2025 \$4,945 \$4,852 \$4,759 \$5,747 \$5,610 \$5,472 \$7,289 \$7,071 \$6,853 \$5,811 \$5,708 \$5,605 \$2,456 \$2,409 3-167 ------- Technologies Considered in the Agencies' Analysis 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 20% 10% 15% 20% 10% 15% 20% 9% 14% 19% 10% 15% 20% 10% 15% 20% 10% 15% 20% 9% 14% 19% \$3,998 \$4,829 \$4,713 \$4,598 \$6,124 \$5,941 \$5,758 \$4,883 \$4,796 \$4,709 \$13,812 \$13,552 \$13,293 \$16,055 \$15,670 \$15,285 \$20,360 \$19,752 \$19,144 \$16,232 \$15,945 \$15,657 \$3,861 \$4,664 \$4,552 \$4,440 \$5,915 \$5,738 \$5,561 \$4,715 \$4,632 \$4,548 \$11,738 \$11,518 \$11,297 \$13,644 \$13,317 \$12,990 \$17,303 \$16,786 \$16,269 \$13,795 \$13,551 \$13,306 \$3,861 \$4,664 \$4,552 \$4,440 \$5,915 \$5,738 \$5,561 \$4,715 \$4,632 \$4,548 \$11,738 \$11,518 \$11,297 \$13,644 \$13,317 \$12,990 \$17,303 \$16,786 \$16,269 \$13,795 \$13,551 \$13,306 \$3,752 \$4,532 \$4,423 \$4,314 \$5,747 \$5,575 \$5,403 \$4,582 \$4,501 \$4,419 \$10,079 \$9,890 \$9,700 \$11,716 \$11,435 \$11,154 \$14,858 \$14,414 \$13,970 \$11,846 \$11,636 \$11,426 \$3,752 \$4,532 \$4,423 \$4,314 \$5,747 \$5,575 \$5,403 \$4,582 \$4,501 \$4,419 \$10,079 \$9,890 \$9,700 \$11,716 \$11,435 \$11,154 \$14,858 \$14,414 \$13,970 \$11,846 \$11,636 \$11,426 \$3,752 \$4,532 \$4,423 \$4,314 \$5,747 \$5,575 \$5,403 \$4,582 \$4,501 \$4,419 \$10,079 \$9,890 \$9,700 \$11,716 \$11,435 \$11,154 \$14,858 \$14,414 \$13,970 \$11,846 \$11,636 \$11,426 \$3,752 \$4,532 \$4,423 \$4,314 \$5,747 \$5,575 \$5,403 \$4,582 \$4,501 \$4,419 \$10,079 \$9,890 \$9,700 \$11,716 \$11,435 \$11,154 \$14,858 \$14,414 \$13,970 \$11,846 \$11,636 \$11,426 \$3,752 \$4,532 \$4,423 \$4,314 \$5,747 \$5,575 \$5,403 \$4,582 \$4,501 \$4,419 \$10,079 \$9,890 \$9,700 \$11,716 \$11,435 \$11,154 \$14,858 \$14,414 \$13,970 \$11,846 \$11,636 \$11,426 \$2,363 \$2,854 \$2,786 \$2,717 \$3,620 \$3,512 \$3,403 \$2,886 \$2,835 \$2,784 \$7,400 \$7,261 \$7,122 \$8,602 \$8,396 \$8,190 \$10,909 \$10,583 \$10,257 \$8,697 \$8,543 \$8,389 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-84 Costs for EV75 Battery Packs for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% Net WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 9% 14% 19% 2017 \$9,886 \$9,718 \$9,551 \$11,456 \$11,174 \$10,892 \$14,592 \$14,148 \$13,704 \$11,470 \$11,260 \$11,049 2018 \$7,909 \$7,775 \$7,640 \$9,164 \$8,939 \$8,713 \$11,673 \$11,318 \$10,964 \$9,176 \$9,008 \$8,840 2019 \$7,909 \$7,775 \$7,640 \$9,164 \$8,939 \$8,713 \$11,673 \$11,318 \$10,964 \$9,176 \$9,008 \$8,840 2020 \$6,327 \$6,220 \$6,112 \$7,332 \$7,151 \$6,971 \$9,339 \$9,055 \$8,771 \$7,341 \$7,206 \$7,072 2021 \$6,327 \$6,220 \$6,112 \$7,332 \$7,151 \$6,971 \$9,339 \$9,055 \$8,771 \$7,341 \$7,206 \$7,072 2022 \$6,327 \$6,220 \$6,112 \$7,332 \$7,151 \$6,971 \$9,339 \$9,055 \$8,771 \$7,341 \$7,206 \$7,072 2023 \$6,327 \$6,220 \$6,112 \$7,332 \$7,151 \$6,971 \$9,339 \$9,055 \$8,771 \$7,341 \$7,206 \$7,072 2024 \$6,327 \$6,220 \$6,112 \$7,332 \$7,151 \$6,971 \$9,339 \$9,055 \$8,771 \$7,341 \$7,206 \$7,072 2025 \$5,062 \$4,976 \$4,890 \$5,865 \$5,721 \$5,577 \$7,471 \$7,244 \$7,017 \$5,873 \$5,765 \$5,657 3-168 ------- Technologies Considered in the Agencies' Analysis 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 9% 14% 19% 10% 15% 20% 10% 15% 20% 10% 15% 20% 9% 14% 19% \$4,253 \$4,181 \$4,109 \$4,928 \$4,807 \$4,685 \$6,277 \$6,086 \$5,895 \$4,934 \$4,844 \$4,753 \$14,139 \$13,899 \$13,659 \$16,384 \$15,980 \$15,577 \$20,869 \$20,234 \$19,600 \$16,405 \$16,104 \$15,803 \$4,107 \$4,038 \$3,968 \$4,759 \$4,642 \$4,525 \$6,062 \$5,878 \$5,694 \$4,765 \$4,678 \$4,591 \$12,016 \$11,812 \$11,608 \$13,924 \$13,581 \$13,238 \$17,736 \$17,196 \$16,657 \$13,942 \$13,686 \$13,430 \$4,107 \$4,038 \$3,968 \$4,759 \$4,642 \$4,525 \$6,062 \$5,878 \$5,694 \$4,765 \$4,678 \$4,591 \$12,016 \$11,812 \$11,608 \$13,924 \$13,581 \$13,238 \$17,736 \$17,196 \$16,657 \$13,942 \$13,686 \$13,430 \$3,991 \$3,923 \$3,855 \$4,624 \$4,511 \$4,397 \$5,890 \$5,711 \$5,532 \$4,630 \$4,545 \$4,460 \$10,318 \$10,143 \$9,968 \$11,956 \$11,662 \$11,367 \$15,229 \$14,766 \$14,303 \$11,971 \$11,752 \$11,532 \$3,991 \$3,923 \$3,855 \$4,624 \$4,511 \$4,397 \$5,890 \$5,711 \$5,532 \$4,630 \$4,545 \$4,460 \$10,318 \$10,143 \$9,968 \$11,956 \$11,662 \$11,367 \$15,229 \$14,766 \$14,303 \$11,971 \$11,752 \$11,532 \$3,991 \$3,923 \$3,855 \$4,624 \$4,511 \$4,397 \$5,890 \$5,711 \$5,532 \$4,630 \$4,545 \$4,460 \$10,318 \$10,143 \$9,968 \$11,956 \$11,662 \$11,367 \$15,229 \$14,766 \$14,303 \$11,971 \$11,752 \$11,532 \$3,991 \$3,923 \$3,855 \$4,624 \$4,511 \$4,397 \$5,890 \$5,711 \$5,532 \$4,630 \$4,545 \$4,460 \$10,318 \$10,143 \$9,968 \$11,956 \$11,662 \$11,367 \$15,229 \$14,766 \$14,303 \$11,971 \$11,752 \$11,532 \$3,991 \$3,923 \$3,855 \$4,624 \$4,511 \$4,397 \$5,890 \$5,711 \$5,532 \$4,630 \$4,545 \$4,460 \$10,318 \$10,143 \$9,968 \$11,956 \$11,662 \$11,367 \$15,229 \$14,766 \$14,303 \$11,971 \$11,752 \$11,532 \$2,514 \$2,471 \$2,428 \$2,913 \$2,841 \$2,769 \$3,710 \$3,597 \$3,485 \$2,917 \$2,863 \$2,809 \$7,575 \$7,447 \$7,318 \$8,778 \$8,562 \$8,346 \$11,181 \$10,841 \$10,501 \$8,789 \$8,628 \$8,467 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-85 Costs for EV100 Battery Packs for the 2008 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% Net WR 4% 9% 14% 4% 9% 14% 5% 10% 15% 3% 2017 \$11,774 \$11,563 \$11,351 \$13,550 \$13,261 \$12,973 \$16,403 \$15,949 \$15,495 \$14,089 2018 \$9,420 \$9,250 \$9,081 \$10,840 \$10,609 \$10,378 \$13,122 \$12,759 \$12,396 \$11,271 2019 \$9,420 \$9,250 \$9,081 \$10,840 \$10,609 \$10,378 \$13,122 \$12,759 \$12,396 \$11,271 2020 \$7,536 \$7,400 \$7,265 \$8,672 \$8,487 \$8,302 \$10,498 \$10,207 \$9,917 \$9,017 2021 \$7,536 \$7,400 \$7,265 \$8,672 \$8,487 \$8,302 \$10,498 \$10,207 \$9,917 \$9,017 2022 \$7,536 \$7,400 \$7,265 \$8,672 \$8,487 \$8,302 \$10,498 \$10,207 \$9,917 \$9,017 2023 \$7,536 \$7,400 \$7,265 \$8,672 \$8,487 \$8,302 \$10,498 \$10,207 \$9,917 \$9,017 2024 \$7,536 \$7,400 \$7,265 \$8,672 \$8,487 \$8,302 \$10,498 \$10,207 \$9,917 \$9,017 2025 \$6,028 \$5,920 \$5,812 \$6,938 \$6,790 \$6,642 \$8,398 \$8,166 \$7,933 \$7,214 3-169 ------- Technologies Considered in the Agencies' Analysis DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 8% 13% 4% 9% 14% 4% 9% 14% 5% 10% 15% 3% 8% 13% 4% 9% 14% 4% 9% 14% 5% 10% 15% 3% 8% 13% \$13,830 \$13,572 \$5,065 \$4,974 \$4,883 \$5,829 \$5,705 \$5,581 \$7,056 \$6,861 \$6,666 \$6,061 \$5,950 \$5,838 \$16,840 \$16,537 \$16,234 \$19,380 \$18,966 \$18,553 \$23,459 \$22,810 \$22,161 \$20,150 \$19,780 \$19,410 \$11,064 \$10,857 \$4,892 \$4,804 \$4,716 \$5,630 \$5,510 \$5,390 \$6,815 \$6,626 \$6,438 \$5,853 \$5,746 \$5,638 \$14,311 \$14,054 \$13,797 \$16,470 \$16,119 \$15,768 \$19,937 \$19,385 \$18,833 \$17,125 \$16,810 \$16,496 \$11,064 \$10,857 \$4,892 \$4,804 \$4,716 \$5,630 \$5,510 \$5,390 \$6,815 \$6,626 \$6,438 \$5,853 \$5,746 \$5,638 \$14,311 \$14,054 \$13,797 \$16,470 \$16,119 \$15,768 \$19,937 \$19,385 \$18,833 \$17,125 \$16,810 \$16,496 \$8,851 \$8,686 \$4,753 \$4,668 \$4,582 \$5,470 \$5,353 \$5,237 \$6,621 \$6,438 \$6,255 \$5,687 \$5,583 \$5,479 \$12,289 \$12,068 \$11,847 \$14,142 \$13,841 \$13,539 \$17,119 \$16,645 \$16,172 \$14,705 \$14,434 \$14,164 \$8,851 \$8,686 \$4,753 \$4,668 \$4,582 \$5,470 \$5,353 \$5,237 \$6,621 \$6,438 \$6,255 \$5,687 \$5,583 \$5,479 \$12,289 \$12,068 \$11,847 \$14,142 \$13,841 \$13,539 \$17,119 \$16,645 \$16,172 \$14,705 \$14,434 \$14,164 \$8,851 \$8,686 \$4,753 \$4,668 \$4,582 \$5,470 \$5,353 \$5,237 \$6,621 \$6,438 \$6,255 \$5,687 \$5,583 \$5,479 \$12,289 \$12,068 \$11,847 \$14,142 \$13,841 \$13,539 \$17,119 \$16,645 \$16,172 \$14,705 \$14,434 \$14,164 \$8,851 \$8,686 \$4,753 \$4,668 \$4,582 \$5,470 \$5,353 \$5,237 \$6,621 \$6,438 \$6,255 \$5,687 \$5,583 \$5,479 \$12,289 \$12,068 \$11,847 \$14,142 \$13,841 \$13,539 \$17,119 \$16,645 \$16,172 \$14,705 \$14,434 \$14,164 \$8,851 \$8,686 \$4,753 \$4,668 \$4,582 \$5,470 \$5,353 \$5,237 \$6,621 \$6,438 \$6,255 \$5,687 \$5,583 \$5,479 \$12,289 \$12,068 \$11,847 \$14,142 \$13,841 \$13,539 \$17,119 \$16,645 \$16,172 \$14,705 \$14,434 \$14,164 \$7,081 \$6,949 \$2,994 \$2,940 \$2,886 \$3,445 \$3,372 \$3,298 \$4,171 \$4,055 \$3,940 \$3,582 \$3,517 \$3,451 \$9,022 \$8,860 \$8,698 \$10,383 \$10,162 \$9,940 \$12,569 \$12,221 \$11,873 \$10,796 \$10,598 \$10,399 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-86 Costs for EV100 Battery Packs for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% Net WR 4% 9% 14% 4% 9% 14% 5% 10% 15% 3% 8% 13% 4% 9% 2017 \$12,046 \$11,831 \$11,615 \$13,794 \$13,512 \$13,231 \$16,845 \$16,382 \$15,919 \$14,246 \$13,982 \$13,718 \$5,182 \$5,089 2018 \$9,637 \$9,465 \$9,292 \$11,035 \$10,810 \$10,585 \$13,476 \$13,106 \$12,735 \$11,397 \$11,185 \$10,974 \$5,005 \$4,915 2019 \$9,637 \$9,465 \$9,292 \$11,035 \$10,810 \$10,585 \$13,476 \$13,106 \$12,735 \$11,397 \$11,185 \$10,974 \$5,005 \$4,915 2020 \$7,710 \$7,572 \$7,434 \$8,828 \$8,648 \$8,468 \$10,781 \$10,484 \$10,188 \$9,117 \$8,948 \$8,779 \$4,863 \$4,776 2021 \$7,710 \$7,572 \$7,434 \$8,828 \$8,648 \$8,468 \$10,781 \$10,484 \$10,188 \$9,117 \$8,948 \$8,779 \$4,863 \$4,776 2022 \$7,710 \$7,572 \$7,434 \$8,828 \$8,648 \$8,468 \$10,781 \$10,484 \$10,188 \$9,117 \$8,948 \$8,779 \$4,863 \$4,776 2023 \$7,710 \$7,572 \$7,434 \$8,828 \$8,648 \$8,468 \$10,781 \$10,484 \$10,188 \$9,117 \$8,948 \$8,779 \$4,863 \$4,776 2024 \$7,710 \$7,572 \$7,434 \$8,828 \$8,648 \$8,468 \$10,781 \$10,484 \$10,188 \$9,117 \$8,948 \$8,779 \$4,863 \$4,776 2025 \$6,168 \$6,057 \$5,947 \$7,062 \$6,918 \$6,774 \$8,625 \$8,388 \$8,150 \$7,294 \$7,159 \$7,023 \$3,063 \$3,008 3-170 ------- Technologies Considered in the Agencies' Analysis 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 14% 4% 9% 14% 5% 10% 15% 3% 8% 13% 4% 9% 14% 4% 9% 14% 5% 10% 15% 3% 8% 13% \$4,997 \$5,934 \$5,813 \$5,692 \$7,247 \$7,047 \$6,848 \$6,128 \$6,015 \$5,901 \$17,229 \$16,920 \$16,612 \$19,728 \$19,325 \$18,923 \$24,092 \$23,429 \$22,767 \$20,375 \$19,997 \$19,619 \$4,826 \$5,731 \$5,614 \$5,497 \$6,999 \$6,806 \$6,614 \$5,919 \$5,809 \$5,699 \$14,642 \$14,380 \$14,118 \$16,766 \$16,424 \$16,082 \$20,475 \$19,912 \$19,348 \$17,316 \$16,994 \$16,673 \$4,826 \$5,731 \$5,614 \$5,497 \$6,999 \$6,806 \$6,614 \$5,919 \$5,809 \$5,699 \$14,642 \$14,380 \$14,118 \$16,766 \$16,424 \$16,082 \$20,475 \$19,912 \$19,348 \$17,316 \$16,994 \$16,673 \$4,689 \$5,568 \$5,455 \$5,341 \$6,800 \$6,613 \$6,426 \$5,751 \$5,644 \$5,538 \$12,573 \$12,348 \$12,122 \$14,396 \$14,103 \$13,809 \$17,581 \$17,097 \$16,614 \$14,868 \$14,593 \$14,317 \$4,689 \$5,568 \$5,455 \$5,341 \$6,800 \$6,613 \$6,426 \$5,751 \$5,644 \$5,538 \$12,573 \$12,348 \$12,122 \$14,396 \$14,103 \$13,809 \$17,581 \$17,097 \$16,614 \$14,868 \$14,593 \$14,317 \$4,689 \$5,568 \$5,455 \$5,341 \$6,800 \$6,613 \$6,426 \$5,751 \$5,644 \$5,538 \$12,573 \$12,348 \$12,122 \$14,396 \$14,103 \$13,809 \$17,581 \$17,097 \$16,614 \$14,868 \$14,593 \$14,317 \$4,689 \$5,568 \$5,455 \$5,341 \$6,800 \$6,613 \$6,426 \$5,751 \$5,644 \$5,538 \$12,573 \$12,348 \$12,122 \$14,396 \$14,103 \$13,809 \$17,581 \$17,097 \$16,614 \$14,868 \$14,593 \$14,317 \$4,689 \$5,568 \$5,455 \$5,341 \$6,800 \$6,613 \$6,426 \$5,751 \$5,644 \$5,538 \$12,573 \$12,348 \$12,122 \$14,396 \$14,103 \$13,809 \$17,581 \$17,097 \$16,614 \$14,868 \$14,593 \$14,317 \$2,953 \$3,507 \$3,436 \$3,364 \$4,283 \$4,165 \$4,048 \$3,622 \$3,555 \$3,488 \$9,231 \$9,066 \$8,900 \$10,570 \$10,354 \$10,138 \$12,908 \$12,553 \$12,198 \$10,916 \$10,714 \$10,511 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-87 Costs for EV150 Battery Packs for the 2008 Baseline (2010\$) Cost type DMC DMC DMC DMC 1C 1C 1C 1C TC TC TC TC Vehicle class Small car Standard car Large car Small MPV Small car Standard car Large car Small MPV Small car Standard car Large car Small MPV Applied WR 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% Net WR 2% 2% 3% 1% 2% 2% 3% 1% 2% 2% 3% 1% 2017 \$15,701 \$18,950 \$21,552 \$19,744 \$6,755 \$8,152 \$9,272 \$8,493 \$22,456 \$27,102 \$30,824 \$28,237 2018 \$12,561 \$15,160 \$17,242 \$15,795 \$6,523 \$7,873 \$8,954 \$8,203 \$19,084 \$23,033 \$26,196 \$23,998 2019 \$12,561 \$15,160 \$17,242 \$15,795 \$6,523 \$7,873 \$8,954 \$8,203 \$19,084 \$23,033 \$26,196 \$23,998 2020 \$10,049 \$12,128 \$13,793 \$12,636 \$6,338 \$7,650 \$8,700 \$7,970 \$16,387 \$19,777 \$22,494 \$20,606 2021 \$10,049 \$12,128 \$13,793 \$12,636 \$6,338 \$7,650 \$8,700 \$7,970 \$16,387 \$19,777 \$22,494 \$20,606 2022 \$10,049 \$12,128 \$13,793 \$12,636 \$6,338 \$7,650 \$8,700 \$7,970 \$16,387 \$19,777 \$22,494 \$20,606 2023 \$10,049 \$12,128 \$13,793 \$12,636 \$6,338 \$7,650 \$8,700 \$7,970 \$16,387 \$19,777 \$22,494 \$20,606 2024 \$10,049 \$12,128 \$13,793 \$12,636 \$6,338 \$7,650 \$8,700 \$7,970 \$16,387 \$19,777 \$22,494 \$20,606 2025 \$8,039 \$9,702 \$11,035 \$10,109 \$3,992 \$4,818 \$5,480 \$5,020 \$12,031 \$14,520 \$16,515 \$15,129 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3-171 ------- Technologies Considered in the Agencies' Analysis Table 3-88 Costs for EV150 Battery Packs for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC 1C 1C 1C 1C TC TC TC TC Vehicle class Small car Standard car Large car Small MPV Small car Standard car Large car Small MPV Small car Standard car Large car Small MPV Applied WR 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% Net WR 2% 2% 3% 1% 2% 2% 3% 1% 2% 2% 3% 1% 2017 \$16,102 \$19,265 \$22,080 \$19,976 \$6,927 \$8,287 \$9,498 \$8,593 \$23,028 \$27,552 \$31,578 \$28,569 2018 \$12,881 \$15,412 \$17,664 \$15,981 \$6,690 \$8,004 \$9,173 \$8,299 \$19,571 \$23,415 \$26,837 \$24,280 2019 \$12,881 \$15,412 \$17,664 \$15,981 \$6,690 \$8,004 \$9,173 \$8,299 \$19,571 \$23,415 \$26,837 \$24,280 2020 \$10,305 \$12,329 \$14,131 \$12,784 \$6,500 \$7,777 \$8,913 \$8,064 \$16,805 \$20,106 \$23,044 \$20,848 2021 \$10,305 \$12,329 \$14,131 \$12,784 \$6,500 \$7,777 \$8,913 \$8,064 \$16,805 \$20,106 \$23,044 \$20,848 2022 \$10,305 \$12,329 \$14,131 \$12,784 \$6,500 \$7,777 \$8,913 \$8,064 \$16,805 \$20,106 \$23,044 \$20,848 2023 \$10,305 \$12,329 \$14,131 \$12,784 \$6,500 \$7,777 \$8,913 \$8,064 \$16,805 \$20,106 \$23,044 \$20,848 2024 \$10,305 \$12,329 \$14,131 \$12,784 \$6,500 \$7,777 \$8,913 \$8,064 \$16,805 \$20,106 \$23,044 \$20,848 2025 \$8,244 \$9,863 \$11,305 \$10,228 \$4,094 \$4,898 \$5,614 \$5,079 \$12,338 \$14,762 \$16,919 \$15,307 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost For Mild HEV batteries, the agencies used a similar approach to estimating the cost of the battery pack but used a different approach to determining its size. Our Mild HEV system used in the analyses is based, largely, on the Buick eAssist system.37 According to the press releases, it includes a 15 kW motor and a 15 kW/0.5kWh/l 15 Volt two-module battery. For the agencies' analyses, a 15kW/0.25kWh/l 10 Volt single-module battery was selected for several reasons. First, the Buick system uses a 20% state-of-charge (SOC) swing for the battery. We believe that, in the 2017-2025 timeframe, a 40% SOC swing is reasonable. As such, the energy capacity of the battery can be halved (from 0.5 to 0.25 kWh).22 The 110V system used in the analysis is essentially the same as Buick's 115V system. The voltage change is due to our use of a 28 cell single-module battery pack rather than the 32 cell double- module battery pack which is used in the eAssist system. Such changes are consistent with our expectation that cells will increase in size allowing for fewer cells and fewer modules. Further, for the Mild HEV technology, the agencies are using the same system regardless of vehicle class or subclass. In other words, the Mild HEV system is a stand-alone technology that can be applied to any subclass without unique modifications for each class or subclass. As such, it adds more weight as a percentage to a smaller vehicle than to a larger vehicle but it provides more effectiveness to a smaller vehicle than to a larger vehicle. Since the same system is used regardless of vehicle class or subclass, the costs are identical regardless of vehicle class or subclass. Using the ANL BatPaC model, the Mild HEV battery DMC was calculated as \$553 and is considered applicable to the MY 2017. The agencies derived the Mild HEV battery pack cost using the same methodology that was used for the P2 HEV yy "eAssist" is a Buick (or General Motors) term and is not a generic term for this technology, hence our use of the term mild hybrid. zz Note that projected battery cost is relatively insensitive to kWh capacity at the high power-to-energy ratio of these batteries. A 0.5 kWh battery could alternatively be specified at a similar cost. 3-172 ------- Technologies Considered in the Agencies' Analysis battery pack, and consider cost to be on the flat portion of the learning curve during the 2017- 2025 timeframe. The agencies have applied a highl complexity ICM of 1.56 through 2024 then 1.35 thereafter. The resultant Mild HEV battery pack costs are as shown in Table 3-89. The associated weight penalties are as shown in Table 3-90. Table 3-89 Costs for Mild Hybrid (MHEV) Battery Packs for both the 2008 and 2010 Baselines (2010\$) Cost type DMC 1C TC Vehicle class All All All 2017 \$553 \$312 \$865 2018 \$536 \$311 \$847 2019 \$520 \$310 \$830 2020 \$505 \$309 \$813 2021 \$490 \$308 \$797 2022 \$475 \$307 \$782 2023 \$461 \$306 \$766 2024 \$447 \$305 \$752 2025 \$433 \$187 \$621 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-90 EPA and NHTSA Weight Reduction Offset Associated with MHEV for both the 2008 and 2010 Baselines Vehicle class Small car Standard car Large car Small MPV Large MPV Truck Weight penalty 3.5% 3.0% 2.5% 2.5% 2.5% 2.0% The CAFE model does not use pre-built packages and it applies technologies incrementally as necessary to meet the fuel consumption reduction requirement, so the cost interaction between any particular technology and other technologies (cost synergies) must be defined. This allows flexibility so that when a technology is picked, the model will automatically look through the cost synergy defined in a table and apply cost adjustments accordingly. The total cost for mass reduction and electrification is composed of the following four parts: (1) Cost of net mass reduction; (2) Cost of electrification with zero mass reduction; (3) Mass reduction cost synergy for increased or decreased amount of mass reduction due to switching from conventional powertrain to electrification systems as defined in Figure 3-25. For an example, if a midsize passenger car needs both 10 percent net mass reduction and P2 hybrid to meet the CAFE target, the model will need to find the cost of additional 5 percent of mass reduction to consider the vehicle weight increase due to switching from conventional powertrain system to P2 electrification packages. This additional 5 percent of mass reduction is calculated starting from 10 percent mass reduction, not zero as shown in Figure 3-25 because mass reduction cost versus mass reduction percent is not a linear function. The cost increases faster as the 3-173 ------- Technologies Considered in the Agencies' Analysis amount of mass reduction becomes higher. (4) Electrification system cost synergies (battery and non-battery components) due to mass reduction as defined in Table 3-76 and Table 3-103: Continuing the example in the steps above, if a midsize passenger car needs both 10 percent net mass reduction and P2 hybrid to meet the CAFE target, after calculating the costs above, the model will need to find the cost of electrification systems, including battery system and non-battery system, with the required net amount of mass reduction using the equations in Table 3-76 and Table 3-103. Then the delta cost between this cost and the cost calculated in step 2, i.e. electrification system cost with zero applied mass reduction is calculated and treated as a cost synergy. These cost deltas are normally negative, i.e., a cost reduction, due to the downsizing of the electrification system resulting from mass reduction The sum of item (3) and (4) in the above list are calculated as cost synergies and stored in the cost synergy table as defined in NHTSA's RIA. Figure 3-25 Mass Reduction Cost Example for Applied and Net Mass Reduction \$600 5 \$500 +-* m 2 \$400 o '•g \$300 3 1 \$200 | \$100 \$- 0 Example of Applied and Net Mass Reduction Costs j? . ro* , ^>; o^' ^\ "f // ^ ^ ^ff ^ C0^%H ^^ K^^^ ^^ X fc 5% 10% 15'\. 20% Amount of Mass Reduction [%] The agencies have also carefully reconsidered the power and energy requirements for each electrified vehicle type, which has a significant impact on the cost estimates for HEVs, PHEVs, and EVs as compared to the estimates used in the 2012-2016 rulemaking. The agencies note that, for this analysis, the agencies have assumed batteries will be capable of lasting the lifetime of the vehicle, which is consistent with the expected customer demands from this technology (as manufacturers have confirmed). Lastly, the agencies have focused attention on an emerging HEV technology known as a P2-hybrid, a technology not considered in the 2012-2016 light-duty rule. 3-174 ------- Technologies Considered in the Agencies' Analysis The agencies have also considered, for this analysis, the costs associated with in-home chargers expected to be necessary for PHEVs and EVs. Further details on in-home chargers and their estimated costs are presented in Section 3.4.4. 3.4.3.10 Non-battery costs for MHEVs, HEVs, PHEVs, EVs and FCEVs This section addresses the costs of non-battery components which are required for electric drive vehicles. Some of these components are not found in every electric-drive vehicle (e.g. an HEV does not have an on-board battery charger as found in a PHEV or EV). Others are found in all electric drive vehicles and/or must be scaled to the vehicle type or class to properly represent the cost. The agencies derived the costs of these components from the FEV teardown study and the 2010 TAR. Where appropriate, costs were scaled to vehicle class and in the case of the motor and inverter, the sizing methodology used for battery sizing was applied. The electric drive motor and inverter provide the motive power for any electric-drive vehicle converting electrical energy from the battery into kinetic energy for propulsion. In an electric-drive vehicle, energy stored in the battery is routed to the inverter which converts it to a voltage and wave form that can be used by the motor. In many cases, such as HEVs, the combined cost of the motor and inverter exceed the battery cost. As batteries become larger in PHEVs and EVs, the battery cost grows faster than motor and inverter cost. For this analysis, the agencies used the vehicle power requirement calculation discussed in 3.4.3.8 to calculate the required motor and inverter size for each vehicle class at each weight reduction point. Then, for the HEVs and PHEVs, a regression was created from the FEV teardown data for motors and inverters and this regression was used to calculate the motor and inverter cost for each combination of vehicle class and weight reduction. This regression for use with the 2008 baseline was \$13.78x(motor size in kW)+\$781.50 (values in 2010\$), and for use with the 2010 baseline was \$14.13x(motor size in kW)+\$771.21 (values in 2010\$). The results are shown as the "Motor assembly" line item in Table 3-91 through Table 3-96, which show our scaled DMC for P2 HEV, PHEV20 and PHEV40, respectively, for both the 2008 and 2010 baselines. For EVs, the agencies used the motor and inverter cost regression from the 2010 TAR (see 2010 TAR at page B-21) and we used that regression for both the 2008 and 2010 baselines. Since the FEV teardown was conducted on an HEV Ford Fusion, the agencies believe the technology for an EV is different enough to warrant using the TAR regression. The regression presented in the TAR showed the DMC being equal to \$8.45x(motor size in kW)+\$185.05 (values in 2010\$). The results are presented as separate line items for "Motor inverter" and "Motor assembly" in Table 3-97 through Table 3-102, which show our scaled DMC for EV75, EV100 and EV150, respectively, for both the 2008 and 2010 baselines. In addition to electric drive motors and inverters, there are several other components in electric drive vehicles that are required. These components include the following: • Body Modifications which are required on HEVs and PHEVs include changes to sheet metal to accommodate electric drive components and the addition of fasteners to secure components such as electric cables. These costs come from the FEV teardown and are scaled by vehicle class. For EVs, these costs are assumed to be included in the base vehicle because they are less likely to be adapted from conventional vehicles. 3-175 ------- Technologies Considered in the Agencies' Analysis • Brake System changes include the addition of a braking system that can control the vehicle's regenerative braking system—a key enabler of electric drive vehicle efficiency. The brake system costs are from the FEV teardown and are scaled to vehicle class. • Climate Control System includes components such as an electric air conditioning compressor that enables operation while the engine is off for HEVs and PHEVs as well as for an EV which has no engine. Climate control system costs come from the FEV teardown and are scaled to vehicle class. • Conventional vehicle battery and alternator are deleted in these vehicles, for a cost savings, replaced by the DC-DC converter which converts the high-voltage traction battery to a nominal 12V DC to operate the vehicle's accessories. This credit comes from the FEV teardown study and is scaled to vehicle class. • DC-DC converter converts the high-voltage battery voltage to a nominal 12V battery voltage to run vehicle accessories such as the radio, lights and wipers. This cost comes from the FEV teardown study and is scaled to vehicle class. • Power distribution and Control consists of those components which route electricity to the motor, inverter and contains the controllers to operate and monitor the electric drive system. This cost applies to HEVs and PHEVs and comes from the FEV teardown study. It is scaled to vehicle class. • On-Vehicle Charger consists of the components necessary to charge a PHEV or EV from an outlet. It includes the charging port, wiring and electronics necessary to convert a 120V or 240V AC input to the high-voltage DC power necessary to charge the battery. Because the FEV teardown study subject vehicle did not have an on-vehicle charger, the costs from the TAR were used for this item. It is not scaled to vehicle class, however the EV charger is assumed to cost twice the amount of the PHEV charger to account for a higher current capacity. This cost does not include off-vehicle charger components which are discussed in Section 3.4.4, below. • Supplemental heating is required for passenger comfort on PHEVs and EVs which may operate for long periods with no engine heat available. This cost comes from the FEV teardown study and is scaled to vehicle class. The supplemental heater on the EV is assumed to be three times more costly than the PHEV because the entire cabin comfort is dependent on the supplemental heater. In a PHEV, it is assumed that in extreme conditions, the internal combustion engine will start to provide additional cabin heat and defrost functions. • High Voltage Wiring is an item used on EVs only. It includes the high voltage cabling from the battery to the inverter and motor as well as control components. It is equivalent to the power distribution and control used on HEVs and PHEVs and comes from the FEV teardown study. It is scaled to vehicle class. • Delete Internal Combustion Engine and Transmission For EVs, the engine and transmission are deleted and a credit is applied. These credits come from work done in support of the 2010 TAR and are scaled to vehicle class. • Battery Discharge System For HEVs, PHEVs and EVs, it is expected that manufacturers will provide the means to safely discharge battery packs following a vehicle 3-176 ------- Technologies Considered in the Agencies' Analysis crash. The agencies have assumed that this would include dedicated DC terminals, an access panel for the terminals, and a diagnostics port. The estimated cost of this capability is the same for all vehicle classes, but is different for HEVs than for PHEVs and EVs. The results of the scaling exercise applied to non-battery components are presented in Table 3-91 through Table 3-102 for P2 HEVs, PHEV20, PHEV40, EV75, EV100 and EV150, for the 2008 and 2010 baselines, respectively. Table 3-91 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for P2 HEV for the 2008 Baseline (2010\$) System 0%WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control Battery discharge system Motor assembly Total 2%WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control Battery discharge system Motor assembly Total 7.5% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control Battery discharge system Motor assembly Total 10% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control Battery discharge system Motor assembly Small car \$6 \$221 \$140 -\$60 \$121 \$196 \$6 \$1,045 \$1,675 \$6 \$221 \$140 -\$60 \$121 \$196 \$6 \$1,039 \$1,670 \$6 \$221 \$140 -\$60 \$121 \$196 \$6 \$1,025 \$1,655 \$6 \$221 \$140 -\$60 \$121 \$196 \$6 \$1,018 Standard car \$6 \$228 \$157 -\$65 \$152 \$201 \$6 \$1,172 \$1,857 \$6 \$228 \$157 -\$65 \$152 \$201 \$6 \$1,164 \$1,849 \$6 \$228 \$157 -\$65 \$152 \$201 \$6 \$1,143 \$1,828 \$6 \$228 \$157 -\$65 \$152 \$201 \$6 \$1,133 Large car \$6 \$231 \$168 -\$82 \$162 \$204 \$6 \$1,480 \$2,175 \$6 \$231 \$168 -\$82 \$162 \$204 \$6 \$1,467 \$2,161 \$6 \$231 \$168 -\$82 \$162 \$204 \$6 \$1,428 \$2,123 \$6 \$231 \$168 -\$82 \$162 \$204 \$6 \$1,411 Small MPV \$6 \$225 \$164 -\$86 \$152 \$200 \$6 \$1,112 \$1,777 \$6 \$225 \$164 -\$86 \$152 \$200 \$6 \$1,106 \$1,771 \$6 \$225 \$164 -\$86 \$152 \$200 \$6 \$1,088 \$1,752 \$6 \$225 \$164 -\$86 \$152 \$200 \$6 \$1,079 Large MPV \$6 \$233 \$250 -\$86 \$152 \$206 \$6 \$1,287 \$2,052 \$6 \$233 \$250 -\$86 \$152 \$206 \$6 \$1,277 \$2,042 \$6 \$233 \$250 -\$86 \$152 \$206 \$6 \$1,249 \$2,014 \$6 \$233 \$250 -\$86 \$152 \$206 \$6 \$1,237 Truck \$6 \$240 \$186 -\$94 \$177 \$220 \$6 \$1,429 \$2,169 \$6 \$240 \$186 -\$94 \$177 \$220 \$6 \$1,416 \$2,156 \$6 \$240 \$186 -\$94 \$177 \$220 \$6 \$1,381 \$2,121 \$6 \$240 \$186 -\$94 \$177 \$220 \$6 \$1,364 3-177 ------- Technologies Considered in the Agencies' Analysis Total 20% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control Battery discharge system Motor assembly Total \$1,649 \$6 \$221 \$140 -\$60 \$121 \$196 \$6 \$1,007 \$1,637 \$1,818 \$6 \$228 \$157 -\$65 \$152 \$201 \$6 \$1,115 \$1,800 \$2,105 \$6 \$231 \$168 -\$82 \$162 \$204 \$6 \$1,377 \$2,071 \$1,744 \$6 \$225 \$164 -\$86 \$152 \$200 \$6 \$1,064 \$1,729 \$2,002 \$6 \$233 \$250 -\$86 \$152 \$206 \$6 \$1,212 \$1,977 \$2,104 \$6 \$240 \$186 -\$94 \$177 \$220 \$6 \$1,337 \$2,077 Table 3-92 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for P2 HEV for the 2010 Baseline (2010\$) System 0%WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control Battery discharge system Motor assembly Total 2%WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control Battery discharge system Motor assembly Total 7.5% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control Battery discharge system Motor assembly Total 10% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control Small car \$6 \$223 \$140 -\$60 \$121 \$197 \$6 \$1,051 \$1,683 \$6 \$223 \$140 -\$60 \$121 \$197 \$6 \$1,045 \$1,677 \$6 \$223 \$140 -\$60 \$121 \$197 \$6 \$1,030 \$1,662 \$6 \$223 \$140 -\$60 \$121 \$197 Standard car \$6 \$229 \$157 -\$65 \$152 \$202 \$6 \$1,191 \$1,878 \$6 \$229 \$157 -\$65 \$152 \$202 \$6 \$1,183 \$1,869 \$6 \$229 \$157 -\$65 \$152 \$202 \$6 \$1,159 \$1,846 \$6 \$229 \$157 -\$65 \$152 \$202 Large car \$6 \$232 \$168 -\$82 \$177 \$205 \$6 \$1,512 \$2,224 \$6 \$232 \$168 -\$82 \$177 \$205 \$6 \$1,497 \$2,210 \$6 \$232 \$168 -\$82 \$177 \$205 \$6 \$1,457 \$2,169 \$6 \$232 \$168 -\$82 \$177 \$205 Small MPV \$6 \$225 \$164 -\$86 \$162 \$201 \$6 \$1,134 \$1,811 \$6 \$225 \$164 -\$86 \$162 \$201 \$6 \$1,127 \$1,804 \$6 \$225 \$164 -\$86 \$162 \$201 \$6 \$1,107 \$1,784 \$6 \$225 \$164 -\$86 \$162 \$201 Large MPV \$6 \$232 \$250 -\$86 \$162 \$206 \$6 \$1,299 \$2,073 \$6 \$232 \$250 -\$86 \$162 \$206 \$6 \$1,288 \$2,063 \$6 \$232 \$250 -\$86 \$162 \$206 \$6 \$1,259 \$2,034 \$6 \$232 \$250 -\$86 \$162 \$206 Truck \$6 \$242 \$186 -\$94 \$177 \$221 \$6 \$1,445 \$2,188 \$6 \$242 \$186 -\$94 \$177 \$221 \$6 \$1,432 \$2,175 \$6 \$242 \$186 -\$94 \$177 \$221 \$6 \$1,395 \$2,138 \$6 \$242 \$186 -\$94 \$177 \$221 3-178 ------- Technologies Considered in the Agencies' Analysis Battery discharge system Motor assembly Total 20% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control Battery discharge system Motor assembly Total \$6 \$1,023 \$1,655 \$6 \$223 \$140 -\$60 \$121 \$197 \$6 \$1,010 \$1,642 \$6 \$1,149 \$1,836 \$6 \$229 \$157 -\$65 \$152 \$202 \$6 \$1,129 \$1,816 \$6 \$1,438 \$2,150 \$6 \$232 \$168 -\$82 \$177 \$205 \$6 \$1,402 \$2,114 \$6 \$1,098 \$1,775 \$6 \$225 \$164 -\$86 \$162 \$201 \$6 \$1,081 \$1,757 \$6 \$1,246 \$2,021 \$6 \$232 \$250 -\$86 \$162 \$206 \$6 \$1,220 \$1,994 \$6 \$1,378 \$2,121 \$6 \$242 \$186 -\$94 \$177 \$221 \$6 \$1,350 \$2,093 Table 3-93 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for PHEV20 for the 2008 Baseline (2010\$) System 0% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 2% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 7.5% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Small car \$6 \$221 \$140 -\$60 \$121 \$196 \$105 \$38 \$2,097 \$13 \$2,878 \$6 \$221 \$140 -\$60 \$121 \$196 \$105 \$38 \$2,071 \$13 \$2,852 \$6 \$221 \$140 -\$60 \$121 \$196 \$105 \$38 \$1,999 \$13 Standard car \$6 \$228 \$157 -\$65 \$152 \$201 \$105 \$43 \$2,735 \$13 \$3,575 \$6 \$228 \$157 -\$65 \$152 \$201 \$105 \$43 \$2,695 \$13 \$3,536 \$6 \$228 \$157 -\$65 \$152 \$201 \$105 \$43 \$2,588 \$13 Large car \$6 \$231 \$168 -\$82 \$162 \$204 \$105 \$45 \$4,276 \$13 \$5,129 \$6 \$231 \$168 -\$82 \$162 \$204 \$105 \$45 \$4,207 \$13 \$5,059 \$6 \$231 \$168 -\$82 \$162 \$204 \$105 \$45 \$4,014 \$13 Small MPV \$6 \$225 \$164 -\$86 \$152 \$200 \$105 \$44 \$2,436 \$13 \$3,258 \$6 \$225 \$164 -\$86 \$152 \$200 \$105 \$44 \$2,403 \$13 \$3,225 \$6 \$225 \$164 -\$86 \$152 \$200 \$105 \$44 \$2,312 \$13 3-179 ------- Technologies Considered in the Agencies' Analysis Total 10% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 20% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total \$2,780 \$6 \$221 \$140 -\$60 \$121 \$196 \$105 \$38 \$1,966 \$13 \$2,747 \$6 \$221 \$140 -\$60 \$121 \$196 \$105 \$38 \$1,943 \$13 \$2,724 \$3,428 \$6 \$228 \$157 -\$65 \$152 \$201 \$105 \$43 \$2,539 \$13 \$3,379 \$6 \$228 \$157 -\$65 \$152 \$201 \$105 \$43 \$2,500 \$13 \$3,341 \$4,867 \$6 \$231 \$168 -\$82 \$162 \$204 \$105 \$45 \$3,927 \$13 \$4,780 \$6 \$231 \$168 -\$82 \$162 \$204 \$105 \$45 \$3,861 \$13 \$4,714 \$3,134 \$6 \$225 \$164 -\$86 \$152 \$200 \$105 \$44 \$2,271 \$13 \$3,093 \$6 \$225 \$164 -\$86 \$152 \$200 \$105 \$44 \$2,235 \$13 \$3,057 a The agencies have not estimated PHEV or EV costs for the large MPV and track vehicle classes since we do not believe these vehicle classes would use the technologies. Table 3-94 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for PHEV20 for the 2010 Baseline (2010\$) System 0% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 2% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Small car \$6 \$223 \$140 -\$60 \$121 \$197 \$105 \$38 \$2,169 \$13 \$2,951 \$6 \$223 \$140 -\$60 \$121 \$197 \$105 \$38 \$2,141 Standard car \$6 \$229 \$157 -\$65 \$152 \$202 \$105 \$43 \$2,870 \$13 \$3,712 \$6 \$229 \$157 -\$65 \$152 \$202 \$105 \$43 \$2,828 Large car \$6 \$232 \$168 -\$82 \$177 \$205 \$105 \$45 \$4,476 \$13 \$5,347 \$6 \$232 \$168 -\$82 \$177 \$205 \$105 \$45 \$4,402 Small MPV \$6 \$225 \$164 -\$86 \$162 \$201 \$105 \$44 \$2,586 \$13 \$3,419 \$6 \$225 \$164 -\$86 \$162 \$201 \$105 \$44 \$2,549 5-180 ------- Technologies Considered in the Agencies' Analysis Battery discharge system Total 7.5% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 10% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 20% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total \$13 \$2,924 \$6 \$223 \$140 -\$60 \$121 \$197 \$105 \$38 \$2,064 \$13 \$2,847 \$6 \$223 \$140 -\$60 \$121 \$197 \$105 \$38 \$2,029 \$13 \$2,812 \$6 \$223 \$140 -\$60 \$121 \$197 \$105 \$38 \$2,002 \$13 \$2,785 \$13 \$3,670 \$6 \$229 \$157 -\$65 \$152 \$202 \$105 \$43 \$2,712 \$13 \$3,554 \$6 \$229 \$157 -\$65 \$152 \$202 \$105 \$43 \$2,660 \$13 \$3,502 \$6 \$229 \$157 -\$65 \$152 \$202 \$105 \$43 \$2,616 \$13 \$3,458 \$13 \$5,272 \$6 \$232 \$168 -\$82 \$177 \$205 \$105 \$45 \$4,198 \$13 \$5,069 \$6 \$232 \$168 -\$82 \$177 \$205 \$105 \$45 \$4,106 \$13 \$4,976 \$6 \$232 \$168 -\$82 \$177 \$205 \$105 \$45 \$4,031 \$13 \$4,901 \$13 \$3,383 \$6 \$225 \$164 -\$86 \$162 \$201 \$105 \$44 \$2,450 \$13 \$3,283 \$6 \$225 \$164 -\$86 \$162 \$201 \$105 \$44 \$2,404 \$13 \$o T5 o J,ZJO \$6 \$225 \$164 -\$86 \$162 \$201 \$105 \$44 \$2,364 \$13 \$3,197 a The agencies have not estimated PHEV or EV costs for the large MPV and track vehicle classes since we do not believe these vehicle classes would use the technologies. Table 3-95 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for PHEV40 for the 2008 Baseline (2010\$) System 0% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Small car \$6 \$221 \$140 -\$60 \$121 \$196 \$105 Standard car \$6 \$228 \$157 -\$65 \$152 \$201 \$105 Large car \$6 \$231 \$168 -\$82 \$162 \$204 \$105 Small MPV \$6 \$225 \$164 -\$86 \$152 \$200 \$105 5-181 ------- Technologies Considered in the Agencies' Analysis Supplemental heater Motor assembly Battery discharge system Total 2% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 7.5% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 10% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 20% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total \$38 \$2,097 \$13 \$2,878 \$6 \$221 \$140 -\$60 \$121 \$196 \$105 \$38 \$2,071 \$13 \$2,852 \$6 \$221 \$140 -\$60 \$121 \$196 \$105 \$38 \$2,007 \$13 \$2,788 \$6 \$221 \$140 -\$60 \$121 \$196 \$105 \$38 \$2,007 \$13 \$2,788 \$6 \$221 \$140 -\$60 \$121 \$196 \$105 \$38 \$2,007 \$13 \$2,788 \$43 \$2,735 \$13 \$3,575 \$6 \$228 \$157 -\$65 \$152 \$201 \$105 \$43 \$2,695 \$13 \$3,536 \$6 \$228 \$157 -\$65 \$152 \$201 \$105 \$43 \$2,591 \$13 \$3,432 \$6 \$228 \$157 -\$65 \$152 \$201 \$105 \$43 \$2,591 \$13 \$3,432 \$6 \$228 \$157 -\$65 \$152 \$201 \$105 \$43 \$2,591 \$13 \$3,432 \$45 \$4,276 \$13 \$5,129 \$6 \$231 \$168 -\$82 \$162 \$204 \$105 \$45 \$4,207 \$13 \$5,059 \$6 \$231 \$168 -\$82 \$162 \$204 \$105 \$45 \$4,025 \$13 \$4,878 \$6 \$231 \$168 -\$82 \$162 \$204 \$105 \$45 \$4,025 \$13 \$4,878 \$6 \$231 \$168 -\$82 \$162 \$204 \$105 \$45 \$4,025 \$13 \$4,878 \$44 \$2,436 \$13 \$3,258 \$6 \$225 \$164 -\$86 \$152 \$200 \$105 \$44 \$2,403 \$13 \$3,225 \$6 \$225 \$164 -\$86 \$152 \$200 \$105 \$44 \$2,313 \$13 \$3,135 \$6 \$225 \$164 -\$86 \$152 \$200 \$105 \$44 \$2,312 \$13 \$3,134 \$6 \$225 \$164 -\$86 \$152 \$200 \$105 \$44 \$2,312 \$13 \$3,134 a The agencies have not estimated PHEV or EV costs for the large MPV and track vehicle classes since we do not believe these vehicle classes would use the technologies. 5-182 ------- Technologies Considered in the Agencies' Analysis Table 3-96 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for PHEV40 for the 2010 Baseline (2010\$)a System 0% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 2% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 7.5% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 10% WR Body system Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total 20% WR Body system Small car \$6 \$223 \$140 -\$60 \$121 \$197 \$105 \$38 \$2,169 \$13 \$2,951 \$6 \$223 \$140 -\$60 \$121 \$197 \$105 \$38 \$2,141 \$13 \$2,924 \$6 \$223 \$140 -\$60 \$121 \$197 \$105 \$38 \$2,068 \$13 \$2,851 \$6 \$223 \$140 -\$60 \$121 \$197 \$105 \$38 \$2,068 \$13 \$2,851 \$6 Standard car \$6 \$229 \$157 -\$65 \$152 \$202 \$105 \$43 \$2,870 \$13 \$3,712 \$6 \$229 \$157 -\$65 \$152 \$202 \$105 \$43 \$2,828 \$13 \$3,670 \$6 \$229 \$157 -\$65 \$152 \$202 \$105 \$43 \$2,714 \$13 \$3,556 \$6 \$229 \$157 -\$65 \$152 \$202 \$105 \$43 \$2,714 \$13 \$3,556 \$6 Large car \$6 \$232 \$168 -\$82 \$177 \$205 \$105 \$45 \$4,476 \$13 \$5,347 \$6 \$232 \$168 -\$82 \$177 \$205 \$105 \$45 \$4,402 \$13 \$5,272 \$6 \$232 \$168 -\$82 \$177 \$205 \$105 \$45 \$4,206 \$13 \$5,076 \$6 \$232 \$168 -\$82 \$177 \$205 \$105 \$45 \$4,206 \$13 \$5,076 \$6 Small MPV \$6 \$225 \$164 -\$86 \$162 \$201 \$105 \$44 \$2,586 \$13 \$3,419 \$6 \$225 \$164 -\$86 \$162 \$201 \$105 \$44 \$2,549 \$13 \$3,383 \$6 \$225 \$164 -\$86 \$162 \$201 \$105 \$44 \$2,450 \$13 \$3,283 \$6 \$225 \$164 -\$86 \$162 \$201 \$105 \$44 \$2,449 \$13 \$3,283 \$6 5-183 ------- Technologies Considered in the Agencies' Analysis Brake system Climate controls Delete electrical DC-DC converter Power Distr & control On vehicle charger Supplemental heater Motor assembly Battery discharge system Total \$223 \$140 -\$60 \$121 \$197 \$105 \$38 \$2,068 \$13 \$2,851 \$229 \$157 -\$65 \$152 \$202 \$105 \$43 \$2,714 \$13 \$3,556 \$232 \$168 -\$82 \$177 \$205 \$105 \$45 \$4,206 \$13 \$5,076 \$225 \$164 -\$86 \$162 \$201 \$105 \$44 \$2,449 \$13 \$3,283 a The agencies have not estimated PHEV or EV costs for the large MPV and track vehicle classes since we do not believe these vehicle classes would use the technologies. Table 3-97 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for EV75 for the 2008 Baseline (2010\$) a System 0% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 2% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 7.5% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater Small car \$221 \$140 -\$60 \$121 \$196 \$76 \$316 \$703 \$121 -\$1,596 -\$894 \$992 \$13 \$350 \$221 \$140 -\$60 \$121 \$196 \$76 \$316 \$689 \$121 -\$1,596 -\$894 \$976 \$13 \$320 \$221 \$140 -\$60 \$121 \$196 \$76 Standard car \$228 \$157 -\$65 \$152 \$201 \$85 \$316 \$1,044 \$121 -\$1,596 -\$894 \$1,383 \$13 \$1,145 \$228 \$157 -\$65 \$152 \$201 \$85 \$316 \$1,023 \$121 -\$1,596 -\$894 \$1,359 \$13 \$1,100 \$228 \$157 -\$65 \$152 \$201 \$85 Large car \$231 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,868 \$121 -\$2,466 -\$894 \$2,329 \$13 \$2,060 \$231 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,831 \$121 -\$2,466 -\$894 \$2,286 \$13 \$1,979 \$231 \$168 -\$82 \$162 \$204 \$91 Small MPV \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$885 \$121 -\$2,394 -\$894 \$1,200 \$13 -\$12 \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$867 \$121 -\$2,394 -\$894 \$1,180 \$13 -\$50 \$225 \$164 -\$86 \$152 \$200 \$89 5-184 ------- Technologies Considered in the Agencies' Analysis On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 10% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 20% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total \$316 \$650 \$121 -\$1,596 -\$894 \$932 \$13 \$237 \$221 \$140 -\$60 \$121 \$196 \$76 \$316 \$633 \$121 -\$1,596 -\$894 \$911 \$13 \$199 \$221 \$140 -\$60 \$121 \$196 \$76 \$316 \$571 \$121 -\$1,596 -\$894 \$840 \$13 \$65 \$316 \$966 \$121 -\$1,596 -\$894 \$1,293 \$13 \$977 \$228 \$157 -\$65 \$152 \$201 \$85 \$316 \$939 \$121 -\$1,596 -\$894 \$1,263 \$13 \$921 \$228 \$157 -\$65 \$152 \$201 \$85 \$316 \$851 \$121 -\$1,596 -\$894 \$1,162 \$13 \$731 \$316 \$1,728 \$121 -\$2,466 -\$894 \$2,168 \$13 \$1,759 \$231 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,681 \$121 -\$2,466 -\$894 \$2,114 \$13 \$1,659 \$231 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,519 \$121 -\$2,466 -\$894 \$1,928 \$13 \$1,309 \$316 \$818 \$121 -\$2,394 -\$894 \$1,124 \$13 -\$154 \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$796 \$121 -\$2,394 -\$894 \$1,099 \$13 -\$202 \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$727 \$121 -\$2,394 -\$894 \$1,020 \$13 -\$350 a The agencies have not estimated PHEV or EV costs for the large MPV and truck vehicle classes since we do not believe these vehicle classes would use the technologies. Table 3-98 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for EV75 for the 2010 Baseline (2010\$) a System 0% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Small car \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$729 Standard car \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$1,094 Large car \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,932 Small MPV \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$946 3-185 ------- Technologies Considered in the Agencies' Analysis Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 2% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 7.5% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 10% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 20% WR Brake system \$121 -\$1,596 -\$894 \$1,021 \$13 \$406 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$714 \$121 -\$1,596 -\$894 \$1,004 \$13 \$375 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$674 \$121 -\$1,596 -\$894 \$958 \$13 \$289 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$656 \$121 -\$1,596 -\$894 \$938 \$13 \$250 \$223 \$121 -\$1,596 -\$894 \$1,441 \$13 \$1,255 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$1,072 \$121 -\$1,596 -\$894 \$1,416 \$13 \$1,208 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$1,012 \$121 -\$1,596 -\$894 \$1,347 \$13 \$1,079 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$985 \$121 -\$1,596 -\$894 \$1,315 \$13 \$1,020 \$229 \$121 -\$2,466 -\$894 \$2,402 \$13 \$2,214 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,893 \$121 -\$2,466 -\$894 \$2,358 \$13 \$2,131 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,787 \$121 -\$2,466 -\$894 \$2,236 \$13 \$1,903 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,739 \$121 -\$2,466 -\$894 \$2,180 \$13 \$1,799 \$232 \$121 -\$2,394 -\$894 \$1,271 \$13 \$132 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$927 \$121 -\$2,394 -\$894 \$1,249 \$13 \$92 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$875 \$121 -\$2,394 -\$894 \$1,189 \$13 -\$20 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$851 \$121 -\$2,394 -\$894 \$1,162 \$13 -\$71 \$225 3-186 ------- Technologies Considered in the Agencies' Analysis Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total \$140 -\$60 \$121 \$197 \$76 \$316 \$595 \$121 -\$1,596 -\$894 \$867 \$13 \$118 \$157 -\$65 \$152 \$202 \$85 \$316 \$895 \$121 -\$1,596 -\$894 \$1,212 \$13 \$828 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,580 \$121 -\$2,466 -\$894 \$1,998 \$13 \$1,458 \$164 -\$86 \$162 \$201 \$89 \$316 \$780 \$121 -\$2,394 -\$894 \$1,080 \$13 -\$225 a The agencies have not estimated PHEV or EV costs for the large MPV and track vehicle classes since we do not believe these vehicle classes would use the technologies. Table 3-99 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for EV100 for the 2008 Baseline (2010\$)a System 0% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 2% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 7.5% WR Brake system Climate controls Delete electrical Small car \$221 \$140 -\$60 \$121 \$196 \$76 \$316 \$703 \$121 -\$1,596 -\$894 \$992 \$13 \$350 \$221 \$140 -\$60 \$121 \$196 \$76 \$316 \$689 \$121 -\$1,596 -\$894 \$976 \$13 \$320 \$221 \$140 -\$60 Standard car \$228 \$157 -\$65 \$152 \$201 \$85 \$316 \$1,044 \$121 -\$1,596 -\$894 \$1,383 \$13 \$1,145 \$228 \$157 -\$65 \$152 \$201 \$85 \$316 \$1,023 \$121 -\$1,596 -\$894 \$1,359 \$13 \$1,100 \$228 \$157 -\$65 Large car \$231 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,868 \$121 -\$2,466 -\$894 \$2,329 \$13 \$2,060 \$231 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,831 \$121 -\$2,466 -\$894 \$2,286 \$13 \$1,979 \$231 \$168 -\$82 Small MPV \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$885 \$121 -\$2,394 -\$894 \$1,200 \$13 -\$12 \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$867 \$121 -\$2,394 -\$894 \$1,180 \$13 -\$50 \$225 \$164 -\$86 5-187 ------- Technologies Considered in the Agencies' Analysis DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 10% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 20% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total \$121 \$196 \$76 \$316 \$650 \$121 -\$1,596 -\$894 \$932 \$13 \$237 \$221 \$140 -\$60 \$121 \$196 \$76 \$316 \$633 \$121 -\$1,596 -\$894 \$911 \$13 \$199 \$221 \$140 -\$60 \$121 \$196 \$76 \$316 \$608 \$121 -\$1,596 -\$894 \$883 \$13 \$146 \$152 \$201 \$85 \$316 \$966 \$121 -\$1,596 -\$894 \$1,293 \$13 \$977 \$228 \$157 -\$65 \$152 \$201 \$85 \$316 \$939 \$121 -\$1,596 -\$894 \$1,263 \$13 \$921 \$228 \$157 -\$65 \$152 \$201 \$85 \$316 \$906 \$121 -\$1,596 -\$894 \$1,224 \$13 \$848 \$162 \$204 \$91 \$316 \$1,728 \$121 -\$2,466 -\$894 \$2,168 \$13 \$1,759 \$231 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,681 \$121 -\$2,466 -\$894 \$2,114 \$13 \$1,659 \$231 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,617 \$121 -\$2,466 -\$894 \$2,041 \$13 \$1,521 \$152 \$200 \$89 \$316 \$818 \$121 -\$2,394 -\$894 \$1,124 \$13 -\$154 \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$796 \$121 -\$2,394 -\$894 \$1,099 \$13 -\$202 \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$774 \$121 -\$2,394 -\$894 \$1,073 \$13 -\$249 a The agencies have not estimated PHEV or EV costs for the large MPV and truck vehicle classes since we do not believe these vehicle classes would use the technologies. Table 3-100 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for EV100 for the 2010 Baseline (2010\$)a System 0% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Small car \$223 \$140 -\$60 \$121 \$197 Standard car \$229 \$157 -\$65 \$152 \$202 Large car \$232 \$168 -\$82 \$177 \$205 Small MPV \$225 \$164 -\$86 \$162 \$201 5-188 ------- Technologies Considered in the Agencies' Analysis Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 2% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 7.5% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 10% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system \$76 \$316 \$729 \$121 -\$1,596 -\$894 \$1,021 \$13 \$406 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$714 \$121 -\$1,596 -\$894 \$1,004 \$13 \$375 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$674 \$121 -\$1,596 -\$894 \$958 \$13 \$289 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$656 \$121 -\$1,596 -\$894 \$938 \$13 \$85 \$316 \$1,094 \$121 -\$1,596 -\$894 \$1,441 \$13 \$1,255 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$1,072 \$121 -\$1,596 -\$894 \$1,416 \$13 \$1,208 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$1,012 \$121 -\$1,596 -\$894 \$1,347 \$13 \$1,079 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$985 \$121 -\$1,596 -\$894 \$1,315 \$13 \$91 \$316 \$1,932 \$121 -\$2,466 -\$894 \$2,402 \$13 \$2,214 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,893 \$121 -\$2,466 -\$894 \$2,358 \$13 \$2,131 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,787 \$121 -\$2,466 -\$894 \$2,236 \$13 \$1,903 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,739 \$121 -\$2,466 -\$894 \$2,180 \$13 \$89 \$316 \$946 \$121 -\$2,394 -\$894 \$1,271 \$13 \$132 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$927 \$121 -\$2,394 -\$894 \$1,249 \$13 \$92 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$875 \$121 -\$2,394 -\$894 \$1,189 \$13 -\$20 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$851 \$121 -\$2,394 -\$894 \$1,162 \$13 3-189 ------- Technologies Considered in the Agencies' Analysis Total 20% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total \$250 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$633 \$121 -\$1,596 -\$894 \$912 \$13 \$201 \$1,020 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$954 \$121 -\$1,596 -\$894 \$1,280 \$13 \$954 \$1,799 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,684 \$121 -\$2,466 -\$894 \$2,118 \$13 \$1,682 -\$71 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$829 \$121 -\$2,394 -\$894 \$1,137 \$13 -\$118 a The agencies have not estimated PHEV or EV costs for the large MPV and track vehicle classes since we do not believe these vehicle classes would use the technologies. Table 3-101 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for EV150 for the 2008 Baseline (2010\$)a System 0% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 2% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 7.5% WR Small car \$223 \$140 -\$60 \$121 \$196 \$76 \$316 \$703 \$121 -\$1,596 -\$894 \$992 \$13 \$351 \$223 \$140 -\$60 \$121 \$196 \$76 \$316 \$692 \$121 -\$1,596 -\$894 \$979 \$13 \$328 Standard car \$229 \$157 -\$65 \$152 \$201 \$85 \$316 \$1,044 \$121 -\$1,596 -\$894 \$1,383 \$13 \$1,146 \$229 \$157 -\$65 \$152 \$201 \$85 \$316 \$1,028 \$121 -\$1,596 -\$894 \$1,364 \$13 \$1,111 Large car \$232 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,868 \$121 -\$2,466 -\$894 \$2,329 \$13 \$2,061 \$232 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,837 \$121 -\$2,466 -\$894 \$2,293 \$13 \$1,995 Small MPV \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$885 \$121 -\$2,394 -\$894 \$1,200 \$13 -\$11 \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$878 \$121 -\$2,394 -\$894 \$1,193 \$13 -\$26 5-190 ------- Technologies Considered in the Agencies' Analysis Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 10% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 20% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total \$223 \$140 -\$60 \$121 \$196 \$76 \$316 \$692 \$121 -\$1,596 -\$894 \$979 \$13 \$328 \$223 \$140 -\$60 \$121 \$196 \$76 \$316 \$692 \$121 -\$1,596 -\$894 \$979 \$13 \$328 \$223 \$140 -\$60 \$121 \$196 \$76 \$316 \$692 \$121 -\$1,596 -\$894 \$979 \$13 \$328 \$229 \$157 -\$65 \$152 \$201 \$85 \$316 \$1,028 \$121 -\$1,596 -\$894 \$1,364 \$13 \$1,111 \$229 \$157 -\$65 \$152 \$201 \$85 \$316 \$1,028 \$121 -\$1,596 -\$894 \$1,364 \$13 \$1,111 \$229 \$157 -\$65 \$152 \$201 \$85 \$316 \$1,028 \$121 -\$1,596 -\$894 \$1,364 \$13 \$1,111 \$232 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,837 \$121 -\$2,466 -\$894 \$2,293 \$13 \$1,995 \$232 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,837 \$121 -\$2,466 -\$894 \$2,293 \$13 \$1,995 \$232 \$168 -\$82 \$162 \$204 \$91 \$316 \$1,837 \$121 -\$2,466 -\$894 \$2,293 \$13 \$1,995 \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$878 \$121 -\$2,394 -\$894 \$1,193 \$13 -\$26 \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$878 \$121 -\$2,394 -\$894 \$1,193 \$13 -\$26 \$225 \$164 -\$86 \$152 \$200 \$89 \$316 \$878 \$121 -\$2,394 -\$894 \$1,193 \$13 -\$26 a The agencies have not estimated PHEV or EV costs for the large MPV and truck vehicle classes since we do not believe these vehicle classes would use the technologies. Table 3-102 Scaled Non-battery DMC by Applied Vehicle Weight Reduction for EV150 for the 2010 Baseline (2010\$)a System 0% WR Brake system Climate controls Small car \$223 \$140 Standard car \$229 \$157 Large car \$232 \$168 Small MPV \$225 \$164 5-191 ------- Technologies Considered in the Agencies' Analysis Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 2% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 7.5% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total 10% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine -\$60 \$121 \$197 \$76 \$316 \$729 \$121 -\$1,596 -\$894 \$1,021 \$13 \$406 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$720 \$121 -\$1,596 -\$894 \$1,011 \$13 \$387 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$720 \$121 -\$1,596 -\$894 \$1,011 \$13 \$387 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$720 \$121 -\$1,596 -\$65 \$152 \$202 \$85 \$316 \$1,094 \$121 -\$1,596 -\$894 \$1,441 \$13 \$1,255 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$1,081 \$121 -\$1,596 -\$894 \$1,425 \$13 \$1,226 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$1,081 \$121 -\$1,596 -\$894 \$1,425 \$13 \$1,226 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$1,081 \$121 -\$1,596 -\$82 \$177 \$205 \$91 \$316 \$1,932 \$121 -\$2,466 -\$894 \$2,402 \$13 \$2,214 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,910 \$121 -\$2,466 -\$894 \$2,377 \$13 \$2,167 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,910 \$121 -\$2,466 -\$894 \$2,377 \$13 \$2,167 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,910 \$121 -\$2,466 -\$86 \$162 \$201 \$89 \$316 \$946 \$121 -\$2,394 -\$894 \$1,271 \$13 \$132 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$941 \$121 -\$2,394 -\$894 \$1,265 \$13 \$121 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$941 \$121 -\$2,394 -\$894 \$1,265 \$13 \$121 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$941 \$121 -\$2,394 3-192 ------- Technologies Considered in the Agencies' Analysis Delete transmission Motor assembly Battery discharge system Total 20% WR Brake system Climate controls Delete electrical DC-DC converter High voltage wiring Supplemental heater On vehicle charger Motor inverter Controls Delete 1C engine Delete transmission Motor assembly Battery discharge system Total -\$894 \$1,011 \$13 \$387 \$223 \$140 -\$60 \$121 \$197 \$76 \$316 \$720 \$121 -\$1,596 -\$894 \$1,011 \$13 \$387 -\$894 \$1,425 \$13 \$1,226 \$229 \$157 -\$65 \$152 \$202 \$85 \$316 \$1,081 \$121 -\$1,596 -\$894 \$1,425 \$13 \$1,226 -\$894 \$2,377 \$13 \$2,167 \$232 \$168 -\$82 \$177 \$205 \$91 \$316 \$1,910 \$121 -\$2,466 -\$894 \$2,377 \$13 \$2,167 -\$894 \$1,265 \$13 \$121 \$225 \$164 -\$86 \$162 \$201 \$89 \$316 \$941 \$121 -\$2,394 -\$894 \$1,265 \$13 \$121 a The agencies have not estimated PHEV classes would use the technologies. or EV costs for the large MPV and truck vehicle classes since we do not believe these vehicle Similar to the approach taken for battery pack costs, the agencies generated linear regressions of non-battery system costs against percent of net mass reduction and the results are shown in Table 3-103. This was done using the same weight reduction offsets as used for battery packs as presented in Table 3-75. The agencies separated battery pack costs from the remainder of the systems for each type of electrified vehicle. The advantage of separating the battery pack costs from other system costs is that it allows each to carry unique indirect cost multipliers and learning effects which are important given that battery technology is an emerging technology, while electric motors and inverters are more stable technologies. Table 3-103 Linear Regressions of Non-Battery System Direct Manufacturing Costs vs Net Mass reduction (2010\$) Vehicle Class P2HEV PHEV20 PHEV40 EV75 EV100 EV150 2008 Baseline Small car Standard car Large car Small MPV Large MPV Truck -\$263x+\$ 1,675 -\$391x+\$l,857 -\$699x+\$2,175 -\$331x+\$l,777 -\$506x+\$2,052 -\$648x+\$2,169 -\$l,316x+\$2,878 -\$l,953x+\$3,575 -\$3,495x+\$5,129 -\$l,655x+\$3,258 -\$l,316x+\$2,878 -\$l,953x+\$3,575 -\$3,495x+\$5,129 -\$l,655x+\$3,258 -\$l,510x+\$350 -\$2,242x+\$l,145 -\$4,012x+\$2,060 -\$l,900x+-\$12 -\$l,510x+\$350 -\$2,242x+\$l,145 -\$4,012x+\$2,060 -\$l,900x+-\$12 -\$l,510x+\$351 -\$2,242x+\$l,146 -\$4,012x+\$2,061 -\$l,900x+-\$ll 20 10 Baseline Small car Standard car Large car Small MPV Large MPV Truck -\$279x+\$ 1,683 -\$420x+\$l,878 -\$741x+\$2,224 -\$363x+\$l,811 -\$528x+\$2,073 -\$674x+\$2,188 -\$l,397x+\$2,951 -\$2,099x+\$3,712 -\$3,705x+\$5,347 -\$l,814x+\$3,419 -\$l,397x+\$2,951 -\$2,099x+\$3,712 -\$3,705x+\$5,347 -\$l,814x+\$3,419 -\$l,565x+\$406 -\$2,350x+\$l,255 -\$4,149x+\$2,214 -\$2,032x+\$132 -\$l,565x+\$406 -\$2,350x+\$l,255 -\$4,149x+\$2,214 -\$2,032x+\$132 -\$l,565x+\$406 -\$2,350x+\$l,255 -\$4,149x+\$2,214 -\$2,032x+\$132 Notes: 3-193 ------- Technologies Considered in the Agencies' Analysis "x" in the equations represents the net weight reduction as a percentage, so the non-battery components for a small car P2 HEV (2008 baseline) with a 20% applied weight reduction and, therefore, a 15% net weight reduction would cost (-\$263)x(15%)+\$l,675=\$l,635.The agencies did not regress PHEV or EV costs for the large MPV and truck vehicle classes since we do not believe these vehicle classes would use the technologies. For P2 HEV and PHEV non-battery components, the direct manufacturing costs shown in Table 3-103 are considered applicable to the 2012MY. The agencies consider the P2 and PHEV non-battery component technologies to be on the flat portion of the learning curve during the 2017-2025 timeframe. The agencies have applied a highl complexity ICM of 1.56 through 2018 then 1.35 thereafter. For EV non-battery components, the direct manufacturing costs shown in Table 3-103 are considered applicable to the 2017MY. The agencies consider the EV non-battery component technologies to be on the flat portion of the learning curve during the 2017-2025 timeframe. The agencies have applied a high2 complexity ICM of 1.77 through 2024 then 1.50 thereafter. The resultant costs for P2 HEV, PHEV20, PHEV40, EV75, EV100 and EV150 non-battery components for the 2008 and 2010 baselines are shown in Table 3-104 through Table 3-115, respectively.aaa Table 3-104 Costs for P2 HEV Non-Battery Components for the 2008 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Large MPV Large MPV Large MPV Truck Truck Truck Small car Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% Net WR 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 6% 11% 16% 5% 2017 \$1,442 \$1,430 \$1,419 \$1,594 \$1,577 \$1,560 \$1,857 \$1,826 \$1,796 \$1,528 \$1,513 \$1,499 \$1,759 \$1,737 \$1,715 \$1,848 \$1,820 \$1,792 \$922 2018 \$1,413 \$1,402 \$1,391 \$1,562 \$1,546 \$1,529 \$1,820 \$1,790 \$1,760 \$1,497 \$1,483 \$1,469 \$1,723 \$1,702 \$1,680 \$1,811 \$1,784 \$1,756 \$920 2019 \$1,385 \$1,374 \$1,363 \$1,531 \$1,515 \$1,498 \$1,783 \$1,754 \$1,725 \$1,467 \$1,453 \$1,440 \$1,689 \$1,668 \$1,647 \$1,775 \$1,748 \$1,721 \$565 2020 \$1,357 \$1,346 \$1,335 \$1,500 \$1,484 \$1,468 \$1,747 \$1,719 \$1,690 \$1,438 \$1,424 \$1,411 \$1,655 \$1,634 \$1,614 \$1,739 \$1,713 \$1,686 \$564 2021 \$1,330 \$1,319 \$1,309 \$1,470 \$1,455 \$1,439 \$1,713 \$1,685 \$1,657 \$1,409 \$1,396 \$1,383 \$1,622 \$1,602 \$1,582 \$1,705 \$1,679 \$1,653 \$563 2022 \$1,303 \$1,293 \$1,283 \$1,441 \$1,426 \$1,410 \$1,678 \$1,651 \$1,623 \$1,381 \$1,368 \$1,355 \$1,590 \$1,570 \$1,550 \$1,670 \$1,645 \$1,620 \$563 2023 \$1,277 \$1,267 \$1,257 \$1,412 \$1,397 \$1,382 \$1,645 \$1,618 \$1,591 \$1,353 \$1,340 \$1,328 \$1,558 \$1,538 \$1,519 \$1,637 \$1,612 \$1,587 \$562 2024 \$1,252 \$1,242 \$1,232 \$1,384 \$1,369 \$1,354 \$1,612 \$1,585 \$1,559 \$1,326 \$1,314 \$1,301 \$1,527 \$1,508 \$1,489 \$1,604 \$1,580 \$1,556 \$561 2025 \$1,227 \$1,217 \$1,207 \$1,356 \$1,342 \$1,327 \$1,580 \$1,554 \$1,528 \$1,300 \$1,287 \$1,275 \$1,496 \$1,477 \$1,459 \$1,572 \$1,548 \$1,524 \$560 aaa Note that, in the draft Joint TSD, we inadvertently stated the following with respect to the years in which costs were considered valid and the years for which near term and long term ICMs were applied: "For P2 HEV non- battery components, the direct manufacturing costs shown in Table 3-103 are considered applicable to the 2017MY. The agencies consider the P2 non-battery component technologies to be on the flat portion of the learning curve during the 2017-2025 timeframe. The agencies have applied a highl complexity ICM of 1.56 through 2018 then 1.35 thereafter. For PHEV and EV non-battery components, the direct manufacturing costs shown in Table 3-103 are considered applicable to the 2025MY. The agencies consider the PHEV and EV non- battery component technologies to be on the flat portion of the learning curve during the 2017-2025 timeframe. The agencies have applied a high2 complexity ICM of 1.77 through 2024 then 1.50 thereafter." Importantly, the costs then (and now) were calculated according to the corrected text shown in this final Joint TSD. 3-194 ------- Technologies Considered in the Agencies' Analysis 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC TC TC TC TC TC TC Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Large MPV Large MPV Large MPV Truck Truck Truck Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Large MPV Large MPV Large MPV Truck Truck Truck 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 6% 11% 16% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 6% 11% 16% \$915 \$908 \$1,020 \$1,009 \$998 \$1,188 \$1,168 \$1,149 \$977 \$968 \$959 \$1,125 \$1,111 \$1,097 \$1,182 \$1,164 \$1,146 \$2,364 \$2,345 \$2,327 \$2,614 \$2,586 \$2,558 \$3,044 \$2,995 \$2,945 \$2,505 \$2,481 \$2,458 \$2,884 \$2,848 \$2,812 \$3,030 \$2,984 \$2,938 \$913 \$906 \$1,018 \$1,007 \$996 \$1,185 \$1,166 \$1,147 \$975 \$966 \$957 \$1,123 \$1,109 \$1,095 \$1,180 \$1,162 \$1,144 \$2,333 \$2,315 \$2,296 \$2,580 \$2,552 \$2,525 \$3,005 \$2,956 \$2,907 \$2,472 \$2,449 \$2,426 \$2,846 \$2,811 \$2,775 \$2,991 \$2,946 \$2,900 \$561 \$556 \$625 \$618 \$612 \$728 \$716 \$704 \$599 \$593 \$588 \$689 \$681 \$672 \$724 \$713 \$702 \$1,950 \$1,934 \$1,919 \$2,156 \$2,133 \$2,110 \$2,511 \$2,470 \$2,429 \$2,066 \$2,046 \$2,027 \$2,378 \$2,349 \$2,319 \$2,499 \$2,461 \$2,423 \$560 \$555 \$624 \$617 \$611 \$727 \$715 \$703 \$598 \$592 \$587 \$688 \$680 \$671 \$723 \$712 \$701 \$1,921 \$1,906 \$1,891 \$2,124 \$2,102 \$2,079 \$2,474 \$2,434 \$2,393 \$2,036 \$2,016 \$1,997 \$2,343 \$2,314 \$2,285 \$2,463 \$2,425 \$2,388 \$559 \$554 \$623 \$616 \$610 \$726 \$714 \$702 \$597 \$591 \$586 \$687 \$679 \$670 \$722 \$711 \$700 \$1,893 \$1,878 \$1,863 \$2,093 \$2,071 \$2,049 \$2,438 \$2,398 \$2,358 \$2,006 \$1,987 \$1,968 \$2,309 \$2,280 \$2,252 \$2,427 \$2,390 \$2,353 \$558 \$554 \$622 \$615 \$609 \$724 \$713 \$701 \$596 \$590 \$585 \$686 \$678 \$669 \$721 \$710 \$699 \$1,866 \$1,851 \$1,836 \$2,063 \$2,041 \$2,019 \$2,403 \$2,364 \$2,324 \$1,977 \$1,958 \$1,940 \$2,276 \$2,247 \$2,219 \$2,392 \$2,355 \$2,319 \$557 \$553 \$621 \$614 \$608 \$723 \$712 \$700 \$595 \$590 \$584 \$685 \$677 \$668 \$720 \$709 \$698 \$1,839 \$1,824 \$1,810 \$2,033 \$2,012 \$1,990 \$2,368 \$2,329 \$2,291 \$1,948 \$1,930 \$1,912 \$2,243 \$2,215 \$2,187 \$2,357 \$2,321 \$2,285 \$556 \$552 \$620 \$614 \$607 \$722 \$711 \$699 \$594 \$589 \$583 \$684 \$676 \$667 \$719 \$708 \$697 \$1,813 \$1,798 \$1,784 \$2,004 \$1,983 \$1,961 \$2,334 \$2,296 \$2,258 \$1,920 \$1,902 \$1,884 \$2,211 \$2,183 \$2,156 \$2,323 \$2,288 \$2,253 \$556 \$551 \$619 \$613 \$606 \$721 \$710 \$698 \$593 \$588 \$582 \$683 \$675 \$666 \$718 \$707 \$696 \$1,787 \$1,773 \$1,758 \$1,975 \$1,954 \$1,933 \$2,301 \$2,263 \$2,226 \$1,893 \$1,875 \$1,857 \$2,179 \$2,152 \$2,125 \$2,290 \$2,255 \$2,221 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-105 Costs for P2 HEV Non-Battery Components for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Large MPV Large MPV Large MPV Truck Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% Net WR 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 6% 2017 \$1,448 \$1,436 \$1,423 \$1,611 \$1,593 \$1,574 \$1,898 \$1,866 \$1,833 \$1,555 \$1,540 \$1,524 \$1,776 \$1,753 \$1,730 \$1,863 2018 \$1,419 \$1,407 \$1,395 \$1,579 \$1,561 \$1,543 \$1,860 \$1,828 \$1,797 \$1,524 \$1,509 \$1,493 \$1,740 \$1,718 \$1,696 \$1,826 2019 \$1,390 \$1,379 \$1,367 \$1,547 \$1,529 \$1,512 \$1,823 \$1,792 \$1,761 \$1,494 \$1,479 \$1,464 \$1,706 \$1,684 \$1,662 \$1,790 2020 \$1,363 \$1,351 \$1,340 \$1,516 \$1,499 \$1,482 \$1,786 \$1,756 \$1,726 \$1,464 \$1,449 \$1,434 \$1,672 \$1,650 \$1,628 \$1,754 2021 \$1,335 \$1,324 \$1,313 \$1,486 \$1,469 \$1,452 \$1,750 \$1,721 \$1,691 \$1,435 \$1,420 \$1,406 \$1,638 \$1,617 \$1,596 \$1,719 2022 \$1,309 \$1,298 \$1,287 \$1,456 \$1,440 \$1,423 \$1,715 \$1,686 \$1,657 \$1,406 \$1,392 \$1,377 \$1,605 \$1,585 \$1,564 \$1,684 2023 \$1,282 \$1,272 \$1,261 \$1,427 \$1,411 \$1,395 \$1,681 \$1,653 \$1,624 \$1,378 \$1,364 \$1,350 \$1,573 \$1,553 \$1,533 \$1,651 2024 \$1,257 \$1,246 \$1,236 \$1,398 \$1,383 \$1,367 \$1,647 \$1,620 \$1,592 \$1,350 \$1,337 \$1,323 \$1,542 \$1,522 \$1,502 \$1,618 2025 \$1,232 \$1,221 \$1,211 \$1,370 \$1,355 \$1,339 \$1,614 \$1,587 \$1,560 \$1,323 \$1,310 \$1,296 \$1,511 \$1,491 \$1,472 \$1,585 3-195 ------- Technologies Considered in the Agencies' Analysis DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC TC TC TC TC TC TC Truck Truck Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Large MPV Large MPV Large MPV Truck Truck Truck Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Large MPV Large MPV Large MPV Truck Truck Truck 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 11% 16% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 6% 11% 16% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 5% 10% 15% 6% 11% 16% \$1,834 \$1,805 \$926 \$918 \$911 \$1,030 \$1,019 \$1,007 \$1,214 \$1,193 \$1,173 \$995 \$985 \$975 \$1,136 \$1,122 \$1,107 \$1,192 \$1,173 \$1,155 \$2,374 \$2,354 \$2,334 \$2,641 \$2,611 \$2,582 \$3,112 \$3,059 \$3,006 \$2,550 \$2,525 \$2,499 \$2,912 \$2,875 \$2,837 \$3,056 \$3,008 \$2,960 \$1,798 \$1,769 \$924 \$917 \$909 \$1,028 \$1,017 \$1,005 \$1,212 \$1,191 \$1,171 \$993 \$983 \$973 \$1,134 \$1,119 \$1,105 \$1,190 \$1,171 \$1,152 \$2,343 \$2,323 \$2,304 \$2,607 \$2,577 \$2,548 \$3,071 \$3,019 \$2,967 \$2,517 \$2,492 \$2,466 \$2,874 \$2,837 \$2,800 \$3,016 \$2,969 \$2,921 \$1,762 \$1,733 \$568 \$563 \$558 \$631 \$624 \$617 \$744 \$731 \$719 \$610 \$604 \$597 \$696 \$687 \$678 \$730 \$719 \$708 \$1,958 \$1,942 \$1,925 \$2,178 \$2,154 \$2,129 \$2,566 \$2,523 \$2,479 \$2,103 \$2,082 \$2,061 \$2,402 \$2,371 \$2,340 \$2,520 \$2,481 \$2,441 \$1,726 \$1,699 \$567 \$562 \$557 \$630 \$623 \$616 \$743 \$730 \$718 \$609 \$603 \$596 \$695 \$686 \$677 \$729 \$718 \$706 \$1,929 \$1,913 \$1,897 \$2,146 \$2,122 \$2,098 \$2,529 \$2,486 \$2,443 \$2,073 \$2,052 \$2,031 \$2,367 \$2,336 \$2,306 \$2,483 \$2,444 \$2,405 \$1,692 \$1,665 \$566 \$561 \$556 \$629 \$622 \$615 \$742 \$729 \$716 \$608 \$602 \$596 \$694 \$685 \$676 \$728 \$717 \$705 \$1,901 \$1,885 \$1,869 \$2,115 \$2,091 \$2,067 \$2,492 \$2,450 \$2,408 \$2,042 \$2,022 \$2,001 \$2,332 \$2,302 \$2,272 \$2,447 \$2,409 \$2,370 \$1,658 \$1,632 \$565 \$560 \$555 \$629 \$621 \$614 \$740 \$728 \$715 \$607 \$601 \$595 \$693 \$684 \$675 \$727 \$716 \$704 \$1,874 \$1,858 \$1,842 \$2,085 \$2,061 \$2,037 \$2,456 \$2,414 \$2,373 \$2,013 \$1,992 \$1,972 \$2,298 \$2,269 \$2,239 \$2,412 \$2,374 \$2,336 \$1,625 \$1,599 \$564 \$559 \$555 \$628 \$620 \$613 \$739 \$727 \$714 \$606 \$600 \$594 \$692 \$683 \$674 \$726 \$715 \$703 \$1,847 \$1,831 \$1,816 \$2,054 \$2,031 \$2,008 \$2,420 \$2,379 \$2,338 \$1,984 \$1,964 \$1,944 \$2,265 \$2,236 \$2,207 \$2,377 \$2,339 \$2,302 \$1,592 \$1,567 \$563 \$559 \$554 \$627 \$620 \$613 \$738 \$726 \$713 \$605 \$599 \$593 \$691 \$682 \$673 \$725 \$714 \$702 \$1,820 \$1,805 \$1,790 \$2,025 \$2,002 \$1,979 \$2,386 \$2,345 \$2,305 \$1,955 \$1,936 \$1,916 \$2,233 \$2,204 \$2,175 \$2,343 \$2,306 \$2,269 \$1,560 \$1,536 \$562 \$558 \$553 \$626 \$619 \$612 \$737 \$725 \$712 \$604 \$598 \$592 \$690 \$681 \$672 \$724 \$713 \$701 \$1,794 \$1,779 \$1,764 \$1,996 \$1,974 \$1,951 \$2,352 \$2,312 \$2,272 \$1,927 \$1,908 \$1,888 \$2,201 \$2,173 \$2,144 \$2,309 \$2,273 \$2,237 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-106 Costs for PHEV20 Non-Battery Components for the 2008 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% NetWR 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 2017 \$2,463 \$2,406 \$2,349 \$3,050 \$2,966 \$2,881 \$4,389 \$4,238 \$4,086 \$2,784 2018 \$2,414 \$2,358 \$2,302 \$2,989 \$2,906 \$2,823 \$4,301 \$4,153 \$4,004 \$2,728 2019 \$2,365 \$2,311 \$2,256 \$2,930 \$2,848 \$2,767 \$4,215 \$4,070 \$3,924 \$2,673 2020 \$2,318 \$2,264 \$2,211 \$2,871 \$2,791 \$2,712 \$4,131 \$3,988 \$3,846 \$2,620 2021 \$2,272 \$2,219 \$2,166 \$2,814 \$2,735 \$2,657 \$4,049 \$3,909 \$3,769 \$2,568 2022 \$2,226 \$2,175 \$2,123 \$2,757 \$2,681 \$2,604 \$3,968 \$3,831 \$3,693 \$2,516 2023 \$2,182 \$2,131 \$2,081 \$2,702 \$2,627 \$2,552 \$3,888 \$3,754 \$3,620 \$2,466 2024 \$2,138 \$2,089 \$2,039 \$2,648 \$2,575 \$2,501 \$3,810 \$3,679 \$3,547 \$2,417 2025 \$2,095 \$2,047 \$1,998 \$2,595 \$2,523 \$2,451 \$3,734 \$3,605 \$3,476 \$2,368 3-196 ------- Technologies Considered in the Agencies' Analysis DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 8% 13% 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 8% 13% 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 8% 13% \$2,712 \$2,640 \$1,576 \$1,539 \$1,503 \$1,951 \$1,897 \$1,843 \$2,808 \$2,711 \$2,614 \$1,781 \$1,735 \$1,689 \$4,039 \$3,945 \$3,851 \$5,002 \$4,863 \$4,724 \$7,197 \$6,949 \$6,700 \$4,565 \$4,447 \$4,329 \$2,658 \$2,587 \$1,572 \$1,536 \$1,500 \$1,948 \$1,893 \$1,839 \$2,802 \$2,706 \$2,609 \$1,777 \$1,731 \$1,686 \$3,986 \$3,894 \$3,801 \$4,937 \$4,800 \$4,663 \$7,104 \$6,858 \$6,613 \$4,505 \$4,389 \$4,273 \$2,604 \$2,536 \$965 \$943 \$921 \$1,196 \$1,163 \$1,129 \$1,721 \$1,661 \$1,602 \$1,091 \$1,063 \$1,035 \$3,331 \$3,254 \$3,177 \$4,125 \$4,011 \$3,896 \$5,936 \$5,731 \$5,526 \$3,765 \$3,668 \$3,570 \$2,552 \$2,485 \$964 \$942 \$919 \$1,194 \$1,161 \$1,128 \$1,718 \$1,659 \$1,599 \$1,089 \$1,061 \$1,033 \$3,282 \$3,206 \$3,130 \$4,065 \$3,952 \$3,839 \$5,849 \$5,647 \$5,445 \$3,709 \$3,614 \$3,518 \$2,501 \$2,435 \$963 \$940 \$918 \$1,192 \$1,159 \$1,126 \$1,715 \$1,656 \$1,597 \$1,088 \$1,060 \$1,032 \$3,234 \$3,159 \$3,084 \$4,006 \$3,894 \$3,783 \$5,764 \$5,565 \$5,366 \$3,655 \$3,561 \$3,467 \$2,451 \$2,386 \$961 \$939 \$917 \$1,190 \$1,157 \$1,124 \$1,713 \$1,654 \$1,594 \$1,086 \$1,058 \$1,030 \$3,187 \$3,114 \$3,040 \$3,948 \$3,838 \$3,728 \$5,680 \$5,484 \$5,288 \$3,602 \$3,510 \$3,417 \$2,402 \$2,339 \$960 \$937 \$915 \$1,189 \$1,156 \$1,123 \$1,710 \$1,651 \$1,592 \$1,085 \$1,057 \$1,029 \$3,141 \$3,069 \$2,996 \$3,891 \$3,783 \$3,675 \$5,598 \$5,405 \$5,212 \$3,550 \$3,459 \$3,367 \$2,354 \$2,292 \$958 \$936 \$914 \$1,187 \$1,154 \$1,121 \$1,708 \$1,649 \$1,590 \$1,083 \$1,055 \$1,027 \$3,096 \$3,025 \$2,953 \$3,835 \$3,728 \$3,622 \$5,518 \$5,328 \$5,137 \$3,500 \$3,409 \$3,319 \$2,307 \$2,246 \$957 \$935 \$913 \$1,185 \$1,152 \$1,119 \$1,705 \$1,646 \$1,587 \$1,081 \$1,054 \$1,026 \$3,052 \$2,982 \$2,911 \$3,780 \$3,675 \$3,570 \$5,440 \$5,252 \$5,064 \$3,450 \$3,361 \$3,272 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-107 Costs for PHEV20 Non-Battery Components for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC Vehicle class Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV Small car Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% NetWR 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 8% 13% 3% 8% 13% 3% 8% 13% 2% 7% 12% 3% 8% 13% 3% 2017 \$2,524 \$2,464 \$2,403 \$3,166 \$3,075 \$2,984 \$4,574 \$4,414 \$4,253 \$2,919 \$2,841 \$2,762 \$1,615 \$1,576 \$1,537 \$2,025 \$1,967 \$1,909 \$2,926 \$2,824 \$2,721 \$1,868 \$1,817 \$1,767 \$4,139 2018 \$2,474 \$2,414 \$2,355 \$3,102 \$3,013 \$2,924 \$4,483 \$4,325 \$4,168 \$2,861 \$2,784 \$2,707 \$1,612 \$1,573 \$1,534 \$2,021 \$1,963 \$1,905 \$2,920 \$2,818 \$2,715 \$1,864 \$1,814 \$1,763 \$4,085 2019 \$2,424 \$2,366 \$2,308 \$3,040 \$2,953 \$2,865 \$4,393 \$4,239 \$4,084 \$2,804 \$2,728 \$2,653 \$990 \$966 \$942 \$1,241 \$1,205 \$1,170 \$1,793 \$1,730 \$1,667 \$1,144 \$1,114 \$1,083 \$3,414 2020 \$2,376 \$2,319 \$2,262 \$2,979 \$2,894 \$2,808 \$4,305 \$4,154 \$4,003 \$2,748 \$2,674 \$2,600 \$988 \$964 \$941 \$1,239 \$1,203 \$1,168 \$1,790 \$1,727 \$1,664 \$1,143 \$1,112 \$1,081 \$3,364 2021 \$2,328 \$2,272 \$2,217 \$2,920 \$2,836 \$2,752 \$4,219 \$4,071 \$3,923 \$2,693 \$2,620 \$2,548 \$986 \$963 \$939 \$1,237 \$1,202 \$1,166 \$1,788 \$1,725 \$1,662 \$1,141 \$1,110 \$1,079 \$3,315 2022 \$2,282 \$2,227 \$2,172 \$2,861 \$2,779 \$2,697 \$4,135 \$3,989 \$3,844 \$2,639 \$2,568 \$2,497 \$985 \$961 \$938 \$1,235 \$1,200 \$1,164 \$1,785 \$1,722 \$1,659 \$1,139 \$1,108 \$1,078 \$3,267 2023 \$2,236 \$2,182 \$2,129 \$2,804 \$2,724 \$2,643 \$4,052 \$3,910 \$3,767 \$2,586 \$2,516 \$2,447 \$984 \$960 \$936 \$1,233 \$1,198 \$1,162 \$1,782 \$1,720 \$1,657 \$1,137 \$1,107 \$1,076 \$3,220 2024 \$2,191 \$2,139 \$2,086 \$2,748 \$2,669 \$2,590 \$3,971 \$3,832 \$3,692 \$2,534 \$2,466 \$2,398 \$982 \$959 \$935 \$1,232 \$1,196 \$1,161 \$1,780 \$1,717 \$1,655 \$1,136 \$1,105 \$1,075 \$3,174 2025 \$2,148 \$2,096 \$2,044 \$2,693 \$2,616 \$2,538 \$3,892 \$3,755 \$3,618 \$2,484 \$2,417 \$2,350 \$981 \$957 \$934 \$1,230 \$1,195 \$1,159 \$1,777 \$1,715 \$1,652 \$1,134 \$1,104 \$1,073 \$3,128 3-197 ------- Technologies Considered in the Agencies' Analysis TC TC TC TC TC TC TC TC TC TC TC Small car Small car Standard car Standard car Standard car Large car Large car Large car Small MPV Small MPV Small MPV 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 8% 13% 3% 8% 13% 2% 7% 12% 3% 8% 13% \$4,040 \$3,940 \$5,191 \$5,042 \$4,892 \$7,501 \$7,237 \$6,974 \$4,787 \$4,658 \$4,529 \$3,987 \$3,889 \$5,123 \$4,976 \$4,829 \$7,403 \$7,143 \$6,883 \$4,725 \$4,597 \$4,470 \$3,332 \$3,250 \$4,281 \$4,158 \$4,035 \$6,186 \$5,969 \$5,752 \$3,948 \$3,842 \$3,735 \$3,283 \$3,202 \$4,218 \$4,097 \$3,976 \$6,096 \$5,881 \$5,667 \$3,890 \$3,785 \$3,681 \$3,235 \$3,156 \$4,157 \$4,037 \$3,918 \$6,007 \$5,796 \$5,585 \$3,834 \$3,730 \$3,627 \$3,188 \$3,110 \$4,097 \$3,979 \$3,861 \$5,920 \$5,712 \$5,504 \$3,778 \$3,676 \$3,574 \$3,142 \$3,065 \$4,038 \$3,922 \$3,805 \$5,834 \$5,629 \$5,424 \$3,724 \$3,623 \$3,523 \$3,097 \$3,021 \$3,980 \$3,865 \$3,751 \$5,751 \$5,549 \$5,347 \$3,670 \$3,571 \$3,472 \$3,053 \$2,978 \$3,923 \$3,810 \$3,697 \$5,669 \$5,470 \$5,270 \$3,618 \$3,520 \$3,423 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-108 Costs for PHEV40 Non-Battery Components for the 2008 Baseline (2010\$) Cost type Vehicle class DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC Applied WR Small car Small car Stdcar Stdcar Large car Large car Small MPV Small MPV Small car Small car Stdcar Stdcar Large car Large car Small MPV Small MPV Small car Small car Stdcar Stdcar Large car Large car Small MPV Small MPV Net WR 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 2017 2% 7% 3% 8% 1% 6% 3% 8% 2% 7% 3% 8% 1% 6% 3% 8% 2% 7% 3% 8% 1% 6% 3% 8% 2018 \$2,474 \$2,417 \$3 \$2 050 966 \$4,420 \$4 \$2 \$2 \$1 \$1 \$1 \$1 \$2 \$2 \$1 \$1 \$4 \$3 \$5 \$4 268 784 712 583 546 951 897 827 730 781 735 057 964 002 863 \$7,247 \$6 \$4 998 565 \$4,447 2019 \$2,425 \$2,369 \$2 989 \$2,906 \$4,331 \$4 183 \$2,728 \$2,658 \$1,580 \$1,543 \$1 948 \$1,893 \$2 822 \$2,725 \$1,777 \$1,731 \$4,005 \$3 912 \$4,937 \$4,800 \$7 153 \$6,907 \$4 \$4 505 389 \$2 \$2 \$2 \$2 2020 376 322 930 848 \$4,245 \$4 \$2 \$2 099 673 604 \$970 \$948 \$1 \$1 \$1 \$1 \$1 \$1 \$3 \$3 \$4 \$4 \$5 \$5 \$3 \$3 196 163 732 673 091 063 346 269 125 Oil 977 772 765 668 \$2 2021 329 \$2,275 \$2 871 \$2,791 \$4,160 \$4,017 \$2,620 \$2 552 \$968 \$946 \$1 194 \$1,161 \$1,730 \$1,670 \$1,089 \$1,061 \$3,297 \$3,221 \$4,065 \$3,952 \$5 889 \$5,687 \$3,709 \$3,614 2022 \$2,282 \$2,230 \$2 814 \$2,735 \$4,076 \$3 937 \$2,568 \$2 501 \$967 \$945 \$1 192 \$1,159 \$1,727 \$1,668 \$1,088 \$1,060 \$3,249 \$3 174 \$4,006 \$3,894 \$5 804 \$5,605 \$3,655 \$3 561 2023 \$2,237 \$2,185 \$2,757 \$2,681 \$3,995 \$3 858 \$2,516 \$2,451 \$966 \$943 \$1 190 \$1,157 \$1,725 \$1,665 \$1,086 \$1,058 \$3,202 \$3 128 \$3,948 \$3,838 \$5,719 \$5,523 \$3,602 \$3 510 \$2 2024 192 \$2,141 \$2,702 \$2,627 \$3,915 \$3,781 \$2,466 \$2,402 \$964 \$942 \$1 189 \$1,156 \$1,722 \$1,663 \$1,085 \$1,057 \$3,156 \$3,083 \$3,891 \$3,783 \$5,637 \$5,444 \$3,550 \$3,459 \$2 2025 148 \$2,099 \$2,648 \$2,575 \$3,837 \$3,705 \$2,417 \$2 354 \$963 \$940 \$1 187 \$1,154 \$1,719 \$1,661 \$1,083 \$1,055 \$3,111 \$3,039 \$3,835 \$3,728 \$5 556 \$5,366 \$3 500 \$3,409 \$2 \$2 \$2 \$2 \$3 \$3 \$2 \$2 105 057 595 523 760 631 368 307 \$961 \$939 \$1 \$1 \$1 \$1 \$1 \$1 \$3 \$2 \$3 \$3 185 152 717 658 081 054 066 996 780 675 \$5,477 \$5,289 \$3,450 \$3 361 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-109 Costs for PHEV40 Non-Battery Components for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC Vehicle class Small car Small car Stdcar Std car Large car Large car Small MPV Small MPV Applied WR 15% 20% 15% 20% 15% 20% 15% 20% Net WR 3% 8% 3% 8% 2% 7% 3% 8% 2017 \$2,524 \$2,464 \$3,166 \$3,075 \$4,574 \$4,414 \$2,919 \$2,841 2018 \$2,474 \$2,414 \$3,102 \$3,013 \$4,483 \$4,325 \$2,861 \$2,784 2019 \$2,424 \$2,366 \$3,040 \$2,953 \$4,393 \$4,239 \$2,804 \$2,728 2020 \$2,376 \$2,319 \$2,979 \$2,894 \$4,305 \$4,154 \$2,748 \$2,674 2021 \$2,328 \$2,272 \$2,920 \$2,836 \$4,219 \$4,071 \$2,693 \$2,620 2022 \$2,282 \$2,227 \$2,861 \$2,779 \$4,135 \$3,989 \$2,639 \$2,568 2023 \$2,236 \$2,182 \$2,804 \$2,724 \$4,052 \$3,910 \$2,586 \$2,516 2024 \$2,191 \$2,139 \$2,748 \$2,669 \$3,971 \$3,832 \$2,534 \$2,466 2025 \$2,148 \$2,096 \$2,693 \$2,616 \$3,892 \$3,755 \$2,484 \$2,417 3-198 ------- Technologies Considered in the Agencies' Analysis 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC Small car Small car Stdcar Std car Large car Large car Small MPV Small MPV Small car Small car Std car Std car Large car Large car Small MPV Small MPV 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 15% 20% 3% 8% 3% 8% 2% 7% 3% 8% 3% 8% 3% 8% 2% 7% 3% 8% \$1,615 \$1,576 \$2,025 \$1,967 \$2,926 \$2,824 \$1,868 \$1,817 \$4,139 \$4,040 \$5,191 \$5,042 \$7,501 \$7,237 \$4,787 \$4,658 \$1,612 \$1,573 \$2,021 \$1,963 \$2,920 \$2,818 \$1,864 \$1,814 \$4,085 \$3,987 \$5,123 \$4,976 \$7,403 \$7,143 \$4,725 \$4,597 \$990 \$966 \$1,241 \$1,205 \$1,793 \$1,730 \$1,144 \$1,114 \$3,414 \$3,332 \$4,281 \$4,158 \$6,186 \$5,969 \$3,948 \$3,842 \$988 \$964 \$1,239 \$1,203 \$1,790 \$1,727 \$1,143 \$1,112 \$3,364 \$3,283 \$4,218 \$4,097 \$6,096 \$5,881 \$3,890 \$3,785 \$986 \$963 \$1,237 \$1,202 \$1,788 \$1,725 \$1,141 \$1,110 \$3,315 \$3,235 \$4,157 \$4,037 \$6,007 \$5,796 \$3,834 \$3,730 \$985 \$961 \$1,235 \$1,200 \$1,785 \$1,722 \$1,139 \$1,108 \$3,267 \$3,188 \$4,097 \$3,979 \$5,920 \$5,712 \$3,778 \$3,676 \$984 \$960 \$1,233 \$1,198 \$1,782 \$1,720 \$1,137 \$1,107 \$3,220 \$3,142 \$4,038 \$3,922 \$5,834 \$5,629 \$3,724 \$3,623 \$982 \$959 \$1,232 \$1,196 \$1,780 \$1,717 \$1,136 \$1,105 \$3,174 \$3,097 \$3,980 \$3,865 \$5,751 \$5,549 \$3,670 \$3,571 \$981 \$957 \$1,230 \$1,195 \$1,777 \$1,715 \$1,134 \$1,104 \$3,128 \$3,053 \$3,923 \$3,810 \$5,669 \$5,470 \$3,618 \$3,520 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-110 Costs for EV75 Non-Battery Components for the 2008 Baseline (2010\$) Cost type Vehicle class DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC Applied WR Small car Small car Small car Stdcar Stdcar Stdcar Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Stdcar Stdcar Stdcar Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Stdcar Stdcar Stdcar Large car Large car Large car Small MPV Small MPV Net 2017 WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 10% 15% 20% 10% 15% 20% 10% 15% 20% 9% 14% 19% 10% 15% 20% 10% 15% 20% 10% 15% 20% 9% 14% 19% 10% 15% 20% 10% 15% 20% 10% 15% 20% 9% 14% 2018 \$199 \$124 \$48 \$921 \$809 \$697 \$1,659 \$1,458 \$1,257 -\$183 -\$278 -\$373 \$153 \$95 \$37 \$709 \$623 \$536 \$1,277 \$1,123 \$968 -\$141 -\$214 -\$287 \$352 \$219 \$85 \$1,630 \$1,431 \$1,233 \$2,936 \$2,581 \$2,226 -\$324 -\$492 2019 \$193 \$120 \$47 \$893 \$784 \$676 \$1,609 \$1,414 \$1,220 -\$177 -\$269 -\$362 \$153 \$95 \$37 \$707 \$621 \$535 \$1,273 \$1,119 \$965 -\$140 -\$213 -\$286 \$346 \$215 \$84 \$1,600 \$1,405 \$1,211 \$2,882 \$2,534 \$2,185 -\$318 -\$483 2020 \$187 \$116 \$45 \$866 \$761 \$655 \$1,560 \$1,372 \$1,183 -\$172 -\$261 -\$351 \$152 \$95 \$37 \$705 \$619 \$533 \$1,270 \$1,116 \$963 -\$140 -\$213 -\$285 \$340 \$211 \$82 \$1,571 \$1,380 \$1,189 \$2,830 \$2,488 \$2,146 -\$312 -\$474 2021 \$182 \$113 \$44 \$840 \$738 \$636 \$1 \$1 \$1 ,514 ,331 ,148 -\$167 -\$254 -\$340 \$152 \$94 \$37 \$703 \$618 \$532 \$1 \$1 ,266 ,113 \$960 -\$140 -\$212 -\$285 \$334 \$207 \$81 \$1 \$1 \$1 ,543 ,356 ,168 \$2,780 \$2,444 \$2,108 -\$306 -\$466 2022 \$176 \$109 \$43 \$815 \$716 \$617 \$1,468 \$1,291 \$1,113 -\$162 -\$246 -\$330 \$152 \$94 \$37 \$701 \$616 \$531 \$1,263 \$1,110 \$958 -\$139 -\$212 -\$284 \$328 \$204 \$79 \$1,516 \$1,332 \$1,147 \$2,731 \$2,401 \$2,071 -\$301 -\$458 2023 \$171 \$106 \$41 \$791 \$694 \$598 \$1,424 \$1 \$1 ,252 ,080 -\$157 -\$239 -\$320 \$151 \$94 \$37 \$699 \$614 \$529 \$1 \$1 ,260 ,107 \$955 -\$139 -\$211 -\$283 \$322 \$200 \$78 \$1,490 \$1 \$1 ,309 ,127 \$2,684 \$2,359 \$2,035 -\$296 -\$450 2024 \$168 \$104 \$40 \$775 \$681 \$586 \$1,396 \$1,227 \$1,058 -\$154 -\$234 -\$314 \$151 \$94 \$36 \$698 \$613 \$528 \$1,258 \$1,106 \$954 -\$139 -\$211 -\$283 \$319 \$198 \$77 \$1,473 \$1,294 \$1,114 \$2,654 \$2,333 \$2,012 -\$293 -\$444 2025 \$164 \$102 \$40 \$759 \$667 \$574 \$1 \$1 \$1 ,368 ,202 ,037 -\$151 -\$229 -\$307 \$151 \$94 \$36 \$697 \$612 \$527 \$1 \$1 ,256 ,104 \$952 -\$138 -\$210 -\$282 \$315 \$195 \$76 \$1,457 \$1 \$1 ,279 ,102 \$2,624 \$2,306 \$1 ,989 -\$289 -\$439 \$161 \$100 \$39 \$744 \$654 \$563 \$1 \$1 \$1 ,340 ,178 ,016 -\$148 -\$225 -\$301 \$97 \$60 \$23 \$449 \$394 \$339 \$808 \$710 \$613 -\$89 -\$135 -\$182 \$258 \$160 \$62 \$1 \$1 ,193 ,048 \$902 \$2,149 \$1 \$1 ,889 ,629 -\$237 -\$360 3-199 ------- Technologies Considered in the Agencies' Analysis TC Small MPV 20% 19% -\$660 -\$636 -\$625 -\$614 -\$603 -\$596 -\$590 -\$483 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-111 Costs for EV75 Non-Battery Components for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC Vehicle class Small car Small car Small car Std car Std car Std car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Std car Std car Std car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Std car Std car Std car Large car Large car Large car Small MPV Small MPV Small MPV Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% Net WR 10% 15% 20% 9% 14% 19% 10% 15% 20% 9% 14% 19% 10% 15% 20% 9% 14% 19% 10% 15% 20% 9% 14% 19% 10% 15% 20% 9% 14% 19% 10% 15% 20% 9% 14% 19% 2017 \$250 \$172 \$93 \$1,043 \$926 \$808 \$1,799 \$1,592 \$1,385 -\$51 -\$152 -\$254 \$192 \$132 \$72 \$803 \$713 \$623 \$1,386 \$1,226 \$1,066 -\$39 -\$117 -\$195 \$442 \$304 \$165 \$1,847 \$1,639 \$1,431 \$3,185 \$2,818 \$2,451 -\$90 -\$270 -\$449 2018 \$242 \$166 \$91 \$1,012 \$898 \$784 \$1,745 \$1,544 \$1,343 -\$49 -\$148 -\$246 \$192 \$132 \$72 \$801 \$711 \$621 \$1,382 \$1,222 \$1,063 -\$39 -\$117 -\$195 \$434 \$298 \$162 \$1,813 \$1,609 \$1,405 \$3,127 \$2,767 \$2,406 -\$88 -\$265 -\$441 2019 \$235 \$161 \$88 \$982 \$871 \$761 \$1,693 \$1,498 \$1,303 -\$48 -\$143 -\$239 \$191 \$131 \$72 \$799 \$709 \$619 \$1,378 \$1,219 \$1,060 -\$39 -\$117 -\$194 \$426 \$293 \$159 \$1,781 \$1,580 \$1,380 \$3,071 \$2,717 \$2,363 -\$86 -\$260 -\$433 2020 \$228 \$157 \$85 \$952 \$845 \$738 \$1,642 \$1,453 \$1,264 -\$46 -\$139 -\$232 \$191 \$131 \$71 \$797 \$707 \$617 \$1,374 \$1,216 \$1,057 -\$39 -\$116 -\$194 \$419 \$288 \$157 \$1,749 \$1,552 \$1,355 \$3,016 \$2,669 \$2,321 -\$85 -\$255 -\$426 2021 \$221 \$152 \$83 \$924 \$820 \$716 \$1,593 \$1,409 \$1,226 -\$45 -\$135 -\$225 \$190 \$131 \$71 \$795 \$705 \$616 \$1,370 \$1,212 \$1,054 -\$39 -\$116 -\$193 \$412 \$283 \$154 \$1,718 \$1,525 \$1,331 \$2,963 \$2,622 \$2,280 -\$83 -\$251 -\$418 2022 \$215 \$147 \$80 \$896 \$795 \$694 \$1,545 \$1,367 \$1,189 -\$44 -\$131 -\$218 \$190 \$130 \$71 \$793 \$703 \$614 \$1,367 \$1,209 \$1,052 -\$38 -\$116 -\$193 \$404 \$278 \$151 \$1,689 \$1,498 \$1,308 \$2,912 \$2,576 \$2,241 -\$82 -\$246 -\$411 2023 \$210 \$144 \$79 \$878 \$779 \$680 \$1,514 \$1,340 \$1,165 -\$43 -\$128 -\$214 \$189 \$130 \$71 \$791 \$702 \$613 \$1,365 \$1,207 \$1,050 -\$38 -\$115 -\$193 \$400 \$275 \$149 \$1,669 \$1,481 \$1,293 \$2,879 \$2,547 \$2,215 -\$81 -\$244 -\$406 2024 \$206 \$142 \$77 \$861 \$764 \$667 \$1,484 \$1,313 \$1,142 -\$42 -\$126 -\$209 \$189 \$130 \$71 \$790 \$701 \$612 \$1,362 \$1,205 \$1,048 -\$38 -\$115 -\$192 \$395 \$271 \$148 \$1,651 \$1,465 \$1,279 \$2,846 \$2,518 \$2,190 -\$80 -\$241 -\$402 2025 \$202 \$139 \$75 \$843 \$748 \$653 \$1,454 \$1,287 \$1,119 -\$41 -\$123 -\$205 \$122 \$84 \$46 \$508 \$451 \$394 \$877 \$776 \$675 -\$25 -\$74 -\$124 \$324 \$222 \$121 \$1,352 \$1,199 \$1,047 \$2,331 \$2,062 \$1,794 -\$66 -\$197 -\$329 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-112 Costs for EV100 Non-Battery Components for the 2008 Baseline (2010\$) Cost type EPA Vehicle class DMC DMC DMC DMC DMC Applied WR Small car Small car Small car Std car Std car Net WR 10% 15% 20% 10% 15% 2017 4% 9% 14% 4% 9% 2018 \$290 \$214 \$139 \$1,055 \$943 2019 \$281 \$208 \$135 \$1,024 \$915 2020 \$273 \$202 \$130 \$993 \$887 2021 \$264 \$195 \$127 \$963 \$861 2022 \$256 \$190 \$123 \$934 \$835 2023 \$249 \$184 \$119 \$906 \$810 2024 \$244 \$180 \$117 \$888 \$794 2025 \$239 \$177 \$114 \$870 \$778 \$234 \$173 \$112 \$853 \$762 3-200 ------- Technologies Considered in the Agencies' Analysis DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC Stdcar Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Stdcar Stdcar Stdcar Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Stdcar Stdcar Stdcar Large car Large car Large car Small MPV Small MPV Small MPV 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 14% 5% 10% 15% 3% 8% 13% 4% 9% 14% 4% 9% 14% 5% 10% 15% 3% 8% 13% 4% 9% 14% 4% 9% 14% 5% 10% 15% 3% 8% 13% \$831 \$1,859 \$1,659 \$1,458 -\$69 -\$164 -\$259 \$223 \$165 \$107 \$813 \$726 \$640 \$1,432 \$1,277 \$1,123 -\$53 -\$126 -\$199 \$513 \$379 \$245 \$1,868 \$1,669 \$1,471 \$3,291 \$2,936 \$2,581 -\$122 -\$290 -\$458 \$806 \$1,803 \$1,609 \$1,414 -\$67 -\$159 -\$251 \$222 \$164 \$106 \$810 \$724 \$638 \$1,427 \$1,273 \$1,119 -\$53 -\$126 -\$199 \$503 \$372 \$241 \$1,834 \$1,639 \$1,444 \$3,231 \$2,882 \$2,534 -\$120 -\$285 -\$450 \$782 \$1,749 \$1,560 \$1,372 -\$65 -\$154 -\$244 \$222 \$164 \$106 \$808 \$722 \$636 \$1,423 \$1,270 \$1,116 -\$53 -\$125 -\$198 \$494 \$366 \$237 \$1,801 \$1,610 \$1,418 \$3,173 \$2,830 \$2,488 -\$117 -\$280 -\$442 \$759 \$1,697 \$1,514 \$1,331 -\$63 -\$150 -\$236 \$221 \$164 \$106 \$806 \$720 \$635 \$1,420 \$1,266 \$1,113 -\$53 -\$125 -\$198 \$486 \$359 \$232 \$1,769 \$1,581 \$1,393 \$3,116 \$2,780 \$2,444 -\$115 -\$275 -\$434 \$736 \$1,646 \$1,468 \$1,291 -\$61 -\$145 -\$229 \$221 \$163 \$106 \$804 \$718 \$633 \$1,416 \$1,263 \$1,110 -\$52 -\$125 -\$197 \$477 \$353 \$228 \$1,738 \$1,553 \$1,369 \$3,062 \$2,731 \$2,401 -\$113 -\$270 -\$426 \$714 \$1,596 \$1,424 \$1,252 -\$59 -\$141 -\$222 \$220 \$163 \$105 \$802 \$716 \$631 \$1,412 \$1,260 \$1,107 -\$52 -\$124 -\$197 \$469 \$347 \$224 \$1,708 \$1,526 \$1,345 \$3,009 \$2,684 \$2,359 -\$111 -\$265 -\$419 \$699 \$1,565 \$1,396 \$1,227 -\$58 -\$138 -\$218 \$220 \$162 \$105 \$800 \$715 \$630 \$1,410 \$1,258 \$1,106 -\$52 -\$124 -\$196 \$464 \$343 \$222 \$1,688 \$1,509 \$1,330 \$2,974 \$2,654 \$2,333 -\$110 -\$262 -\$414 \$685 \$1,533 \$1,368 \$1,202 -\$57 -\$135 -\$213 \$219 \$162 \$105 \$799 \$714 \$629 \$1,408 \$1,256 \$1,104 -\$52 -\$124 -\$196 \$458 \$339 \$219 \$1,669 \$1,492 \$1,315 \$2,941 \$2,624 \$2,306 -\$109 -\$259 -\$409 \$672 \$1,503 \$1,340 \$1,178 -\$56 -\$132 -\$209 \$141 \$104 \$68 \$514 \$460 \$405 \$906 \$808 \$710 -\$34 -\$80 -\$126 \$375 \$277 \$180 \$1,367 \$1,222 \$1,077 \$2,408 \$2,149 \$1,889 -\$89 -\$212 -\$335 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-113 Costs for EV100 Non-Battery Components for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC DMC 1C 1C 1C 1C 1C 1C 1C EPA Vehicle class Small car Small car Small car Std car Std car Std car Large car Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Std car Std car Std car Large car Applied WR 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% Net WR 4% 9% 14% 3% 8% 13% 4% 9% 14% 3% 8% 13% 4% 9% 14% 3% 8% 13% 4% 2017 \$344 \$266 \$187 \$1,184 \$1,067 \$949 \$2,048 \$1,841 \$1,633 \$71 -\$30 -\$132 \$265 \$204 \$144 \$912 \$822 \$731 \$1,577 2018 \$333 \$258 \$182 \$1,149 \$1,035 \$921 \$1,987 \$1,786 \$1,584 \$69 -\$29 -\$128 \$264 \$204 \$144 \$909 \$819 \$729 \$1,573 2019 \$323 \$250 \$176 \$1,114 \$1,004 \$893 \$1,927 \$1,732 \$1,537 \$67 -\$29 -\$124 \$263 \$203 \$143 \$907 \$817 \$727 \$1,568 2020 \$314 \$242 \$171 \$1,081 \$974 \$866 \$1,869 \$1,680 \$1,491 \$65 -\$28 -\$120 \$262 \$203 \$143 \$904 \$815 \$725 \$1,564 2021 \$304 \$235 \$166 \$1,049 \$945 \$841 \$1,813 \$1,630 \$1,446 \$63 -\$27 -\$117 \$262 \$202 \$143 \$902 \$813 \$723 \$1,560 2022 \$295 \$228 \$161 \$1,017 \$916 \$815 \$1,759 \$1,581 \$1,403 \$61 -\$26 -\$113 \$261 \$202 \$142 \$900 \$810 \$721 \$1,556 2023 \$289 \$223 \$158 \$997 \$898 \$799 \$1,724 \$1,549 \$1,375 \$60 -\$26 -\$111 \$261 \$201 \$142 \$898 \$809 \$720 \$1,553 2024 \$284 \$219 \$154 \$977 \$880 \$783 \$1,689 \$1,518 \$1,347 \$59 -\$25 -\$109 \$260 \$201 \$142 \$897 \$808 \$719 \$1,551 2025 \$278 \$215 \$151 \$957 \$862 \$767 \$1,656 \$1,488 \$1,320 \$58 -\$25 -\$107 \$167 \$129 \$91 \$577 \$520 \$463 \$998 3-201 ------- Technologies Considered in the Agencies' Analysis 1C 1C 1C 1C 1C TC TC TC TC TC TC TC TC TC TC TC TC Large car Large car Small MPV Small MPV Small MPV Small car Small car Small car Std car Std car Std car Large car Large car Large car Small MPV Small MPV Small MPV 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 15% 20% 9% 14% 3% 8% 13% 4% 9% 14% 3% 8% 13% 4% 9% 14% 3% 8% 13% \$1,418 \$1,258 \$55 -\$23 -\$102 \$608 \$470 \$331 \$2,096 \$1,888 \$1,680 \$3,626 \$3,258 \$2,891 \$126 -\$54 -\$234 \$1,413 \$1,254 \$55 -\$23 -\$101 \$597 \$461 \$325 \$2,058 \$1,854 \$1,650 \$3,560 \$3,199 \$2,839 \$124 -\$53 -\$229 \$1,410 \$1,251 \$55 -\$23 -\$101 \$587 \$453 \$320 \$2,021 \$1,821 \$1,620 \$3,496 \$3,142 \$2,788 \$122 -\$52 -\$225 \$1,406 \$1,247 \$54 -\$23 -\$101 \$576 \$445 \$314 \$1,985 \$1,788 \$1,591 \$3,434 \$3,086 \$2,738 \$119 -\$51 -\$221 \$1,402 \$1,244 \$54 -\$23 -\$100 \$566 \$437 \$308 \$1,951 \$1,757 \$1,564 \$3,373 \$3,032 \$2,690 \$117 -\$50 -\$217 \$1,398 \$1,241 \$54 -\$23 -\$100 \$556 \$430 \$303 \$1,917 \$1,727 \$1,536 \$3,315 \$2,979 \$2,644 \$115 -\$49 -\$214 \$1,396 \$1,239 \$54 -\$23 -\$100 \$550 \$425 \$300 \$1,895 \$1,707 \$1,519 \$3,277 \$2,945 \$2,613 \$114 -\$49 -\$211 \$1,394 \$1,237 \$54 -\$23 -\$100 \$544 \$420 \$296 \$1,874 \$1,688 \$1,502 \$3,240 \$2,912 \$2,584 \$113 -\$48 -\$209 \$897 \$796 \$35 -\$15 -\$64 \$445 \$344 \$243 \$1,534 \$1,382 \$1,230 \$2,654 \$2,385 \$2,116 \$92 -\$39 -\$171 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-114 Costs for EV150 Non-Battery Components for the 2008 Baseline (2010\$) Cost Vehicle type class DMC DMC DMC DMC 1C 1C 1C 1C TC TC TC TC Applied WR Small car Std car Large car Small MPV Small car Std car Large car Small MPV Small car Std car Large car Small MPV Net WR 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 2017 2% 2% 3% 1% 2% 2% 3% 1% 2% 2% 3% 1% 2018 \$321 \$1,101 \$1,941 -\$30 \$247 \$848 \$1,494 -\$23 \$569 \$1,949 \$3,435 -\$54 2019 \$312 \$1,068 \$1,882 -\$29 \$247 \$845 \$1,490 -\$23 \$558 \$1,913 \$3,373 -\$53 2020 \$302 \$1,036 \$1,826 -\$29 \$246 \$843 \$1,486 -\$23 \$548 \$1,879 \$3,312 -\$52 2021 \$293 \$1,005 \$1,771 -\$28 \$245 \$841 \$1,482 -\$23 \$538 \$1,846 \$3,253 -\$51 2022 2023 \$284 \$975 \$1,718 -\$27 \$245 \$838 \$1,478 -\$23 \$529 \$1,813 \$3,196 -\$50 \$276 \$945 \$1,667 -\$26 \$244 \$836 \$1,474 -\$23 \$520 \$1,782 \$3,141 -\$49 2024 2025 \$270 \$927 \$1,633 -\$26 \$244 \$835 \$1,472 -\$23 \$514 \$1,761 \$3,105 -\$49 \$265 \$908 \$1,601 -\$25 \$243 \$834 \$1,469 -\$23 \$508 \$1,742 \$3,070 -\$48 \$260 \$890 \$1,569 -\$25 \$157 \$536 \$946 -\$15 \$416 \$1,426 \$2,514 -\$39 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Table 3-115 Costs for EV150 Non-Battery Components for the 2010 Baseline (2010\$) Cost type DMC DMC DMC DMC 1C 1C 1C 1C TC TC TC TC Vehicle class Small car Std car Large car Small MPV Small car Std car Large car Small MPV Small car Std car Large car Small MPV Applied WR 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% 20% Net WR 1% 1% 3% 1% 1% 1% 3% 1% 1% 1% 3% 1% 2017 \$391 \$1,231 \$2,090 \$112 \$301 \$948 \$1,609 \$86 \$692 \$2,180 \$3,699 \$198 2018 \$379 \$1,194 \$2,027 \$108 \$300 \$945 \$1,605 \$86 \$679 \$2,140 \$3,632 \$194 2019 \$368 \$1,159 \$1,966 \$105 \$299 \$943 \$1,600 \$86 \$667 \$2,101 \$3,566 \$191 2020 \$357 \$1,124 \$1,907 \$102 \$298 \$940 \$1,596 \$85 \$655 \$2,064 \$3,503 \$187 2021 \$346 \$1,090 \$1,850 \$99 \$298 \$938 \$1,592 \$85 \$643 \$2,028 \$3,442 \$184 2022 \$336 \$1,057 \$1,795 \$96 \$297 \$935 \$1,587 \$85 \$632 \$1,993 \$3,382 \$181 2023 \$329 \$1,036 \$1,759 \$94 \$296 \$934 \$1,585 \$85 \$625 \$1,970 \$3,344 \$179 2024 \$322 \$1,016 \$1,724 \$92 \$296 \$932 \$1,582 \$85 \$618 \$1,948 \$3,306 \$177 2025 \$316 \$995 \$1,689 \$90 \$190 \$600 \$1,018 \$54 \$506 \$1,595 \$2,707 \$145 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost For Mild HEV non-battery components, the agencies have used a combination of cost sources which include the FEV teardown of a Saturn Vue along with estimates used for P2 3-202 ------- Technologies Considered in the Agencies' Analysis HEVs as described above. For the electrical power distribution and control system and the DC-DC converter, estimates presented in the NPRM for subcompacts were used with a presumed 20% weight reduction because those systems were estimated to include a 16 kW motor (essentially the same as the 15 kW motor assumed for the Mild HEV technology). These costs and the FEV Saturn Vue teardown costs we used are shown in Table 3-116. Table 3-116 FEV Teardown Results & P2 HEV Values used for MHEV Non-Battery Direct Manufacturing Cost Estimates System Cooling subsystem (including water pumps) Accessory drive subsystem Body system Brake system Climate control system Transmission oil pump and filter subsystem Generator/alternator and regulatory subsystem Electrical power distribution & control system DC-DC converter Total Teardown result (2007\$) \$88.71 \$30.75 \$14.83 \$42.30 \$0 \$53.86 \$51.94 P2HEV (2009\$)a \$203.22 \$115.33 2010\$ \$92.37 \$32.02 \$15.44 \$44.05 \$0 \$56.09 \$54.09 \$205.25 \$116.48 \$615.79 a See the draft Joint TSD, Table 3-80, 20% WR (EPA-420-D-1 1-901, November 2011). For Mild HEV non-battery components, the direct manufacturing costs shown in Table 3-116 are considered applicable MY 2012. The agencies consider the Mild FIEV non- battery component technologies to be on the flat portion of the learning curve during the 2017-2025 timeframe. The agencies have applied a medium complexity ICM of 1.39 through 2018 then 1.29 thereafter. The resultant costs used in this final analysis are shown in Table 3-117. Table 3-117 Costs for Mild HEV Non-Battery Components for both the 2008 and 2010 Baselines (2010\$) Cost type DMC 1C TC Vehicle class All All All 2017 \$534 \$235 \$769 2018 \$524 \$234 \$758 2019 \$513 \$175 \$688 2020 \$503 \$175 \$678 2021 \$493 \$175 \$667 2022 \$483 \$174 \$657 2023 \$473 \$174 \$647 2024 \$464 \$174 \$637 2025 \$455 \$173 \$628 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3.4.4 Hardware costs for charging grid-connected vehicles Grid-connected vehicles such as EVs and PFLEVs require a means to charge their on- board batteries to enable their electric range capabilities. These vehicles require certain hardware to charge, both on-vehicle and off-vehicle. The agencies' September 2010 Technical 3-203 ------- Technologies Considered in the Agencies' Analysis Assessment Report contains an in-depth analysis of the topic of charging and infrastructure. The TAR analysis and assumptions did not receive any significant comment, and a review of the current state of the industry indicates the assumptions in the TAR are still valid. Therefore, the assumptions for the cost of Electric Vehicle Support Equipment (EVSE) are unchanged. Additionally, while some of the characteristics of the modeled grid-connected vehicles such as battery size and energy demand have changed somewhat due to further analysis, the application of Level 1 and Level 2 charging by vehicle type based on charge time has not changed. Three charging levels are currently under consideration. Level 1 charging uses a standard 120 volt (V), 15-20 amps (A) rated (12-16 A usable) circuit and is available in standard residential and commercial buildings. Level 2 charging uses a single phase, 240 V, 20-80 A circuit and allows much shorter charge times. Level 3 charging—sometimes colloquially called "quick" or "fast" charging—uses a 480 V, three-phase circuit, available in mainly industrial areas, typically providing 60-150 kW of off-board charging power. It is expected that 97 to 99% of charging will take place at home, so a cost for a home charger, appropriate to the duty cycle of the vehicle, is added to the vehicle cost. Level 3 charging is available to commercial users and vehicles that charge at Level 3 stations will be assumed to pay at the charge station for the convenience of fast charging. Therefore Level 3 charger costs are not included in overall vehicle cost. The specific equipment required for charging a grid-connected vehicle consists of the following: Charger: A charger that converts electricity from alternating current (AC) from the electricity source to direct current (DC) required for the battery, and also converts the incoming 120 or 240 volt current to 300 or higher volts. Grid-connected vehicles carry an on- board charger capable of accepting AC current from a wall plug (Level 1 circuit) or, from a Level 2 charging station. On-board charger power capability ranges from 1.4 to 10 kW and is usually proportional to the vehicle's battery capacity. The lowest charging power, 1.4 kW, is expected only when grid-connected vehicles are connected to 120 volt (Level 1) outlets, and all currently known PHEV and EV on-board chargers are expected to provide at least 3.3 kW charging when connected to a Level 2 (220 volt, 20+ A) charging station. The latest SAE connection recommended practice, J1772, allows for delivery of up to -19 kW to an on-board vehicle charger. For higher capacity charging under Level 3, a charging station that delivers DC current directly to the vehicle's battery is incorporated off-board in the wall or pedestal mounted. Charging Station: The charging station needed to safely deliver energy from the electric circuit to the vehicle, called electric vehicle support equipment (EVSE). The EVSE may at a minimum, be a specialized cordset that connects a household Level 1/120V socket to the vehicle; otherwise, the EVSE will include a cordset and a charging station (a wall or pedestal mounted box incorporating a charger and other equipment). Charging stations may include optional advanced features such as timers to delay charging until off-peak hours, communications equipment to allow the utility to regulate charging, or even electricity metering capabilities. Stakeholders are working on which features are best located on the EVSE or on the vehicle itself, and it is possible that redundant capabilities and features may 3-204 ------- Technologies Considered in the Agencies' Analysis be present in both the vehicle and EVSEs in the near future until these issues are worked out. EVSE and vehicle manufacturers are also working to ensure that current SAE-compliant "basic" EVSEs are charge-compatible with future grid-connected vehicles. Dedicated Circuit: A Level 1 circuit is standard household current, 120V AC, rated at 15 or 20 A (12 or 16 A usable). A Level 2 circuit is rated at 208 to 240V and up to 80 A and is similar to the type of circuit that powers electric stoves (up to 50 A) and dryers (usually 30 A). Generally, Level 1 and 2 circuits used for electric vehicle recharging must be dedicated circuits, i.e., there cannot be other appliances on that circuit. For a Level 2 circuit, the homeowner or other user must install a charging station and will need a permit. A homeowner may choose to install the charger on a separately-metered circuit to take advantage of special electrical rates for off-peak charging, where available. In addition to the costs of purchasing and installing charging equipment, charging station installation may include the costs of upgrading existing electrical panels and installing the electrical connection from the panel to the desired station location. These costs may be dramatically lowered if new construction incorporates the panel box and wiring required for charging stations, or even includes charging stations or outlets for charging stations as standard equipment. The current costs of charging stations are highly variable depending on the level of service (and alternative power capabilities within these categories), location (individual residence, grouped residences, retail or business, parking lot or garage), level of sophistication of the station, and installation requirements, including electrical upgrading requirements. Estimated costs for charging stations are included in Table 3-118 below. Table 3-118: Estimated Costs for Charging Stations Used in the 2010 TAR (2008\$) Level Location Equipment Installation 1 Single Residence \$30- \$200 (charge cord only, included at no cost to consumer with EV/PHEV) when an accessible household plug (e.g., in a garage or adjacent to a driveway) with a ground fault interrupter is already available \$400-\$ 1000+ may be necessary depending on difficulty of installing a new circuit at the desired location, but in most cases, owners with sufficient panel capacity would opt for a more capable 220 VAC Level 2 installation instead of a Level 1 dedicated circuit because the additional installation cost is only marginally higher 3-205 ------- Technologies Considered in the Agencies' Analysis Residential, Apartment Complex, or Fleet Depotb 3.3 kW EVSE (each): \$300- \$4,000 6.6 kW EVSE (each): \$400- \$4,000 3.3- 6.6 kW installation cost: \$400-\$2,300 without wiring/service panel upgrade, or \$2,000-\$5,000 with panel upgrade rets: 77,78,79,80,3 a Detailed information on charger cost for each charging level and location and specific sources for cost estimates are available in the TAR, Appendix G. b Level 2 EVSE installation costs vary considerably for single-family residences, multi-family residences, and fleet depots, depending upon the need for wiring and service panel upgrades. The range depicted here reflects the anticipated variability of these costs. However, EPRI estimates that the typical residential Level 2 installation costs to be approximately \$1,500. See the TAR, Appendix G for additional information. 3.4.4.1 Application of charging level by vehicle type The home charging availability for a specific consumer will need to be differentiated among EV/PHEVs with different battery capacity. The electric outlets in existing homes are most likely ready for Level 1 charging, which is about sufficient for fully recharging a PHEV20 SUV during normal nighttime, provided the outlet is not being heavily utilized by other loads. Shorter available charging time or owning a PHEV or an EV with a larger battery make the capability to fully charge overnight with a Level 1 system less likely, but upgrading to a Level 2 system in such cases will allow full recharge to happen more quickly. Table 3-119 shows the application of charge level by vehicle type and range. Charging types were chosen based on nominal time to charge a fully-depleted battery in a vehicle with 0% net weight reduction. Charge times exceeding 9 hours for Level 1 were deemed unacceptable and Level 2 charging was specified. For charge times between 6 hours and 9 hours on Level 1, a mix of Level 1 and Level 2 was specified. This was done to recognize the varying consumer value of faster, but more expensive, Level 2 charging over Level 1 charging. Table 3-119: Charger Type by Vehicle Technology and Class EPA Vehicle Class Small car Standard Car Large Car Small MPV Large MPV Truck PHEV20 100% LI 100% LI 100% LI 100% LI 100% LI 50% LI 50% L2 PHEV40 25% LI 75% L2 10% LI 90% L2 100% L2 100% L2 100% L2 100% L2 EV75 100% L2 100% L2 100% L2 100% L2 100% L2 100% L2 EV100 100% L2 100% L2 100% L2 100% L2 100% L2 100% L2 EV150 100% L2 100% L2 100% L2 100% L2 100% L2 100% L2 3-206 ------- Technologies Considered in the Agencies' Analysis For this final rule, consistent with the proposal, the resultant costs associated with in- home chargers and installation of in-home chargers are included in the total cost for an EV and or PHEV. However, here we summarize specially the costs for chargers and installation labor. The agencies have estimated the DMC of a level 1 charge cord at \$31 (2010\$) based on typical costs of similar electrical equipment sold to consumers today and that for a level 2 charger at \$204 (2010\$). Labor associated with installing either of these chargers is estimated at \$1,020 (2010\$). Further, we have estimated that all PHEV20 vehicles (PHEVs with a 20 mile range) would be charged via a level 1 charger and that all EVs, regardless of range, would be charged via a level 2 charger. For the PFLEV40 vehicles (PFLEVs with a 40 mile range), we have estimated that: 25% of small cars would be charged with a level 1 charger with the remainder charged via a level 2 charger; 10% of standard cars would be charged with a level 1 charger with the remainder charged via a level 2 charger; and all remaining PFLEV 40 vehicles would be charged via a level 2 charger. All costs presented here are considered applicable in the 2025 model year. The agencies have applied the learning curve presented in Section 3.2.3 to all charger costs. The agencies have also applied a Highl ICM of 1.56 through 2024 then 1.34 thereafter. Installation costs, being labor costs, have no learning impacts or ICMs applied. The resultant costs are shown in Table 3-120. Table 3-120 Costs for EV/PHEV In-home Chargers (2010\$) Cost type DMC DMC DMC 1C 1C 1C TC TC TC TC Technology PHEV20 Charger PHEV40 Charger EV Charger PHEV20 Charger PHEV40 Charger EV Charger PHEV20 Charger PHEV40 Charger EV Charger Charger labor EPA Vehicle Class All Small car Std car Large car Small MPV All All Small car Std car Large car Small MPV All All Small car Std car Large car Small MPV All All 2017 \$60 \$314 \$365 \$398 \$398 \$19 \$100 \$117 \$128 \$128 \$79 \$414 \$481 \$526 \$526 \$1,020 2018 \$48 \$251 \$292 \$319 \$319 \$18 \$96 \$112 \$122 \$122 \$66 \$347 \$404 \$441 \$441 \$1,020 2019 \$48 \$251 \$292 \$319 \$319 \$18 \$96 \$112 \$122 \$122 \$66 \$347 \$404 \$441 \$441 \$1,020 2020 \$38 \$201 \$233 \$255 \$255 \$18 \$93 \$108 \$118 \$118 \$56 \$294 \$342 \$373 \$373 \$1,020 2021 \$38 \$201 \$233 \$255 \$255 \$18 \$93 \$108 \$118 \$118 \$56 \$294 \$342 \$373 \$373 \$1,020 2022 \$38 \$201 \$233 \$255 \$255 \$18 \$93 \$108 \$118 \$118 \$56 \$294 \$342 \$373 \$373 \$1,020 2023 \$38 \$201 \$233 \$255 \$255 \$18 \$93 \$108 \$118 \$118 \$56 \$294 \$342 \$373 \$373 \$1,020 2024 \$38 \$201 \$233 \$255 \$255 \$18 \$93 \$108 \$118 \$118 \$56 \$294 \$342 \$373 \$373 \$1,020 2025 \$31 \$161 \$187 \$204 \$204 \$11 \$55 \$64 \$70 \$70 \$41 \$216 \$251 \$274 \$274 \$1,020 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3-207 ------- Technologies Considered in the Agencies' Analysis 3.4.5 Other Technologies Assessed that Reduce CO2 and Improve Fuel Economy In addition to the technologies already mentioned above, the other technologies generally considered in the agencies' analysis are described below. They fall into five broad categories: engine technologies, transmission technologies, vehicle technologies, electrification/accessory technologies, hybrid technologies and mass reduction 3.4.5.1 Lower Rolling Resistance Tires Tire rolling resistance is the frictional loss associated mainly with the energy dissipated in the deformation of the tires under load and thus influences fuel economy and CC>2 emissions. Other tire design characteristics (e.g., materials, construction, and tread design) influence durability, traction (both wet and dry grip), vehicle handling, and ride comfort in addition to rolling resistance. A typical low rolling resistance tire's attributes could include: increased specified tire inflation pressure, material changes, and tire construction with less hysteresis, geometry changes (e.g., reduced aspect ratios), and reduction in sidewall and tread deflection. These changes would generally be accompanied with additional changes to vehicle suspension tuning and/or suspension design. The agencies expect that greater reductions in tire rolling resistance will be possible during the rulemaking timeframe than are currently available, as tire manufacturers continue to improve their products in order to meet increasing demand by auto OEMs for tires that contribute more to their vehicles' fuel efficiency. Thus, for this final rule, consistent with the proposal, the agencies considered two "levels" of lower rolling resistance tires. The first level ("LRR1") is defined as a 10 percent reduction in rolling resistance from a base tire, which was estimated to be a 1 to 2 percent effectiveness improvement in MYs 2012-2016 final rule. Based on the 2011 Ricardo study the agencies are now using 1.9 percent effectiveness improvement for LRRlfor all vehicle classes. LRR1 tires are widely available today, and appear to comprise a larger and larger portion of tire manufacturers' product lines as the technology continues to improve and mature. The second level ("LRR2") is defined as a 20 percent reduction in rolling resistance from a base tire, yielding an estimated 3.9 percent effectiveness improvement. In the CAFE model this results in a 2.0 percent incremental effectiveness increase from LRR1. LRR2 represents an additional level of rolling resistance improvement beyond what the agencies considered in the MYs 2012-2016 rulemaking analysis. NHTSA assumed that the increased traction requirements for braking and handling for performance vehicles could not be fully met with the ROLL2 designs in the MYs 2017- 2025 timeframe. For this reason the CAFE model did not apply ROLL2 to performance vehicle classifications. However, the agency did assume that tractions requirement for ROLL1 could be met in this timeframe and thus allowed ROLL1 to be applied to performance vehicle classifications in the MYs 2017-2025 timeframe. In the 2012-2016 light duty vehicle rule, the agencies estimated the incremental DMC at an increase of \$5 (2007\$) per vehicle. This included costs associated with five tires per 3-208 ------- Technologies Considered in the Agencies' Analysis vehicle, four primary and one spare tire. There is no learning applied to this technology due to the commodity based nature of this technology. Looking forward from 2016, the agencies continue to apply this same estimated DMC adjusted for 2010 dollars.bbb The agencies consider LRR1 to be fully learned out or "off the learning curve (i.e., the DMC does not change year-over-year) and have applied a low complexity ICM of 1.24 through 2018, and then 1.19 thereafter, due to the fact that this technology is already well established in the marketplace. To analyze the feasibility and cost for a second level of rolling resistance improvement, EPA, NHTSA, and CARB met with a number of the largest tire suppliers in the United States. The suppliers were generally optimistic about the ability to reduce tire rolling resistance in the future without the need to sacrifice traction (safety) or tread life (durability). Suppliers all generally stated that rolling resistance levels could be reduced by 20 percent relative to today's tires by MY 2017. As such, the agencies agreed, based on these discussions, to consider LRR2 as initially available for purposes of this analysis in MY 2017, but not widespread in the marketplace until MYs 2022-2023. In alignment with introduction of new technology, the agencies limited the phase-in schedule to 15 percent of a manufacturer's fleet starting in 2017, and did not allow complete application (100 percent of a manufacturer's fleet) until 2023. The agencies believe that this schedule aligns with the necessary efforts for production implementation, such as system and electronic system calibration and verification. LRR2 technology does not yet exist in the marketplace today, making cost estimation challenging without disclosing potentially confidential business information. To develop a transparent cost estimate, the agencies relied on LRR1 history, costs, market implementation, and information provided by the 2010 NAS report. The agencies assumed low rolling resistance technology ("LRR1") first entered the marketplace in the 1993 time frame with more widespread adoption being achieved in recent years, yielding approximately 15 years to maturity and widespread adoption. Then, using MY 2017 as the starting point for market entry for LRR2 and taking into account the advances in industry knowledge and an assumed increase in demand for improvements in this technology, the agencies interpolated DMC for LRR2 at \$10 (2010\$) per tire, or \$40 (\$2010) per vehicle. This estimate is generally fairly consistent with CBI suggestions by tire suppliers. The agencies have not included a cost for the spare tire because we believe manufacturers are not likely to include a LRR2 as a spare given the \$10 DMC. In some cases and when possible pending any state-level requirements, manufacturers have removed spare tires replacing them with tire repair kits to reduce both cost and weight associated with a spare tire.81 The agencies consider this estimated cost for LRR2 to be applicable in MY 2021. Further, the agencies consider LRR2 technology to be on the steep portion of the learning curve where costs would be reduced quickly in a relative bbb As noted elsewhere in this chapter, we show dollar values to the nearest dollar. However, dollars and cents are carried through each agency's respective analysis. Thus, while the cost for lower rolling resistance tires in the 2012-2016 final rule was shown as \$5, the specific value used in that rule was \$5.15 (2007\$) and is now \$5.40 (2010\$). We show \$5 for presentation simplicity. 3-209 ------- Technologies Considered in the Agencies' Analysis short period of time. The agencies have applied a low complexity ICM of 1.24 through 2024, and then 1.19 thereafter. The ICM timing for LRR2 is different from that for LRR1 because LRR2 is brand-new for this rulemaking and is not yet being implemented in the fleet. The resultant costs are shown in Table 3-121. Note that both LRR1 and LRR2 are incremental to the baseline system, so LRR2 is not incremental to LRR1. Table 3-121 Costs for Lower Rolling Resistance Tires Levels 1 & 2 (2010\$) Cost type DMC DMC 1C 1C TC TC Lower Rolling Resistance Tire Technology Level 1 Level 2 Level 1 Level 2 Level 1 Level 2 2017 \$5 \$63 \$1 \$10 \$7 \$73 2018 \$5 \$63 \$1 \$10 \$7 \$73 2019 \$5 \$51 \$1 \$10 \$6 \$60 2020 \$5 \$51 \$1 \$10 \$6 \$60 2021 \$5 \$40 \$1 \$10 \$6 \$50 2022 \$5 \$39 \$1 \$10 \$6 \$49 2023 \$5 \$38 \$1 \$10 \$6 \$48 2024 \$5 \$37 \$1 \$10 \$6 \$47 2025 \$5 \$36 \$1 \$8 \$6 \$44 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Note that both levels of lower rolling resistance tires are incremental to today's baseline tires. Given that the final standards cover such a long timeframe, the agencies also considered introducing a third level of rolling resistance reduction ("LRR3"), defined as a 30 percent reduction in rolling resistance. The agencies evaluated the potential of LRR3 entering the marketplace during this final rulemaking timeframe. Tire technologies that enable improvements of 10 and 20 percent have been in existence for many years. Achieving improvements up to 20 percent involves optimizing and integrating multiple technologies, with a primary contributor being the adoption of a silica tread technology.82 This approach was based on the use of a new silica along with a specific polymer and coupling agent combination. The use of the polymer, coupling agent and silica was known to reduce tire rolling resistance at the expense of tread wear, but new approach using novel silica reduced the tread wear tradeoff. Tire suppliers have indicated there are one or more innovations/inventions that they expect to occur in order to move the industry to the next quantum reduction of rolling resistance. However, based on the historical development and integration of tire technologies, there appears to be little evidence supporting improvements beyond LRR2 by 2025. Therefore, the agencies decided not to incorporate LRR3 at this time. The agencies sought comment on whether we should consider application of a 30 percent reduction from today's rolling resistance levels being available for mass production implementation by MY 2025 or sooner. The agencies also sought comment on the viability of this technology, maturity by MY 2025, as well as market introduction timing and the technological ways that this level of rolling resistance improvement will be achieved without any tradeoffs in terms of vehicle handling capability and tire life from what consumers expect today. Finally, the agencies sought cost information regarding the potential incorporation of LRR3 relative to today's costs as well as during the timeframe covered by this final rule. No comments were submitted on any of these topics. 3-210 ------- Technologies Considered in the Agencies' Analysis 3.4.5.2 Low Drag Brakes Low drag brakes reduce the sliding friction of disc brake pads on rotors when the brakes are not engaged because the brake pads are pulled away from the rotating disc either by mechanical or electric methods The 2012-2016 final rule and TAR estimated the effectiveness of low drag brakes to be as much as 1 percent. NHTSA and EPA have slightly revised the effectiveness down to 0.8 percent based on the 2011 Ricardo study and updated lumped-parameter model. In the 2012-2016 rule, the agencies estimated the DMC at \$57 (2007\$). This DMC becomes \$59 (2010\$) for this analysis after adjusting to 2010 dollars. The agencies consider low drag brake technology to be off the learning curve (i.e., the DMC does not change year- over-year) and have applied a low complexity ICM of 1.24 through 2018 then 1.19 thereafter. The resultant costs are shown in Table 3-122. Table 3-122 Costs for Low Drag Brakes (2010\$) Cost type DMC 1C TC 2017 \$59 \$14 \$74 2018 \$59 \$14 \$74 2019 \$59 \$11 \$71 2020 \$59 \$11 \$71 2021 \$59 \$11 \$71 2022 \$59 \$11 \$71 2023 \$59 \$11 \$71 2024 \$59 \$11 \$71 2025 \$59 \$11 \$71 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3.4.5.3 Front or Secondary Axle Disconnect for Four-Wheel Drive Systems Energy is required to continually drive the front, or secondary, axle in a four-wheel drive system even when the system is not required during most operating conditions. This energy loss directly results in increased fuel consumption and CC>2 emissions. Many part-time four-wheel drive systems use some type of front axle disconnect to provide shift-on-the-fly capabilities. The front axle disconnect is normally part of the front differential assembly. As part of a shift-on-the-fly four-wheel drive system, the front axle disconnect serves two basic purposes. First, in two-wheel drive mode, it disengages the front axle from the front driveline so the front wheels do not turn the front driveline at road speed, saving wear and tear. Second, when shifting from two- to four-wheel drive "on the fly" (while moving), the front axle disconnect couples the front axle to the front differential side gear only when the transfer case's synchronizing mechanism has spun the front driveshaft up to the same speed as the rear driveshaft. Four-wheel drive systems that have a front axle disconnect typically do not have either manual- or automatic-locking hubs. To isolate the front wheels from the rest of the front driveline, front axle disconnects use a sliding sleeve to connect or disconnect an axle shaft from the front differential side gear. NHTSA and EPA are not aware of any manufacturer offering this technology in the U.S. today on unibody frame vehicles; however, it is possible this technology could be introduced by manufacturers within the MYs 2017- 2025 time period. 3-211 ------- Technologies Considered in the Agencies' Analysis The 2012-2016 final rule estimated an effectiveness improvement of 1.0 to 1.5 percent for axle disconnect. Based on the 2011 Ricardo report, NHTSA and EPA refined this range to 1.2 to 1.4 percent. In the 2012-2016 rule, the agencies estimated the DMC at \$78 (2007\$) which was considered applicable to the 2015MY. This DMC becomes \$82 (2010\$) for this analysis after adjusting to 2010 dollars. The agencies consider secondary axle disconnect technology to be on the flat portion of the learning curve and have applied a low complexity ICM of 1.24 through 2018 then 1.19 thereafter. The resultant costs are shown in Table 3-123. Table 3-123 Costs for Secondary Axle Disconnect (2010\$) Cost type DMC 1C TC 2017 \$78 \$20 \$98 2018 \$76 \$20 \$96 2019 \$75 \$16 \$91 2020 \$73 \$16 \$89 2021 \$72 \$16 \$88 2022 \$70 \$16 \$86 2023 \$69 \$16 \$85 2024 \$68 \$16 \$83 2025 \$66 \$16 \$82 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost 3.4.5.4 Aerodynamic Drag Reduction Many factors affect a vehicle's aerodynamic drag and the resulting power required to move it through the air. The overall drag force can be simplified as proportional to vehicle's frontal area, vehicle's drag coefficient, air density and the second order of vehicle's velocity. Therefore reducing vehicle's frontal area and drag coefficient can reduce fuel consumption and CC>2 emissions. Although frontal areas tend to be relatively similar within a vehicle class (mostly due to market-competitive size requirements), significant variations in drag coefficient can be observed. Significant changes to a vehicle's aerodynamic performance may need to be implemented during a redesign (e.g., changes in vehicle shape). However, shorter-term aerodynamic reductions, with a somewhat lower effectiveness, may be achieved through the use of revised exterior components (typically at a model refresh in mid-cycle) and add-on devices that are currently being applied. The latter list would include revised front and rear fascias, modified front air dams and rear valances, addition of rear deck lips and underbody panels, and lower aerodynamic drag exterior mirrors. The 2012-2016 final rule estimated that a fleet average of 10 to 20 percent total aerodynamic drag reduction is attainable which equates to incremental reductions in fuel consumption and CC>2 emissions of 2 to 3 percent for both cars and trucks. These numbers are generally supported by the Ricardo study and public technical literature and therefore NHTSA and EPA are retaining these estimates, as confirmed by joint review, for the purposes of this final rule, consistent with the proposal. Importantly, the effectiveness values presented here represent two-cycle effectiveness. Because active aerodynamic technologies (i.e., aero level 2) provide additional off-cycle benefits, both agencies apply an off-cycle credit value to the technology. Off-cycle credits are discussed in Chapter 5 of this Joint TSD. 3-212 ------- Technologies Considered in the Agencies' Analysis For this final rule, consistent with the proposal, the agencies considered two levels of aero improvements. The first level is that discussed in the 2012-2016 final rule and the 2010 TAR and includes such body features as air dams, tire spats, and perhaps one underbody panel. In the 2012-2016 rule, the agencies estimated the DMC of aero-level 1 at \$39 (2007\$). This DMC becomes \$41 (2010\$) for this analysis, applicable in the 2015MY. The agencies consider aero-level 1 technology to be on the flat portion of the learning curve and have applied a low complexity ICM of 1.24 through 2018 then 1.19 thereafter. The second level of aero—level 2 which includes such body features as active grille shutters000, rear visors, larger under body panels or low-profile roof racks —was discussed in the 2010 TAR where the agencies estimated the DMC at \$120 (2008\$) incremental to the baseline vehicle. The agencies inadvertently used that cost as inclusive of aero-level 1 technologies when it should have been incremental to aero-1 technologies. As a result, the agencies now consider the TAR cost to more appropriately be incremental to aero-level 1 with a DMC for this analysis of \$123 (2010\$). The agencies consider this cost to be applicable in the 2015MY. Further, the agencies consider aero-level 2 technology to be on the flat portion of the learning curve. The agencies have applied a medium complexity ICM of 1.39 through 2024 then 1.29 thereafter. The timing of the aero-level 2 ICMs is different than that for the level 1 technology because the level 2 technology is newer and not yet being implemented in the fleet. The resultant costs are shown in Table 3-124. Table 3-124 Costs for Aerodynamic Drag Improvements - Levels 1 & 2 (2010\$) Cost type DMC DMC 1C 1C TC TC TC Aero Technology Level 1 Level 2 Level 1 Level 2 Level 1 Level 2 Level 2 Incremental to Baseline Aero-level 1 Baseline Aero-level 1 Baseline Aero-level 1 Baseline 2017 \$39 \$117 \$10 \$47 \$49 \$164 \$213 2018 \$38 \$115 \$10 \$47 \$48 \$162 \$210 2019 \$37 \$112 \$8 \$47 \$45 \$160 \$205 2020 \$37 \$110 \$8 \$47 \$45 \$157 \$202 2021 \$36 \$108 \$8 \$47 \$44 \$155 \$199 2022 \$35 \$106 \$8 \$47 \$43 \$153 \$196 2023 \$35 \$104 \$8 \$47 \$42 \$150 \$193 2024 \$34 \$102 \$8 \$47 \$42 \$148 \$190 2025 \$33 \$100 \$8 \$35 \$41 \$135 \$176 DMC=Direct manufacturing cost; IC=Indirect cost; TC=Total cost Because a large percent of the performance vehicles already have some level of aerodynamic treatments, when running the CAFE model NHTSA only applies level 1 of aerodynamic treatment to these vehicles. Also for specific vehicles, such as Toyota Prius, which already have extensive aerodynamic treatment, the level of the aerodynamic that could be further applied by NHTSA in the CAFE model is limited in the market input file. ! For details on how active aerodynamics are considered for off-cycle credits, see TSD Chapter 5.2.2. 3-213 ------- Technologies Considered in the Agencies' Analysis 3.4.5.5 Mass Reduction From 1987-2011, there has been a generally increasing trend in the weight of the light duty vehicle fleet as shown in Figure 3-26 from EPA's Fuel Economy Trends Report83. A number of factors have contributed to this weight increase, including the choices of manufacturers and consumers to build and purchase larger vehicles, including heavier trucks, SUVs, and CUVs. Also contributing to this weight increase has been an increase in vehicle content including: safety features (air bags, antilock brakes, energy absorbent and intrusion resistant vehicle structures, etc.), noise reduction (additional damping material), added comfort and convenience features (air conditioning, power locks and windows), luxury features (infotainment systems, powered seats), etc. 5000 3000 1975 1980 1985 1990 1995 Model year 2000 2005 2010 Figure 3-26 Light duty fleet weight trends: 1975-2011 Despite this increase in weight, the average acceleration of vehicles has grown steadily faster without any marked or consistent reduction in fuel economy since 1987, as shown in Figure 3-27. This combination of increased vehicle performance, stable fuel economy, and increased vehicle weight has been partially enabled by the development and adoption of more efficient technologies, especially in engines and transmissions. The impressive improvements in powertrain efficiency during this period have offset increases in energy consumption that result from improvements in weight carrying, towing and volume capacities, safety, consumer features, vehicle refinement, and acceleration performance. 3-214 ------- Technologies Considered in the Agencies' Analysis Adjusted fuel economy (mpg) 0/1 ic 00 - on j 1 S - i 1 0 - ft - • -*-• ^. *t^«- •»-*" 7i*1"~*~*%»-«, fuel economy -,,/ ***~ ? N^^ ---**....- / ^^ "*~*~^s*~^ acceleration ' ^ 15 ** 1 /i pfl €\ no « 1975 1980 1985 1990 1995 2000 2005 2010 Model year Figure 3-27 Light duty fleet trends for acceleration and fuel economy: 1975-2011 84 Motoi 1950-2C 60 40 20 1! * vehicle crash deaths per billion miles traveled 309 ^ Vx^v ^-v. S50 55 60 65 70 75 80 85 90 ************»•» 11. 3 per billion 95 2000 05 Figure 3-28 U.S. Vehicle Fatality Rates for the past 60 years Vehicle mass reduction (also referred to as "down-weighting" or 'light-weighting"), reduces the energy needed to overcome inertial forces, thus yielding lower fuel consumption and GHG emissions. While keeping everything else constant, a lighter vehicle will require less energy to operate than a heavier vehicle. Mass reduction can be achieved through a number of approaches described below, even while maintaining vehicle size. Alternatively, mass reduction can also be achieved by vehicle "downsizing" which involves reducing vehicle exterior dimensions, such as shifting from a midsize vehicle to a compact vehicle. Consistent with the proposal, the agencies did not analyze downsizing as a mass reduction strategy in this analysis for the final rule. In part, this is because a manufacturer's ability to downsize its vehicles is constrained by consumer preferences (such as for interior passenger or cargo volume), which are in turn influenced by many factors that are difficult to predict in 3-215 ------- Technologies Considered in the Agencies' Analysis the future, such as the consumer's utility needs, fuel prices, economic conditions, etc. Also, the final CAFE and GHG emission standards are based on vehicle footprint (the area bounded by where the four tires contract the ground), and assign higher fuel economy targets (and lower CC>2 emission targets) for vehicles with smaller footprints and lower fuel economy targets (and higher CO2 emission targets) for vehicles with larger footprints. As discussed in Chapter 2 of the joint TSD, the agencies believe the shape of the footprint-based target curves will not create incentives for manufacturers to either upsize or downsize their vehicles. Based on these considerations, the agencies are assuming that manufacturers will favor mass reduction through material substitution, design optimization, and adopting other advanced manufacturing technologies rather than compromising a vehicle's attributes and functionality, such as occupant or cargo space, vehicle safety, comfort, acceleration, etc. Consequently, the compliance paths the agencies have investigated for the promulgated standards do not include downsizing. Mass reduction has an important relationship with vehicle powertrain selection and sizing. Vehicle powertrain selection depends on an OEM's product strategy, and may include a variety of options such as naturally aspirated engines, boosted and downsized gasoline engines, diesel engines, or vehicle electrification (P/H/EV) Regardless of the strategy selected, vehicle mass reduction for non-powertrain systems is an important enabler to further reduce vehicle fuel consumption and reduce the size of the powertrain system. The term "glider" refers to a complete vehicle minus the powertrain. Figure 3-29 illustrates the mass o/r breakdown by system for a typical vehicle . The non-powertrain systems normally account for 75 percent of vehicle weight. The agencies have accounted for some of the costs of engine mass reduction when applying engine downsizing technologies. The agencies have also accounted for the amount of mass change due to the application of hybrid and electrification technologies in the vehicle electrification sections. Therefore, this section focuses on both the mass reduction of the glider as well as mass reduction technologies that are specifically targeted at reducing the weight of the powertrain.ddd rather than on mass reduction resulting from powertrain efficiency improvements. An example of a mass reduction technology for the powertrain that is not related to powertrain efficiency improvement is material substitution, such as changing the engine block from cast iron to aluminum or changing the size of the fuel tank.). Mass reduction is calculated for both the glider and the vehicle including powertrain in the studies sponsored by the agencies as shown later in this section. ddd Rather than on mass reduction resulting from powertrain efficiency improvements, such as in the case of adding a turbocharger to a downsized engine. 3-216 ------- Technologies Considered in the Agencies' Analysis Approximate vehicle mass breakdown" Misc.; Closures, ™ fenders; 8K B°dY; '*• 23-28% Suspension/chassis; 22-27)4 System Body-in-white Powertrain Chassis Interior Closures Miscellaneous Major components in system Passenger compartment frame, cross and side beams, roof structure, front-end structure, underbody floor structure, panels Engine, transmission, exhaust system, fuel tank Chassis, suspension, tires, wheels, steering, brakes Seats, instrument panel, insulation, trim, airbags Front and rear doors, hood, lift gate Electrical, lighting, thermal, windows, glazing "Based on Stodolsky et al, 1995a; Bjelkengren, 2008; Lotus Engineering, 2010; the actual system definitions and system component inclusion can vary, and percentage weight breakdown can vary substantially by vehicle Figure 3-29 Vehicle system mass approximation A vehicle can be divided into 6 major systems, which are shown in Figure 3-29. Mass reduction can potentially be applied to any of a vehicle's subsystems, including the engine, exhaust system, transmission, chassis, suspension, brakes, body, closure panels, glazing, seats and other interior components, engine cooling systems, and HVAC systems. While manufacturers may reduce the mass of some individual components during a vehicle refresh, they generally undertake larger amounts of mass reduction systematically and more broadly across all vehicle systems when redesigning a vehicle. In the redesign process, OEMs normally set weight targets by benchmarking other vehicles in the same segment and projecting weight trends into the future, and then identifying targets for all components and subsystems that support achieving the target. The agencies believe this holistic approach, which takes into consideration all secondary mass savings, is likely the most effective way for OEMs to achieve large amounts of mass reduction. During a vehicle redesign where mass reduction is a strategic vehicle program goal, OEMs can consider modular systems design, secondary mass effects, multi-material concepts, and new manufacturing processes to help optimize the design. There are several studies in the public domain that illustrate the potential for these approaches to achieve significant amounts of mass reduction, although it is important to also recognize that the studies use some assumptions that do not account for some of the considerations that are important to manufacturers. One example is the need to share some components across platforms to manage cost and part complexity for assembly and service, which limits the ability to optimize the amount of mass reduction on every vehicle component. Care must also be taken in any study to assure that vehicle functionality and performance, such as stiffness, NVH, safety and vehicle dynamics, continue to meet manufacturer objectives and consumer demands. It is important for design studies to use tools such as simulation modeling to assess the design's ability to meet functionality and performance targets. In this rulemaking, the agencies have targeted to preserve vehicle function and performance in their analysis of mass reduction. 3-217 ------- Technologies Considered in the Agencies' Analysis An example of this approach is illustrated in Figure 3-30, which summarizes the results of the 2010 phase I Lotus Engineering mass reduction study of a Toyota Venza. Mass-reduction features, findings Redesign conventional mid-size vehicle for mass optimization, with two redesign architectures Low Development vehicle technology with industry-leading manufacturing techniques that were deemed feasible for 2014 (for model year 2017 production) for assembly at existing facilities High Development vehicle technology, with modifications to conventional joining and assembly processes that were deemed feasible for 2017 (for model year 2020) production Extensive use of material substitution with high-strength steel, advanced high-strength steel, aluminum, magnesium, plastics and composites throughout vehicles Conservative use of emerging design and parts integration concepts to minimize technical risk Using synergistic total vehicle substantial mass reduction opportunities found at minimized piece costs The Low Development vehicle was found to have likely piece cost reductions, whereas the High Development vehicle had nominal estimated cost increase of 3% (with potential for cost reduction) Mass-reduction impact • Body structure reduction for Low Development Vehicle: 55 Ib (6.6%) • Body structure reduction for High Development Vehicle: 356 Ib (42%) • Overall glider reduction for Low Development Vehicle: 538 Ib (19%) • Overall glider reduction for High Development Vehicle: 1096 Ib (39%) • Overall vehicle reduction for Low Development Vehicle (with hybrid powertrain): 657 Ib (17.6%) » Overall vehicle reduction for High Development Vehicle (with hybrid powertrain): 1209 Ib (32%) Status Engineering design study conducted by Lotus Engineering First phase of project, development of two mass-reduced vehicle designs completed in April 2010 Second phase to test structural integrity, impact load paths, crash worthiness to validate the vehicle designs. Source Lotus Engineering, Inc. 2010. An Assessment of Mass Reduction Opportunities for a 2017-2020 Model Year Vehicle Program Figure 3-30 Example of a holistic vehicle redesign study from Lotus Engineering8 Mass reduction can be considered in terms of the "percent by which the redesigned vehicle is lighter than the previous version," recognizing that the value likely represents both "primary" mass reduction (that which the manufacturer set out to make lighter), and "secondary" mass reduction (from ancillary systems and components that can now be lighter due to the primary mass reductions). oo As summarized by NAS in its 2011 report, there are two key strategies for primary mass reduction: 1) changing the design to use less material or 2) substituting lighter materials for heavier materials. The first key strategy of using less material compared to the baseline component can be achieved by optimizing the design and structure of the component, system or vehicle structure. For example, a number of "body on frame" vehicles have been redesigned with a lighter "unibody" construction, eliminating components, reducing the weight of the body structure, and resulting in significant reductions in overall mass and related costs. The unibody design currently dominates the passenger car segment and has increased penetration into what used to be mostly body-on-frame vehicles, such as SUVs. 5-218 ------- Technologies Considered in the Agencies' Analysis This technique was used in the 2011 Ford Explorer redesign, which also employed the extensive use of high strength steels.89 Figure 3-31 depicts body-on-frame and unibody designs for two sport utility vehicles. Unibody Figure 3-31 Illustration of Body-on-Frame (BoF) and Unibody vehicle construction To further reduce mass inefficiencies in vehicle design, vehicle manufacturers are using continually-improving Computer Aided Engineering (CAE) tools. For example, the Future Steel Vehicle (FSV) project90 sponsored by WorldAutoSteel used three levels of optimization: topology optimization, low fidelity 3G (Geometry Grade and Gauge) optimization, and sub-system optimization, to achieve 30 percent mass reduction in the body structure of a vehicle with a mild steel unibody structure (see Figure 3-32). Designs similar to those proposed in the FSV project have been applied in production vehicles, such as the B- pillarof2010FordFocus.91 3-219 ------- Technologies Considered in the Agencies' Analysis 2.4 T4: Body Structure Sub-System Optimisation The fmal design attained from the LF3G optimisation was used as the basts for the sub-system optimisation as well as the source of the boundary conditions Load path mapping was conducted on the mode) to identify the most dominant structural sub-systems in the body structure. Load path mapping considers the dominant loads m the structural sub-systems for each of the load cases as shown m Figure 2-7. x figurt 2-7: T4 Load Path Mapping - Major Load Path Components Based on load path mapping, seven structural sub-systems (Figure 2-8) were selected for further optimisation using the spectrum of FSVs potential manufacturing technologies •Shot Gun Figure 2-9. Structural Sub-Sytttmt Selected FutureSteelVehicle ^ WorldAutoStecl Figure 3-32 Example of vehicle body load path mapping for mass optimization Vehicle manufacturers have long used these continually-improving CAE tools to optimize vehicle designs. But because any design must meet component and system 3-220 ------- Technologies Considered in the Agencies' Analysis functionality and manufacturability targets, there are practical limitations to the amount of additional mass reduction that can be achieved through optimization. For example, an optimization program would need to account for safety, stiffness, NVH, manufacturing, and other requirements to assure the design is suitable for its intended function and for mass production. Additionally, ultimate optimization of vehicle design for mass reduction may be limited by an OEM's use of shared components and common platform for multiple vehicle models. While optimization may concentrate on the vehicle that has the largest production volume for a platform, designs must also support the most demanding functional requirements of all of the vehicles that share that platform, or those functional requirements will not be met. In addition, the engineering resources and capital for tooling and equipment that would be needed to optimize every vehicle component at each redesign affects the ability to fully optimize a new vehicle to achieve all of the theoretically possible secondary mass reduction. Therefore, some level of mass inefficiency will inherently exist on many or all of the vehicles that share a platform. The agencies sought comment and information in the NPRM on the degree to which shared vehicle components and architectures affect the feasible amount of mass reduction and the cost for mass reduction relative to what could be achieved if mass reduction was optimized for a single vehicle design. Volkswagen confirmed in its comments that with platform sharing, "a weight reduction technology which may be acceptable in terms of price or performance for one model may disrupt the economics or utility of another."92 Using less material can also be achieved through improving the manufacturing process, such as by using improved joining technologies and parts consolidation. This method is often used in combination with applying new materials. For example, more precise manufacturing techniques such as laser welding may reduce the flange size necessary for welding, and thus marginally decrease the mass of an assembly. Also, when complex assemblies are constructed from fewer pieces, the mass of the assembly tends to be lower. However, while synergies in mass reduction certainly exist, and while certain technologies can enable one another (e.g., parts consolidation and molding of advanced composites), others may be incompatible (e.g., laser welding and magnesium casting). The second key strategy to reduce mass of an assembly or component involves the substitution of lower density and/or higher strength materials. Table 3-125 shows material usage typical of contemporary high-volume vehicles. Material substitution includes replacing materials, such as mild steel, with higher-strength and advanced steels, aluminum, magnesium, and composite materials. The substitution of advanced high strength steel (AHSS) for mild steel can reduce the mass of a strength-critical part because the gauge of the AHSS components can be reduced, despite the fact that the densities of the materials are not significantly different. Aluminum has also been used over the years in a variety of components, such as vehicle closures, suspension parts, engine cradles, etc. Aluminum has one third the density of steel and therefore can provide a notable amount of mass reduction. Changing parts from steel to aluminum generally requires part redesign, and extra material may have to be added for strength or durability. Aluminum also has a shorter fatigue life than steel, and therefore the alloy selected and the application must be carefully considered. Magnesium can provide additional mass reduction as it has lower density than aluminum. It has been used for instrument panel cross-car beams by several OEMs for a number of years. It has also been used in an engine block produced by BMW for several years. Its brittle 3-221 ------- Technologies Considered in the Agencies' Analysis nature must be considered, however, when selecting the alloy and the application within the vehicle. Table 3-125 Distribution of Material in Typical Contemporary Vehicles (e.g., Toyota Camry or Chevrolet Malibu)93 Approximate Concent in Cars Material Comments Today, by Weight (percent) Iron and mild steci High-strength steel Aluminum Plastic Other (magnesium, tuanRim, rubber, eic.) Under 480 Mpa 55 > 4HO Mpa (in body structure) 15 No aluminum closure panels; aluminum engine block and head and wheels 10 Miscellaneous parts, mostly interior trim, tight lenses* facia, instrument panel 10 Miscellaneous, parts 10 Automobiles also utilize a wide range of plastic types, including polypropylenes, polyesters, and vinyl esters. These materials are utilized in hatches, roofs, interior panels, instrument panels, and hundreds of other parts. Although primarily used in nonstructural vehicle components, plastics have continued to make in-roads in bumper systems and in composite beam applications, and some studies have found potential to supplant structural beams and frame components. Lighter plastics have also been developed by the industry, and the application of these materials has been increasing. Included in the category of plastics are composites like glass fiber and carbon fiber reinforced polymers. While these more costly advanced materials have primarily been used in a limited number of low production volume vehicle applications, some manufacturers are considering these composites for broader use. While these materials currently have the potential to be applied to components with little or no exposure to impact pulses, the advanced microstructure and limited industry experience may make these longer-term solutions. For example, advanced composite materials (such as carbon fiber-reinforced plastic), depending on the specific fiber, matrix, reinforcement architecture, and processing method, can be subject to dozens of competing damage and failure mechanisms that may complicate a manufacturer's ability to ensure equivalent levels of durability and crashworthiness. As the industry gains experience with these materials, these concerns will inevitably diminish, but may remain relevant during the timeframe of this final rulemaking. In practice, material substitution tends to be quite specific to the manufacturer and situation. Some materials work better than others for particular vehicle components, and a manufacturer may invest more heavily in adjusting to a particular type of advanced material, thus complicating its ability to consider others. The agencies recognize that like any type of mass reduction, material substitution has to be conducted not only with consideration to maintaining equivalent component strength, but also to maintaining all the other attributes of that component, system or vehicle, such as crashworthiness, durability, and NVH. If vehicle mass is reduced sufficiently through application of the two primary strategies of using less material and material substitution described above, secondary mass reduction options may become available. Secondary mass reduction is enabled when the load requirements of a component are reduced as a result of primary mass reduction. If the 3-222 ------- Technologies Considered in the Agencies' Analysis primary mass reduction reaches a sufficient level, a manufacturer may use a smaller, lighter, and potentially more efficient powertrain while maintaining vehicle acceleration performance. If a powertrain is downsized, approximately half of the mass reduction may be attributed to the reduced torque requirement which results from the lower vehicle mass. The lower torque requirement enables a reduction in engine displacement, changes to transmission including the torque converter and gear ratios, and changes to final drive gear ratio. The reduced powertrain torque enables the downsizing and/or mass reduction of powertrain components and accompanying reduced rotating mass (e.g., for transmission, driveshafts/halfshafts, wheels, and tires) without sacrificing powertrain durability. Likewise, the combined mass reductions of the engine, drivetrain, and body in turn reduce stresses on the suspension components, steering components, wheels, tires, and brakes, which can allow further reductions in the mass of these subsystems. Reducing the unsprung masses such as the brakes, control arms, wheels, and tires further reduce stresses in the suspension mounting points, which will allow for further optimization and potential mass reduction. Secondary mass reduction can occur for each kg of primary mass reduction, when all subsystems are redesigned to take the initial primary mass reduction into account. In the MYs 2012-2016 rulemaking analysis, the agencies assumed that 1 kg of primary mass reduction could enable up to 1.25 kg of secondary mass reduction. In the two most recent mass reduction projects by EPA and NHTSA, every 1 kg of primary mass reduction enabled 0.7 kg of secondary mass reduction. We note that these estimates may not be applicable in all real- world instances of mass reduction, and that the literature indicates that the amount of secondary mass reduction potentially available varies significantly from an additional 0.5 kg to 1.25 kg per 1 kg of primary mass reduction, depending on assumptions such as which components or systems primary mass reduction is applied to, and whether the powertrain is available for downsizing. 94>95>96 The amount of secondary mass reduction is also affected by the degree of component sharing that occurs among a manufacturer's models. Component sharing is used by manufacturers to achieve production economies of scale that affect cost and the number of unique parts that must be managed in production and for service. In addition, the engineering resources and capital for tooling and equipment that would be needed to optimize every vehicle component at each redesign affects the ability to fully optimize a new vehicle to achieve all of the theoretically possible secondary mass reduction. While there is agreement in the literature that primary mass reduction can enable secondary mass reduction, the agencies recognize that care must be taken when reviewing reports on mass reduction methods and practices to ascertain the manner and extent to which compounding effects have been considered. All manufacturers are using some or all of these methods to reduce mass in the vehicles they are producing today, and the agencies expect that the industry will continue to learn and improve the application of these techniques for more vehicles during the rulemaking timeframe. We consider mass reduction in net percentage terms in our analysis not only because effectively determining specific appropriate mass reduction methods for each vehicle in the baseline fleet is a large task beyond the scope of this rulemaking, but also because we recognize that even as manufacturers reduce mass to make vehicles more efficient, they may also be adding mass in the form of increased vehicle features and safety content in response to market forces and other governmental regulations. For these reasons, when the agencies 3-223 ------- Technologies Considered in the Agencies' Analysis discuss the amount of mass reduction that we are assuming is feasible for purposes of our analysis, we are implicitly balancing both the considerable opportunities that we believe exist for mass reduction in the future, and the reality that vehicle manufacturing is complex and that mass reduction methods must be applied thoughtfully and judiciously as safety and content demands on vehicles continue to increase over time. Despite our considerable discussion of the topic, the agencies' application of mass reduction in our analysis is fairly simplified. As applied in our models, the percentage reduction for a given vehicle that is assumed for a given year is an abstraction of all the specific mass reduction methods described above. How much mass reduction do the agencies believe is feasible in the rulemaking timeframe? Feasibility, if narrowly defined as the ability to reduce mass without any constraints, is nearly unbounded. However, in practice, the feasible amount of mass reduction is affected by other considerations. Cost effectiveness is one of those constraints and is discussed further below in the mass reduction cost section. In the analysis for the current rulemaking for MYs 2017-2025, the agencies reviewed a number of public reports and accompanying data, as well as confidential information from manufacturers, and believe that mass reduction of up to 20 percent from a MY 2008 baseline vehicle can be achieved in a cost effective manner using technologies currently in production. More detail on studies reviewed by the agencies and additional studies currently in progress by the agencies is located below in Table 3-9 and in the paragraphs below under the question " What additional studies are the agencies conducting to inform our estimates of mass reduction amounts, cost, and effectiveness?" From a general planning perspective, nearly all automakers have made some public statement regarding vehicle mass reduction being a core part of the overall technology strategy that they will utilize to achieve future fuel economy and CC>2 emission standards. Estimates from Ducker Worldwide indicate that the automobile industry will see an annual increase in AHSS of about 10% through 202097. Ford has stated that it intends to reduce the weight of its vehicles by 250-750 Ib QO per model from 2011 to 2020 . For context, the midpoint of that range of reductions would correspond to a 12% reduction from the current Ford new light duty vehicle sales fleet. Mazda has released a statement about achieving a 220-lb reduction per vehicle by 201699. This is equivalent to about a 6% reduction for the company's current fleet. - Land Rover executives have stated that the company remains committed to a goal of reducing curb weights of its SUVs by as much as 500 kilograms over the next lOyears10^ In its comment to the NPRM, Volkswagen stated that they expect to reduce the mass of their vehicles by 7-10% on average during the period of this regulation. Several reports focusing on the OEM's approaches for light weighting are summarized in the University of California Davis study as shown in Table 3-126 101. 3-224 ------- Technologies Considered in the Agencies' Analysis Table 3-126 Automaker industry statements regarding plans for vehicle mass-reduction technology Affiliation General Motors Ford Nissan BMW Volkswagen Fiat Volkswagen BMW BMW Quote "We use a lot of aluminum today - about 300 pounds per vehicle - and are likely to use more lightweight materials in the future" "The use of advanced materials such as magnesium, aluminum and ultra high- strength boron steel offers automakers structural strength at a reduced weight to help improve fuel economy and meet safety and durability requirements" "We are working to reduce the thickness of steel sheet by enhancing the stregnth, expanding the use of aluminum and other lightweight materials, and reducing vehicle weight by rationalizing vehicle body structure" "Lightweight construction is a core aspect for sustainable mobility improving both fuel consumption and CO2 emissions, two key elements of our Efficient Dynamics strategy ... we will be able to produce carbon fiber components in large volumes at competitive costs for the first time. This is particularly relevant for electric -powered vehicles." "Material design and manufacturing technologies remain key technologies in vehicle development. Only integrated approaches that work on these three key technologies will be successful in the future. In addition to the development of metals and light metals, the research on fibre -reinforced plastics will play a major role." "A reduction of fuel consumption attains big importance because of the possible economical savings. In order to achieve that, different ways are followed: alternative engine concepts (for example electric engines instead of combustion ones) or weight reduction of the vehicle structure. Using lightweight materials and different joining techniques hleps to reach this aim" "Lightweight design is a key measure for reducing vehicle fuel consumption along with powertrain efficiency, aerodynamics and electrical power management" "A dynamic vehicle with a low fuel consumption finally demands a stiff body with a low weight. To achieve the initially mentioned targets, it is therefore necessary to design a body which offers good stiffness values and a high level of passive safety at a low weight." "Light weight design can be achieved by engineering light weight, manufacutring light weight and material light weight design." Source Keith, 2010 BMW and SQL, 2010 Goede et al, 2009 Nunez, 2009 Goede et al, 2009 Nunez, 2009 Krinke, 2009 Prestorf, 2009 Prestorf, 2009 Although the focus on mass reduction by manufacturers is widespread, the agencies believe the practical limits of mass reduction will be different for each vehicle model as each model starts with a different mix of conventional and advanced materials, components, and features intended to meet the function and price of a particular market segment. A vehicle that already has a significant fraction of advanced high strength steel (AHSS) or any other advanced material in its structure, for example, will not have the opportunity to realize the same percentage of mass reduction as a vehicle of more traditional construction. Given the myriad methods of achieving mass reduction, and the difficulty in obtaining data, accounting for the current level of mass reduction technology for every model in production in a baseline model year would be an impractical task. However, the agencies believe that reducing vehicle weight to reduce fuel consumption has a continuum of solutions and the technologies employed will have levels of effectiveness and feasibility that will vary by manufacturers and by vehicle. In estimating the amount of mass reduction for this analysis, the agencies also consider fleet safety effects for mass reduction. See Section II. G of the preamble for a detailed discussion of the safety considerations in establishing CAFE and GHG standards. In the CAFE and OMEGA analyses, the agencies considered several levels of mass reduction applicable to all of the models in each subclass, as discussed below. 3-225 ------- Technologies Considered in the Agencies' Analysis Based on the many aspects of mass reduction (i.e.., feasibility, cost and safety), for the final rule, consistent with the proposal, the agencies believe that mass reduction of up to 20 percent is feasible on light trucks, CUVs and minivans, but that less mass reduction should be implemented on other vehicle types to avoid increases in societal fatalities. While the agencies continue to examine mass reduction, we remain alert to safety considerations and seek to ensure that any CAFE and CC>2 standards can be achieved in a safety-neutral or improved manner. In the CAFE model, NHTSA applied amounts of mass reduction shown in Table 3-127, which was based on the ability to achieve overall fleet fatality estimates of close to zero. The results are described in Preamble Section II.G and Chapter V of NHTSA's RIA. The amount of mass reduction applied in EPA's OMEGA model follows the safety neutral analysis is described in Section II.G of the Preamble with a variety of tables in EPA's RIA (Chapter 3.8.2). Table 3-127 MAXIMUM MASS REDUCTION AMOUNT APPLIED IN CAFE MODEL Absolute % MR1* MR2 MRS MR4 MRS Subcompact and Subcompact Perf. PC Compact and Compact Perf. PC Midsize PC and Midsize Perf. PC Large PC and Large Perf. PC Minivan LT Small, Midsize and Large LT 0.0% 0.0% 1.5% 1.5% 1.5% 1.5% 0.0% 0.0% 3.5% 7.5% 7.5% 7.5% 0.0% 0.0% 0.0% 10.0% 10.0% 10.0% 0.0% 0.0% 0.0% 0.0% 15.0% 15.0% 0.0% 0.0% 0.0% 0.0% 20.0% 20.0% Notes: *MR1-MR5: different levels of mass reduction used in CAFE model The amounts of mass reduction shown in Table 3-127 are for conventional vehicles. The agencies assume that vehicles with hybrid and electric powertrain are heavier than conventional vehicles because of the mass of battery systems. In comparing anecdotal data for HEVs, EPA and NHTSA assume a slight weight increase of 4-5% for HEVs as compared to baseline non-hybridized vehicles. The added weight of the Li-ion pack, motor and other electric hardware were offset partially by the reduced size of the base engine as stated in TSD section 3.4.3.8. We believe that this assumption accurately reflects real-world HEV, PHEV and EV construction. As an example, for a Subcompact PHEV with 20 mile range operating on electricity, the agencies assume that to achieve no change in total vehicle mass, it would be necessary to reduce the mass of the glider by 6 percent because of the additional weight of the electrification system. The mass reduction for P/H/EVs can be found section 3.3.3.9 in the joint TSD, and in EPA's RIA Chapter 1 and Chapter V, section E.3.h.4, of NHTSA's FRIA. How much do the agencies estimate mass reduction will cost in the rulemaking timeframe? Automakers are currently utilizing various mass reduction techniques across the light- duty vehicle fleet, and will continue to use and in some cases expand these approaches for the 3-226 ------- Technologies Considered in the Agencies' Analysis 2017 to 2025 time frame. These approaches may include optimized design, geometry, part consolidations, and materials substitution. Unlike the other technologies described in this chapter, mass reduction is potentially more complex in that we cannot define it as a single piece of equipment or hardware change to implement the technological improvement. Mass reduction, depending upon the level of reduction targeted, has the potential to impact nearly every system on the vehicle. Because of this complexity, there are unique challenges to estimating the cost for mass reduction and for demonstrating the feasibility of reducing vehicle mass by a given amount. This section describes the cost estimates used for the agencies' analysis. In the analysis for the MYs 2012-2016 rulemaking, the agencies assumed a constant cost for mass reduction of \$1.32 for each pound reduced up to a mass reduction level of 10 percent (or \$1.48/lb using an ICM factor of 1.1 for a low-complexity technology). The \$1.32/lb estimate was based on averaging three studies: the 2002 NAS Report, a 2008 study by Sierra Research, and a 2007 study by MIT researchers.eee Since the MYs 2012-2016 final rule, the agencies have given further consideration to the cost of mass reduction, and now believe that a cost that varies with the level of mass reduction provides a better estimate. The agencies believe that as the vehicle fleet progresses from lower to higher levels of mass reduction and becomes increasingly optimized for mass and other attributes, the cost for mass reduction will progressively increase. The higher levels of mass reduction may, for example, require applying more advanced materials and technologies than lower levels of mass reduction, which means that the cost of achieving those higher levels may increase accordingly. The unit cost of mass reduction versus the amount of mass reduction might be linear, parabolic, or some other higher order relationship. In the 2017-2025 Notice of Intent, 75 FR 62739 (Oct. 13, 2010), CARB, EPA and NHTSA derived a second order curve based on a study with two vehicle redesigns conducted by Lotus Engineering completed in 2010, such that zero mass reduction had zero cost, and the dollars per pound increased with greater levels of mass reduction. Since the publication of the TAR, the agencies have identified a number of additional studies in the literature relating to the costs of vehicle mass reduction, which are discussed below. The studies show that for low or high mass reduction, the costs can range from small cost savings to significant cost increases. The economic costs associated with mass reduction are difficult to determine conclusively due to the broad range of methods employed to achieve mass reduction. The costs on a specific vehicle or component depend on many factors, such as the design, materials selected, raw material price, appropriate manufacturing processes, production volume, component functionality, required engineering and development, etc. eee Specifically, the 2002 NAS Report estimated that vehicle weight could be reduced by 5 percent (without engine downsizing) at a cost of \$210-\$350, which translates into \$1.50/lb assuming a 3,800 Ib base vehicle and using the midpoint cost; Sierra Research estimated that a 10 percent reduction (with compounding) could be accomplished for \$1.01/lb, and MIT researchers estimated that a 14 percent reduction (with no compounding) could be accomplished for \$1.36/lb. References for these studies are available in endnotes to Chapter 3 of the TSD for the MYs 2012-2016 final rule. 3-227 ------- Technologies Considered in the Agencies' Analysis Cost data thus varies widely in the literature. Of the various studies reviewed by the agencies, not all are equal in their original intent, rigor, transparency, or applicability to this regulatory purpose. The individual studies range from complete vehicle redesign to advanced optimization of individual components, and were conducted by researchers with a wide range of experience and background. Some of the studies were literature reviews, while others developed new designs for lighter components or complete lighter vehicles, while yet others built physical components or systems, and conducted testing on those components and systems. Some of the studies focused only on a certain sub-system (which is a building block for the overall vehicle design), while some of them took a systematical approach and re- designed the whole vehicle to achieve the maximum mass reduction and cost reduction. The latter studies typically identified a specific baseline vehicle, and then utilized different engineering approaches and investigated a variety of mass-reduction concepts that could be applied to that vehicle. Some of the differences between studies emanate from the characteristics of the baseline vehicle and its adaptability to the new technology or method, and the cost assumptions relating to the original components and the redesigned components. Assumptions regarding the degree and cost of any associated mass de-compounding can also confound comparisons. Despite this variation in the literature, in actual practice, we believe manufacturers will choose a target mass reduction for a whole vehicle and for each sub-system, and work to find the lowest total cost method to achieve those targets. Such a process would consider numerous primary and secondary cost factors (including engineering, facilities, equipment, tooling, and retraining costs) as well as technological and manufacturing risks.ggg Regardless of the confidence in specific estimates, the agencies must select a curve that will be applied to the whole fleet that will define the average cost per kg of mass reduction as a function of total percentage of mass reduction. There are many significant challenges that make it difficult for the agencies to establish an estimated cost curve based on the literature, such as the differences in the baselines used in the studies, whether the studies considered platform sharing and powertrain sharing, and other considerations. fff The concept of secondary weight savings or mass compounding (also called mass decompounding) derives from the qualitative understanding that as vehicle weight decreases, other vehicle systems can also decrease in mass while maintaining the original vehicle level of performance and function. For instance, following a primary weight reduction in the vehicle (e.g. Body in White), the designs of some of the other dependant vehicle subsystems (tires, suspensions, brakes, powertrain, body structure) may be redesigned and reduced in mass to account for the overall lighter vehicle. The lighter vehicle is also associated with lighter loads, less friction and drag, and may require less power to be accelerated, and the powertrain may therefore be scaled down in size with a potential for reduced mass, even while maintaining equivalent acceleration performance and functionality. The compounded or secondary mass savings from these additional systems may then drive further mass reductions in the original primary weight reduction (e.g. Body in White). Mass compounding factors found in literature are rough estimates of the secondary mass reduction amount. 888 We also note that the cost of mass reduction in the CAFE model is quantified on a per pound basis that is a function of the percentage decrease in vehicle mass. We assume that OEMs would find the most cost-effective approach to achieve such a mass reduction. Realistically, this would depend heavily on the baseline vehicle as well as the size and adaptability of the initial design to the new technology. Thus, the CAFE model strives to be realistic in the aggregate while recognizing that the figures proposed for any specific model may be debatable. 3-228 ------- Technologies Considered in the Agencies' Analysis The costs for mass reduction employed for the main analysis for this final rule are the same as those in the NPRM. The agencies considered updating cost estimates based on the studies that were underway when the NPRM was issued. Those studies included the EPA/ICCT funded Phase 2 Toyota Venza Low Development project and the NHTSA funded Honda Accord mass reduction project, which are described in the section titled "What additional studies are the agencies conducting to inform our estimates of mass reduction amounts, cost, and effectiveness? " However, these studies were in the middle of the peer review process and had not yet been finalized at the time when the inputs for the main analysis for this final rule were required. For the final rule, the agencies decided to continue to use the same costs for mass reduction that were used in the NRPM. The agencies examined all the studies in Table 3-128 including information supplied by manufacturers (during meetings held subsequent to the TAR) when deciding the mass reduction cost estimate used for the proposal, which has been carried forward for this FRM. The agencies considered three major factors in examining these studies. First, whether a study was rigorous in terms of how it evaluates and validates mass reduction from technological and design perspectives. This includes consideration of a study's comprehensiveness, the technical rigor of its methodology, the validation methods employed, and the relevance of the technologies evaluated in the study given our rulemaking time frame. Second, whether a study was rigorous in terms of its estimation of costs, including the completeness and rigor of the methodology, such as whether the study includes data for all categories of direct manufacturing costs, and whether the study presents detailed cost information for both the baseline and the light-weighted design. And third, the degree of peer review, including if the study is peer-reviewed, and whether it has effectively addressed any critical technical, methodological, and cost issues raised by the peer-review, if this information is available. Some of the variation may be attributed to the complexity of mass reduction as it is not one single discrete technology and can have direct as well as indirect effects on other systems and components. The 2011 NAS study speaks to this point when it states on page 7-1 that "[t]he term material substitution oversimplifies the complexity of introducing advanced materials, because seldom does one part change without changing others around it." These variations underscore that there is not a unique mass reduction solution as there are many different methods with varying costs for taking mass out of vehicles, and every manufacturer, even every vehicle, could have a different approach depending on the specific vehicle, assembly plant and model year of implementation. The agencies recognize that there are challenges to characterizing the mass reduction plans for the entire future fleet due to the complexity and variety of methods available. So far the agencies have not found any study that addresses how to generalize the mass reduction that is achievable on a single vehicle to ^ The agencies considered confidential cost information provided by OEMs that covered a range of components, systems, designs and materials. Some of these cost estimates are higher than some of the literature studies, and manufacturers provided varying levels of detail on the basis for the costs such as whether mass compounding is included, or whether the costs include markup factors. 3-229 ------- Technologies Considered in the Agencies' Analysis the whole fleet. Table 3-128 contains a summary of the data contained in the studies, and the OEM CBI data, which the agencies reviewed. There is a degree of uncertainty associated with comparing the costs from the range of studies in the literature when trying to summarize them in a single table, and we encourage interested stakeholders to carefully review the information in the literature. For some of the cost estimates presented in the papers there are unknowns such as: what year the costs are estimated for, whether mass decompounding (and potential resultant cost savings) was taken into account, and whether mark-ups or indirect costs were included. The agencies tried to normalize the cost estimations from all these studies by converting them to 2009 year dollar, applying mass compounding factor of 1.35 for mass reduction amount more than 10 percent if it has not been applied in the study and factoring out the RPE specified in the study to derive direct manufacture costs for comparison. There are some papers that give cost for only component mass reduction, others that have more general subsystem costs and others yet that estimate total vehicle mass reduction costs (which often include and present data at the subsystem level). Other studies have multiple scenarios for different materials, different vehicle structures and mass reduction strategies. Thus, a single study which contains more than one vehicle can be broken down into a range of vehicle types, or at the subsystem level, or even at the component level. While Table 3-128 is inclusive of all of the information reviewed by the agencies for the NPRM, for the reasons described above the technical staff for the two agencies applied various approaches in evaluating the information. The linear mass-cost relationship developed for the proposal is carried forward to this final rule and presented below is the consensus assessment from the two agencies of the appropriate mass cost for this final rule. Table 3-128 Mass Reduction Studies Considered for Estimating Mass Reduction Cost for this FRM studies ra 01 tt 8 Cost Information from Studies S1 c .0 « 3 -a oi •— ' 1 £ "oi .SP U) 'at ra -S nn s HO °^, = BO 'c c •?? ^ B 0) Q. DC £ ui o i "§" •in- tt 8 LLJ Q_ ce. O Ol "5. 4-* ^J i_ = S O o O (N tt 8 bo c X = L. W 5 -2 R 1 a § vt VI i — i (U -Q MS" to O -S ts D Di Individual Cost Data Points AISI, 1998(ULSAB) AISI, 2000 (ULSAC) Austin et al, 2008 (Sierra Research) - ULS Unibody Austin et al, 2008 (Sierra Research) - AL Unibody Austin et al, 2008 (Sierra Research) - ULS BoF Austin et al, 2008 (Sierra Research) - AL BoF Bull et al, 2008 (Alum Assoc.) - AL BIW Bull et al, 2008 (Alum Assoc.) - AL Closure Bull et al, 2008 (Alum Assoc.) - Whole Vehicle 1998 2000 2008 2008 2008 2008 2008 2008 2008 103 6 320 573 176 298 279 70 573 1 1 1 1 1 1 1 1 1 103 6 320 573 176 298 279 70 573 2977 2977 3200 3200 4500 4500 3378 3378 3378 3.5% 0.2% 10.0% 17.9% 3.9% 6.6% 8.3% 2.1% 17.0% -\$32 \$15 \$209 \$1,805 \$171 \$1,411 \$455 \$151 \$122 1.0 1.0 1.61 1.61 1.61 1.61 1.0 1.0 1.0 1.28 1.24 1.01 1.01 1.01 1.01 1.01 1.01 1.03 -\$41 \$18 \$131 \$1,134 \$107 \$887 \$460 \$153 \$126 -\$0.40 \$2.99 \$0.41 \$1.98 \$0.61 \$2.98 \$1.65 \$2.17 \$0.22 3-230 ------- Technologies Considered in the Agencies' Analysis Cheah et al, 2007 (MIT) - 20% Das, 2008 (ORNL) - AL Body & Panel Das, 2008 (ORNL)-FRPMC Das, 2009 (ORNL) - CF Body & Panel, AL Chassis Das, 2010 (ORNL) - CF Body & Panel, Mg Chassis EEA, 2007 - Midsize Car - Adv Steel EEA, 2007- Midsize Car- Plast/Comp EEA, 2007- Midsize Car-AI EEA, 2007 - Midsize Car- Mg EEA, 2007 - Light Truck - Adv Steel EEA, 2007 - Light Truck - Plast/Comp EEA, 2007 -Light Truck - Al EEA, 2007 -Light Truck - Mg Geek etal, 2008 (Ford) Lotus, 2010 - LD Lotus, 2010- HD Montalbo et al, 2008 (GM/MIT) - Closure-HSS Montalbo et al, 2008 (GM/MIT) - Closure- AL Montalbo et al, 2008 (GM/MIT) - Closure- Mg/AL Plotkin et al, 2009 (Argonne) 2007 2008 2008 2009 2010 2007 2007 2007 2007 2007 2007 2007 2007 2008 2010 2010 2008 2008 2008 2009 712 637 536 933 1173 236 254 586 712 422 456 873 1026 1310 660 1217 25 120 139 683 1 1 1.0 1 1 1 1 1.35 1.35 1 1 1.35 1.35 1 1 1 1 1 1 1 712 637 536 933 1173 236 254 791 961 422 456 1179 1385 1310 660 1217 25 120 139 683 3560 3363 3363 3363 3363 3350 3350 3350 3350 4750 4750 4750 4750 5250 3740 3740 4000 4000 4000 3250 20.0% 19.0% 15.9% 27.7% 34.9% 7.0% 7.6% 23.6% 28.7% 8.9% 9.6% 24.8% 29.2% 25.0% 17.6% 32.5% 0.6% 3.0% 3.5% 21.0% \$646 \$180 -\$280 \$1,490 \$373 \$179 \$239 \$1,388 \$1,508 \$291 \$398 \$1,830 \$1,976 \$500 -\$121 \$362 \$10 \$110 \$110 \$1,300 1.0 1.5 1.5 1.5 1.5 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.03 1.01 1.01 1.00 1.00 1.03 1.03 1.03 1.03 1.03 1.03 1.03 1.03 1.01 1.00 1.00 1.01 1.01 1.01 1.00 \$667 \$121 -\$189 \$993 \$248 \$185 \$247 \$1,434 \$1,558 \$301 \$411 \$1,891 \$2,042 \$506 -\$120 \$360 \$10 \$111 \$111 \$1,300 \$0.94 \$0.19 -\$0.35 \$1.06 \$0.21 \$0.78 \$0.97 \$1.81 \$1.62 \$0.71 \$0.90 \$1.60 \$1.47 \$0.39 -\$0.18 \$0.30 \$0.41 \$0.92 \$0.80 \$1.90 (... Continued) Mass Reduction Studies Considered for Estimating Mass Reduction Cost for this FRM Studies (5 01 to 8 Cost Information from Studies 5" o 'r, 3 •D Ol CC U) 0 as u. no c O Q. F ,s c K ^ -Q 2 u> H .= = -a 0 C Ol 3 £ O V) °- V, C 3,s £ M '01 |u !c 01 _c !X 5 W) 2, •— W) l! la K £ i "5" ^i to 3 LLJ Q. rf at 8 fM 2 ^ Ol "a. "3 >. J5 O Q 8 t>0 1 3 C as £ a § o ^ O •g 3 01 ce. in ra M- o 3 „ .ti — D S Cost Curves NAS, 2010 2010 2010 2010 2010 2010 1.0% 2.0% 5.0% 10.0% 20.0% \$ 1.41 \$ 1.46 \$ 1.65 \$ 1.52 \$ 1.88 3-231 ------- Technologies Considered in the Agencies' Analysis OEMl OEM2 OEMS OEM4 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 2011 2011 2011 8.0% 9.0% 9.5% 10.0% 11.0% 0.4% 0.9% 1.9% 2.3% 2.4% 3.1% 3.6% 4.0% 4.1% 4.5% 4.8% 5.0% 4.0% 7.5% 10.0% 6.9% 8.1% 16.4% S 6.00 S 7.00 S 8.00 S 12.00 S 25.00 S S o.io S 0.20 S 0.33 S 0.38 S 0.60 S 0.76 S 0.85 S 0.88 S 0.98 S 1.09 S 1.17 S 0.57 S 1.01 S 1.51 S 0.97 S 1.02 S 1.95 EPA and NHTSA scrutinized the various available studies in the literature as well as confidential information provided by several auto firms based on the kinds of factors described above for purposes of estimating the cost of mass-reduction in the 2017-2025 timeframe. We determined that there was wide variation across the studies with respect to costs estimates, applicability to the 2017-2025 time frame, and technical rigor. The mass cost curve that was developed is defined by the following equation and is shown in Figure 3-33: Mass Reduction Direct Manufacturing Cost (DMC) (\$/lb) = \$4.36/(%-lb) x Percentage of Mass Reduction Level (%) (2010\$) 3-232 ------- Technologies Considered in the Agencies' Analysis \$1.00 \$0.90 \$0.80 S- \$0.70 £> \$0.60 £ \$0.50 ~ \$0.40 = \$0.30 \$0.20 \$0.10 \$0.00 0 Mass Reduction Cost jrf X ^S\ ^ J^ XL Slope = 4.S36 ^X* _x X ^ ^T x^ % 5% 10% 15% 20% 25% Percent of Mass Reduction Figure 3-33 NPRM and FRM Mass Reduction Direct Manufacturing Cost For example, this results in an estimated \$175 cost increase for a 10% mass reduction of a 4,0001b vehicle (or \$0.44/lb), and a \$394 cost increase for 15% reduction on the same vehicle (or \$0.66/lb). As mentioned in the NPRM, due to the wide variation in data used to select this estimated cost curve, the agencies have also conducted cost sensitivity studies in their respective RIAs in both the proposal and final rule using values of+/-40%. The wide variability in the applicability and rigor of the studies also provides justification for continued research in this field. The agencies consider this DMC to be applicable to the MY2017 and consider mass reduction technology to be on the flat portion of the learning curve in the MY2017-2025 timeframe. To estimate indirect costs for applied mass reduction of up to 15%, the agencies have applied a low complexity ICM of 1.24 through 2018 and 1.19 thereafter. To estimate indirect costs for applied mass reduction of 15% to 25%, the agencies have applied a medium complexity ICM of 1.39 through 2024 and 1.29 thereafter. To estimate indirect costs for applied mass reduction greater than 25%, the agencies believe it is appropriate to apply a highl complexity ICM of 1.56 through 2024 and 1.35 thereafter. The agencies sought comment in the draft Joint TSD for the NPRM (p. 210) regarding options for realistically and appropriately assessing the degree of feasible mass reduction for vehicles in the rulemaking timeframe and the total costs to achieve that mass reduction, but got no specific response. The agencies also sought comments on what practical limiting factors need to be considered when considering maximum feasible amount of mass reduction; the degree to which these limiting factors will impact the amount of feasible mass reduction 3-233 ------- Technologies Considered in the Agencies' Analysis (in terms of the percent of mass reduction); the best method(s) to assess an appropriate and feasible fleet-wide amount mass reduction amount (because each study mainly focuses on a single vehicle); etc. In its comments, VW stated that it "projects full vehicle weight reductions during the time period of this regulation on average in the order of 7-10%." VW noted that this was lower than the agencies' estimates in the NPRM of upwards of 20% mass reduction for large cars and some trucks, which VW stated may exceed cost effective limits. As stated later in this section, the detailed studies sponsored by the agencies suggest that 20% mass reduction is likely feasible for the rulemaking period without using exotic materials or highly advanced technologies. The accompanying detailed cost analysis indicates that the cost of reducing mass by 20% can potentially be economical. The agencies also noted in the NPRM that we expected to refine our estimate of both the amount and the cost of mass reduction between the NPRM and the final rule based on the agencies' ongoing work described a later section, below. As stated before, due to the limited time and the extensive scope of these studies, the agencies did not finish them in time for inclusion in the final rule analysis. How effective do the agencies estimate that mass reduction will be? A rule of thumb used by researchers and industry, based on testing and simulation, is that 10 percent reduction in vehicle mass can be expected to generate a 6 to 8 percent increase in fuel economy if the vehicle powertrain and other components are also downsized accordingly.102 In the analysis for the MYs 2012-2016 final rule, NHTSA and EPA estimated that a 10 percent mass reduction with engine downsizing would result in a 6.5 percent reduction in fuel consumption while maintaining equivalent vehicle performance (i.e., 0-60 mph time, towing capacity, etc.), consistent with estimates in the 2002 NAS report. For small amounts of mass reduction, such as the 1.5 percent used at vehicle refresh in NHTSA's modeling, no engine downsizing was used, so a 10 percent mass reduction without engine downsizing was assumed to result in a 3.5 percent reduction in fuel consumption. In this FRM, both agencies have chosen to use the effectiveness value for mass reduction from EPA's lumped parameter model to maintain consistency. EPA's lumped parameter model- estimated mass reduction effectiveness is based on a simulation model developed by Ricardo, Inc. under contract to EPA. The 2011 Ricardo simulation results show an effectiveness of 5.1 percent for every 10 percent reduction in mass. NHTSA has assumed that for mass reduction amounts less than 10 percent, the effectiveness is 3.5 percent. For mass reduction greater than 10 percent, NHTSA estimates the effectiveness is 5.1 percent in order to avoid double counting benefits - because the effectiveness of engine downsizing is included in the effectiveness of the engine decision tree when applying engine downsizing, it should appropriately be removed from the mass reduction effectiveness value in the mass reduction decision tree. EPA applies an effectiveness of 5.1 percent for every 10 percent mass reduction, and this scales linearly from 0 percent mass reduction, up to the maximum applied mass reduction for any given vehicle, which in this final rule is never larger than 20 percent. What additional studies are the agencies conducting to inform our estimates of mass reduction amounts, cost, and effectiveness? In the MYs 2012-2016 final rule, the agencies stated that there are several areas concerning vehicle mass reduction and vehicle safety on which the agencies would focus their research efforts and undertake further study. The following vehicle level projects focus on the 3-234 ------- Technologies Considered in the Agencies' Analysis goals stated in the MYs 2012-2016 final rule, which include determining the maximum potential for mass reduction in the MY 2017-2025 timeframe by using advanced materials and improved designs while continuing to meeting safety regulations and voluntary guidelines and while maintaining all aspects of vehicle functionality. The fourth study investigates the effects of resultant study designs on fleet safety by evaluating crash performance with objects and other vehicles of different size and mass. 1. NHTSA sponsored mass reduction study on a Honda Accord 2. EPA sponsored mass reduction study on a Toyota Venza (Phase 2 Low Development) 3. California Air Resources Board mass reduction study on a Toyota Venza (Phase 2 High Development) 4. NHTSA fleet-wide simulation study - crash analysis using the resultant designs from the studies 1-3 with objects and the design models of other vehicles with different size and mass. Due to the extensive scope of work for these studies and tight time schedule, some of the studies were finished, but peer reviews and response to peer reviews were not completed in time to enable the results to inform the final rule. We note, however, that the intermediate results from the mass reduction studies would corroborate the level of feasible amount of mass reduction the agencies chose to apply in the NPRM and FRM analyses. Rulemaking modeling results show that the costs for mass reduction are not sensitive to the cost curve of the rulemaking. In the NPRM, EPA found that a +/- 40% change in the cost of mass reduction had very little impact on the cost of the program. This is largely because of safety restraints imposed in the amount of mass reduction selected for the various vehicle classes primarily drive the penetration rates of the technology, rather than the relative cost- effectiveness of the technology itself. The following sections describe the status and results of the studies sponsored by the agencies. NHTSA Sponsored Mass Reduction Study BACKGROUND: NHTSA awarded a contract in December 2010 to Electricore, with EDAG and George Washington University (GWU) as subcontractors, to study the maximum feasible amount of mass reduction for a mid-size car - specifically, a Honda Accord - while keeping the vehicle functionality the same as the baseline vehicle. The Electricore/EDAG/GWU project team was charged with maximizing the amount of mass reduction using technologies that are considered feasible for production of 200,000 units per year during the time frame of this rulemaking while maintaining retail price in parity (within ±10%) with the baseline vehicle. In addition, all designs, materials, technologies and manufacturing processes must be realistically projected to be feasible for industry-wide application in MYs 2017-2025. The project focused on mass reduction and allowed powertrain downsizing, however alternative powertrains, such as diesels, HEVs and EVs, were not to be considered. 3-235 ------- Technologies Considered in the Agencies' Analysis MATERIAL AND TECHNOLOGY SELECTION: For vehicle redesigns, OEMs normally select technologies, materials and manufacturing processes that are currently in use on existing vehicle platforms or planned to be in use on future vehicle platforms. The use of the same or similar technologies, materials and manufacturing processes helps maintain or improve component and vehicle reliability, manufacturability and cost. New materials, technologies and processes are often introduced in low-volume, high price vehicles first and then migrate to high production volume vehicle lines over time. This significantly reduces the risk to OEMs associated with implementing new technologies. Recognizing this when selecting materials, technologies and manufacturing processes, the Electricore/EDAG/GWU team utilized, to the extent possible, only those materials, technologies and design which are currently used or planned to be introduced in the near term (MY 2012-2015) on low-volume production vehicles. The recommended materials (Advanced High Strength Steels, Aluminum, Magnesium and Plastics) manufacturing processes (Stamping, Hot Stamping, Die Casting, Extrusions, Roll Forming) and assembly methods (Spot welding, Laser welding and Adhesive Bonding) are at present used, some to a lesser degree than others. These technologies can be fully developed within the normal product design cycle using the current design and development methods. The process parameters for manufacturing with Advanced High Strength Steels can be supported by computer simulation. This approach minimized those material and technology options which would likely be overly aggressive or unrealistic to implement in mass production in model years 2017-2025. ENGINEERING APPROACH: The Electricore/EDAG/GWU team took a "clean sheet of paper" approach and adopted collaborative design, engineering and CAE process with built-in feedback loops to incorporate results and outcomes from each of the design steps into the overall vehicle design and analysis. The team torn down and benchmarked 2011 Honda Accord and then undertook a series of baseline, noting the designs, materials, technologies and overall design optimization level of the baseline vehicle. Vehicle performance, safety simulation and cost analyses were run in parallel to the design study to help ensure that the design decisions for the concept vehicle would be informed by a well- documented baseline, thus enabling the resultant design to meet the defined project criteria. While working within the constraint of maintaining the baseline Honda Accord's exterior size and shape, the body structure was first redesigned using topology optimization with six load cases including bending stiffness, torsion stiffness, IIHS frontal impact, IIHS side impact, FMVSS pole impact, FMVSS rear impact and FMVSS roof crush cases. The load paths from topology optimization were analyzed and interpreted by technical experts and the results were then fed into low fidelity 3G (Gauge, Grade and Geometry) optimization programs to further optimize for material properties, material thicknesses and cross-sectional shapes while trying to achieve the maximum amount of mass reduction. The Electricore/EDAG/GWU team carefully reviewed the optimization results and built detailed CAD/CAE models for the body structure, closures, bumpers, suspension, and instrumentation panel. The vehicle designs were also carefully reviewed by manufacturing technical experts to ensure that they could be manufactured at high volume production rates. Detailed manufacturing layouts were created and were later used to estimate costs. Multiple materials were used for this study. The body structure was redesigned using a significant amount of advanced high strength steel (AHSS). The closure and suspension were 3-236 ------- Technologies Considered in the Agencies' Analysis designed using a significant amount of aluminum. Magnesium was used for the instrumentation cross-car beam. A limited amount of composite material was used for the seat structure. Electricore and its sub-contractors consulted industry leaders and experts for each component and sub-system when deciding which mass reduction technologies were feasible. DESIGN AND FUNCTION VALIDATION: In order to ensure that the light weighted vehicle had the same functionality as the baseline vehicle, Electricore and its sub-contractors used the CAD/CAE/powertrain models and conducted simulation modeling. This is the first mass reduction study that has been released publicly that includes such a broad array of vehicle simulation modeling analyses to assess vehicle functionality and performance relative to these critical attributes. These significant additional analyses provide greater confidence that the designs employed in this study are more feasible for production implementation than a study without these analyses, although the agency notes that significantly more testing and validation work is required to refine and finalize a design for production. • Safety: Safety performance of the light-weighted design is compared to the safety rating of the baseline MY2011 Honda Accord for seven consumer information and federal safety crash tests using LS-DYNA111. These seven tests are NCAP frontal test, NCAP lateral MDB test, NCAP lateral pole test, IfflS roof crush, IfflS lateral MDB, IMS front offset test, and FMVSS No. 301 rear impact tests. All tests achieved safety performance equivalent to MY 2011 Honda Accord when comparing crash pulse and passenger compartment intrusion levels, with no damage to the fuel tank. This study does not include restraint systems and dummy which would be part of NHTSA's fleet simulation study. • Body Stiffness/ Ride and Handling/NVH: Vehicle body torsional and bending stiffness are signatures for the vehicle structure performance. Higher stiffness is generally associated with a refined ride and handling qualities. The baseline vehicle body structure underwent testing for normal modes of vibration, and torsion and bending stiffness. A detailed FEA model of the light-weighted structure was created and analyzed using the MSC/NASTRAN simulation. The torsional stiffness of the light-weighted design is 30% higher than the baseline vehicle while the bending stiffness is 40% higher. The normal mode frequency test results for the light-weighted body structure, which represents vehicle dynamic stiffness, also are within 2.3% of the targets. These stiffness and modes results show that the light-weighted design will have improved ride and handling and improved NVH performance comparing to a vehicle with lower stiffness. • Vehicle Ride and Handling: In the light-weighted design, the front suspension is redesigned using a MacPherson strut instead of the heavier double wishbone used in 111 LS-DYNA is a software developed by Livermore Software Technologies Corporation used widely by industry and researchers to perform highly non-linear transient finite element analysis. 3-237 ------- Technologies Considered in the Agencies' Analysis the baseline vehicle. Vehicle ride and handling is evaluated using MSC/ADAMSJJJ modeling on five maneuvers, fish-hook test, double lane change maneuver, pothole test, 0.7G constant radius turn test and 0.8G forward braking test. The results from the fish-hook test show that the light-weighted vehicle can achieve a five-star rating for rollover, same as baseline vehicle. The double lane change maneuver tests according to the ISO standard show that the chosen suspension geometry and vehicle parameter of the light-weighted design are within acceptable range for safe high speed maneuvers. These simulations are performed to further validate the chosen light weighted front suspension design. • Durability: There are two types of durability, stress related and corrosion related. Stress related durability for the light-weighted vehicle is evaluated using strain-based analysis based on pot hole, 0.8G forward braking and 0.7G cornering road load cases using ADAMS model. Results from the simulation show that the life of the light- weighted vehicle body structure exceeds the targets. Although timing and funding did not allow corrosion testing to be conducted, the Electricore/EDAG/GWU team considered the properties of materials used, and the location and the functionality of the components to avoid potential issues with corrosion. • Powertrain Performance: The powertrain of the light-weighted vehicle is downsized from 2.4L naturally aspirated engine to 1.8L naturally aspirated engine to maintain the same vehicle acceleration and towing compared to the baseline 2011 Honda Accord. A powertrain simulation tool PSAT1^ is used to verify and validate the light-weighted vehicle for fuel economy and powertrain performance. The light-weighted vehicle with 1.8L NA engine will have 32 mpg fuel economy with comparable 0-30 mph time, 0-60 mph time, quarter mile time, gradability and maximum speed at grade. The only metrics that the light-weighted vehicle performs less than the baseline vehicle is vehicle maximum speed (127 mph for the baseline Accord and 112 mph for the light- weighted design) which the Electricore/EDAG/GWU team and NHTSA believe is acceptable. As a result of the improved fuel economy, the fuel tank for the light- weighted vehicle can be reduced from 18.5 gallon to 15.8 gallon with the same driving range, which further reduced vehicle weight both by reducing fuel tank mass and the mass of fuel carried by the vehicle. • Manufacturability: The manufacturability of all proposed body structure panels were then assessed using simulation tools, which included HYPER-FORM for stamping parts, and other single step process simulation tools for parts manufactured using other methods, such as hot stamping for B-pillar. JJJ MSC/ ADAMS: Macneal-Schwendler Corporation/Automatic Dynamic Analysis of Mechanical Systems. kkk PSAT is a plug-and-play architecture software that allows the user to build and evaluate a vehicle's fuel economy and powertrain performance under varying load conditions and drive cycles. It uses MATLAB in a Simulink environment to record data, calculate and input powertrain requirements based on driver demand and current powertrain values. The software is sponsored by the U.S. Department of Energy and developed by Argonne National Laboratory (ANL). http://www.transportation.anl.gov/modeling_simulation/PSAT/index.html 3-238 ------- Technologies Considered in the Agencies' Analysis COST ANALYSIS; A detailed cost analysis for the light weighted design and cost estimates for alternative design options were also conducted. For OEM-manufactured parts, a detailed cost model was built based on a Technical Cost Modeling (TCM) approach developed by the Massachusetts Institute of Technology (MIT) Materials Systems Laboratory's research103 for estimating the manufacturing costs of OEM parts. The costs were broken down into each of the operations involved in the manufacturing, such as for a sheet metal part production by starting from blanking the steel coil, until the final operation to fabricate the component. Total costs were then categorized into fixed cost, such as tooling, equipment, and facilities; and variable costs such as labor, material, energy, and maintenance. These costs were assessed through an interactive process between the product designer, manufacturing engineers and cost analysts. For OEM-purchased parts, the costs were estimated by consultation with experienced cost analysts and Tier 1 suppliers. Forty-one concise spreadsheets are created for both the baseline vehicle and the light-weighted design in the cost model to calculate both the manufacturing and assembly costs. FINAL RESULTS: To achieve the same vehicle performanc```