United States         Air and Radiation       EPA420-R-02-004
          Environmental Protection                 January 2002
          Agency                       M6.HDE.003
&EPA    Development of
          Heavy-Duty NOx
          OFF-CYCLE Emission
          Effects for MOBILE6
                                 y£u Printed on Recycled
                                 Paper

-------
                                                        EPA420-R-02-004
                                                            January 2002
    Development of Heavy-Duty NOx OFF-CYCLE
              Emission Effects for MOBILE6

                          M6.HDE.003
                           Edward L. Glover
                     Assessment and Modeling Division
                   Office of Transportation and Air Quality
                   U.S. Environmental Protection Agency
                               NOTICE

  This technical report does not necessarily represent final EPA decisions or positions.
It is intended to present technical analysis of issues using data that are currently available.
       The purpose in the release of such reports is to facilitate the exchange of
     technical information and to inform the public of technical developments which
      may form the basis for a final EPA decision, position, or regulatory action.

-------
1.0    Background
       For purposes of this modeling, the heavy-duty diesel vehicle (HDDV) off-cycle operation is
a vehicle component or software which allows emissions in excess of the FTP certification standards
to be produced during operating modes which are not explicitly covered by a certification test while
still controlling emissions during the certification test. In the case of the heavy-duty vehicles, the
excess NOX emissions that were produced during off-cycle operation mostly occurred during steady-
state operating modes such as cruising down the freeway, and rarely occurred during transient
operation.

       In the late 1980's and early 1990's vehicles which produce off-cycle NOX emissions were
being phased  into the fleet. These vehicles were mostly confined to the heaviest of the heavy-duty
diesels (8a and 8b vehicles).  However, by the mid to late 1990's off-cycle NOX emission effects
were widespread on virtually all of the heavy end engines.  Even some of the medium heavy-duty
diesels contain off-cycle emission effects. Originally, it was believed that off-cycle emission effects
were scheduled to be reduced and eliminated by calendar year 2000. However, recent developments
regarding the Heavy-duty Engine Manufacturer's Consent Decree indicate that limited quantities of
engines will be produced in model years 2001 and 2002 that emit off-cycle NOX emissions.  As a
result, the effects for calendar year 2001 have now been included in MOBILE6.

       This document describes the methodology used to estimate excess NOX emissions produced
by heavy-duty diesel vehicles (HDDV) as the result of Off FTP  Cycle  operation.   In previous
MOBILE6 model development documents and in other related documents it was often referred to
as "built-in defeat devices" or "alleged defeat devices". The term 'defeat device' has been dropped
from  the title and body of this report on the grounds that the resulting Heavy-duty Engine
Manufacturers' Consent Decrees which arose from the investigation into the' defeat devices' did not
result in an admission or determination that the disputed devices were actually defeat  devices. The
term "Off Cycle Operation" meaning Off FTP Cycle operation, is being substituted because it better
describes the  effect which will be  modeled without the presumption of a "defeat device." Also
presented in the document are emission estimates based on two programs designed to mitigate the
effect of the off-cycle operation on fleet NOX emission levels. These include the incorporation of
more stringent HDDV standards earlier than originally required (pull ahead), and accelerated engine
rebuild programs (rebuild) to get in-use engines into better compliance.

       This document is NOT a legal document or an official emission inventory document. Such
official documents can be found in the official EPA Consent Decree documents and supporting
material.  This current document  is merely an attempt to compile estimates of the impact of the
heavy-duty vehicle NOX off-cycle emission  effects,  to account for the effect of the proposed
solutions from official sources, and to account for these estimates in the framework of the MOBILE6
model. These results may differ  from the other EPA emission estimates in tons per year or in other
regards. A reader interested in the fleet impact of the off-cycle operation based on official sources
should refer to the official consent decree document.

-------
2.0    Introduction
       The principal focus of this document is to describe the Heavy-Duty Diesel Vehicle Off Cycle
Operation in the context of how it was developed for MOBILE6. It also presents the underlying
parameters and assumptions, and presents some limited NOx emission results.  The second focus
is the heavy-duty off-cycle operation module that is built into the MOBILE6 emission factor model.
It was  developed and programmed into MOBILE6 using parameters from a spreadsheet model
previously discussed in EPA Report M6.HDE.003.

       This current document FINAL M6.HDE.003 differs from the DRAFT M6.HDE.003 in two
ways. First, it no longer discusses issues and results which were inherent to the previous spreadsheet
models such as by roadway ratios and comparisons between MOBILES based inputs and MOBILE6
based inputs.  These were developed so that off-cycle effects could easily be incorporated into
existing MOBILE type  outputs through the use of multiplicative factors.   Second, it includes
changes made in early calendar year 2001 that reflect updates made to the basic 'Consent Decree'
and assumptions made during the 2000 Heavy-duty emission standards rulemaking process.

       In cases where this current document differs from the previous document, M6.HDE.003, the
user should consider the previous document out of date. These differences are in the areas of Pull
Ahead effects, Engine Rebuild effects, and calendar year extensions.  The effect of all three of these
changes is to increase the emissions of Heavy-duty vehicle Off-Cycle NOx emissions. The previous
document is being kept active because it is the sole documentation for the heavy-duty off-cycle
spreadsheet models based on Draft MOBILE6 and MOBILES which had limited release in calendar
year 2000.
       This document is structured into three principal parts.  The first part describes the structure
and parameters that were used to develop the HDDV NOx Off-cycle effects in MOBILE6.  This part
is composed of several sub-sections.  These include a description of the model's overall structure,
the data parameters of the model and the control parameter.  The second part briefly describes the
the results from the model and references the companion spreadsheets which contain the data and
the results  The third part is an Appendix which contains the FORTRAN code used to implement
the algorithm in MOBILE6, and a list of comments from reviewers and EPA responses.

-------
3.0    Heavy-Duty Diesel Vehicle OFF-CYCLE Operation

       This section describes the overall structure, data parameters, and input / output structure of
the HDDV off-cycle operation. The Excel spreadsheet entitled "DDM6_Data.xls" is available that
contains a summary of the important data parameters.

3.1    Overview of the Basic Structure

       The MOBILE6 model is basically a calculator that weights the individual model year and
vehicle class emission factors with vehicle miles traveled (VMT) inputs and presents these results
as average emission factors in grams per mile.  The  following data parameters are used in the
MOBILE6 model, and are defined in the subsequent sections.
       MOBILE6 Heavy-Duty Vehicle Conversion Factors
       MOBILE6 Basic Emission Factors WITHOUT Off-Cycle Operation Effects
       Basic Emission Factors WITH Off-Cycle Operation Effects
       Off-Cycle Operation Parameter and Fleet Penetration Rates
       Total HDDV Vehicle Miles Traveled (VMT)
       MOBILE6 Distribution of VMT by Roadway Type, Vehicle Class and Model Year
       MOBILE6 HDDV Speed Correction Factors

       Several equations are used to weight together these parameters and sum the results. These
equations are shown in the section labeled "Calculations".
3.2    MOBILE6 Heavy-Duty Vehicle Conversion Factors

       The HDDV conversion factors (CF) used in this model are shown in Excel Spreadsheet
DDM6_Data.xls in worksheet CF.  These conversion factors were taken from MOBILE6 and are
referenced in EPA report "Update Heavy-Duty Engine Emission Conversion Factors for MOBILE6:
Analysis of BSFCs and Calculation of Heavy-Duty Engine Emission Conversi on Factors" -EPA420-
P-98-015.   The conversion factors are used to convert the basic heavy-duty diesel emission factors
from g/bhp-hr into g/mi units. This is necessary so that the four individual heavy-duty diesel vehicle
class emission factors and the individual roadway factors can be properly weighted together by
VMT.  Emission factors in units of g/bhp-hr cannot be VMT weighted because the basis units are
in bhp-hr rather than in miles.

       The conversion factors are shown as a function of model year and vehicle class. The model
years range from 1964 through 2030. Each column in the worksheet CF contains a conversion factor
for one of the four heavy-duty vehicle classes: Light, Medium, Class 8a and ClassSb. In all cases,
the Light class conversion factors are composite values obtained by weighting the individual sub-
classes (light-heavy duty standard classes of 2b, 3,4 and 5) together with the Bus conversion factors
based on VMT. This was done because neither the Light or Bus class contain off-cycle operation
effects in any appreciable quantity.

-------
3.3    MOBILE6 Basic Emission Factors WITHOUT Off-Cycle Operation Effects

       The basic emission factors without off-cycle operation effects used in the spreadsheet model
are shown in Excel Spreadsheet DDM6_Data.xls in worksheet BEFs.  These are the base emission
levels for a particular model year and vehicle class if the off-cycle operation is not present or not
working. The first five columns show the Without off-cycle operation emission factors. These are
shown as a function of vehicle class (Light, Medium, 8a and 8b), and as a function of model year.
Basic emission factors are shown for model years 1964 through 2030, and are in units of grams per
brake-horsepower-hr (g/bhp-hr or g/hp-hr). These values are the default emission levels for HDD Vs
in MOBILE6, and are very heavily based on certification test  results.  The analogous MOBILES
values are also default emission levels and are based on certification test results. The reference for
the MOBILE6 values  is EPA report EPA420-R-99-010  - "Update of Heavy-Duty Emission Levels
(Model Years 1988 -  2004+) for Use in MOBILE6." The MOBILES values were taken from the
MOBILES model code.

3.4    MOBILE6 Basic Emission Factors WITH Off-Cycle Operation Effects

       The last four columns in worksheet BEFs of the Excel  Spreadsheet DDM6_Data.xls show
the basic NOX emission factors in g/bhp-hr units when the off-cycle operation is active and
operating. Like the WITHOUT off-cycle operation emission levels, the WITH off-cycle emission
levels are shown  as a function of vehicle class and model year.  Except for the Pull Ahead effects,
the "Light-Duty"  class is unaffected by the off-cycle operation so it shows the same NOX emission
factors for both "WITH" and "WITHOUT" off-cycle operations. NOX emission factors are shown
for the model year range of 1964 through 2030. One exception is the WITH off-cycle operation
NOX emission factors for model years 2002 and 2003.  These are lower than the WITHOUT off-
cycle operation emission levels as the result of "Pull Ahead" requirements for lower certification
standards. For model years 2004  and later, and model years 1987 and  earlier, the WITH and
WITHOUT off-cycle  operation emission levels are  equivalent, since there would be no off-cycle
operation effects  on emissions for these model year engines.

       The off-cycle operation emission effects were determined on an engine by engine basis using
available data and engineering judgement by experts familiar with engine control software. These
individual estimates were weighted together by engine family using diesel vehicle sales information
to produce model year average emission levels (this document uses a summary of the individual
manufacturer results). The individual engine family or manufacturer specific engine control logic,
emission results and sales projections are proprietary in nature and were made available to the EPA
on a confidential basis as part of the Heavy-Duty Diesel Off-cycle operation Consent Decree. Thus,
they are not provided  in this document nor are they publically available in any other source.

3.5    Off-Cycle Operation Operating Parameter and Fleet Penetration Rates

       The off-cycle operation  operating parameters and fleet penetration rates used in  the
MOBILE6 model are shown in worksheet "Fleet Fractions" of Excel SpreadsheetDDM6_Data.xls.
These are shown as a function of vehicle class (light, medium, 8a and 8b), model year, and roadway
type.  The three roadway types are Urban, Arterial, and Interstate.  There are four sub-types within

-------
each of these three broad classes, and they are also listed in spreadsheet DD_Data.xls worksheet
"Fleet Fractions." The Urban type represents city driving which includes considerable stop-and-go
transient operation.  The Arterial type represents driving on primary roads in which there is both
transient operation and steady-state operation. The Interstate type represents the operation which
includes mostly steady state type driving at higher speeds.  A zero (0.00) means that the off-cycle
operation does not operate during that mode, and a one (1.00) would mean that it operates all of the
time in that mode. This information is on a fleet summary basis. It was calculated from individual
engine family off-cycle  operation  response  data that was determined  from proprietary  and
confidential data submitted by the engine manufacturers, limited testing of affected engines, and
engineering judgment by experts in engine control and emission control software. As a general rule,
the Urban type contains relatively little off-cycle operation and the Interstate type contains a very
high percentage of off-cycle operation as can be seen in the low fractions for Urban and the high
fractions for Interstate. The off-cycle operation operating fractions are also a function of model year.

       The columns entitled "Fleet Fraction" show the percent of a particular model year and vehicle
class that is equipped with the off-cycle operation or affected by the legal settlement.  A value of
zero means  for that particular combination of vehicle class and  model there are no  off-cycle
emissions, and a value of unity (1.0) means that in that model year, all vehicles of that class were
capable of producing off-cycle NOX emissions.  In general, the data  suggests that the  off-cycle
operation penetration began slowly in the 1988 model year on the heavier class diesels, and slowly
progressed to encompass most of the fleet. Virtually full penetration seems to have occurred by the
mid  1990's on the heavier engines. Penetration drops off abruptly around calendar year 2001,
presumably due to the settlement. The only exception to this rule is for the case of the 2002 and
2003 model years. These model year groups contain a coded factor to calculate the impact of "Pull
Ahead."  No off-cycle operations will be produced in those model years.

3.6    Total VMT Estimates

       The  Total VMT Estimates are  shown   in  worksheet  Total_VMT of  Spreadsheet
DDM6_Data.xls. Separate values are shown for calendar years 1988 through 2010.  The VMT
values were obtained from the EPA Trends model,  and are typical HDDV VMT levels used in Air
Quality modeling. Linear equations are used to proj ect future VMT levels. The Total VMT estimate
is an important factor in determining the total tons of NOX produced by the off-cycle operations.
However, it is not explicitly used  in the MOBILE6 model inputs because MOBILE6 does not
produce NOX emissions in units of tons (only in grams per mile).
3.7    Distribution of VMT by Roadway Type. Vehicle Class and Model Year

       The distribution of VMT by roadway type and vehicle class is an important topic that affects
the total quantity of off-cycle NOX emissions. However, all of the assumptions and data regarding
VMT distributions are present in MOBILE6, and are not unique to the HDDV Off-Cycle NOX
emission calculation.  Currently, the default MOBILE6 model assumes that all vehicle classes have
the same distribution of VMT by  roadway type, and that this single distribution was obtained from
analyzing overall vehicle VMT data trends.  The same  distribution was assumed for all vehicle

-------
classes because the roadside survey data could not be disaggregated by vehicle class (These data are
collected in roadway surveys when vehicles run over the rubber hoses on roadways, and the vehicle
class cannot easily be discerned). The effect of this type of data collection is that a heavy-duty diesel
truck is assumed to accumulate on a percentage of total VMT basis, the same amount of expressway
VMT as a passenger car.  This assumption is not likely accurate because it is common knowledge
that heavy-duty trucks spend more of their time on expressways than passenger cars. Therefore, this
assumption makes it likely that the  MOBILE6 model understates the percentage of expressway
driving where the off-cycle emissions are highest, and thereby understates the Off-Cycle NOX
emissions by a significant margin.

       The assumption of equal VMT distributions by roadway and vehicle class is the default
MOBILE6 assumption. However, user specific inputs to the program can override this assumption
and different values can be entered for each vehicle type.  Thus, if the user has good data on VMT
distribution by vehicle class for the  area they which to model, they can utilize it, and potentially
obtain a more accurate result.  See the MOBILE6 User's Guide for more details on how to enter
alternative VMT distribution and a sample alternate VMT distribution.
3.8    MOBILE6 HDDV Speed Correction Factors
       In this analysis, speed correction factors are used only to model the effect of speed on the
heavy-duty emissions which are associated with "without off-cycle operation."  The "with off-cycle
operation heavy-duty emission factors" implicitly contain speed correction factors, and need no
additional speed correction. Thus, the component of the off-cycle emission factor is not corrected
for speed. This implicit relationship is an assumption that the off-cycle operation emission effects
inherently contain all of the remaining speed related NOX emission effects that occur as the result
of vehicle operation outside of the certification test cycle, and the certification cycle's average speed
of 20 MPH.   The basis for this assumption is that the off-cycle operation effects are a strong
function of speed and typically only occur during the steady-state higher speed activity of the trucks.
Also, based on the limited test and engineering data, it is not feasible to separate off-cycle operation
NOX emission effects from normal speed correction NOX emission effects.

       The average speed for  each roadway type is used to calculate the NOX speed correction
factors.  The speed correction factor equation used in this  analysis is the standard MOBILE6
equation for heavy-duty diesels. It is not a function of roadway type.  The speed correction factors
are calculated for each of the roadway types by inserting the appropriate average speed into Equation
1 below:

       NOX Speed CF  =  0.676 - 0.0480*speed + 0.00071 *speed**2                 Eqn la

'**2' means squared.

-------
3.9    Calculations

       The calculation of the HDDV off-cycle effects in MOBILE6 are quite simple. They are
shown in the Appendix in the section with the FORTRAN code.   They are also shown for a given
roadway and vehicle class type using equation Ib.  The overall off-cycle effect is the product of the
base off-cycle effect in grams per mile and the fleet fraction that produce off-cycle emissions and
the percent of time a vehicle class spends in off-cycle mode for a given roadway.
Off-Cycle(g/bhp-hr)   =  BASE Off-Cycle Effect * Fleet Penetration Frac  *
                        Time spent in Mode(iroad)                                  Eqnlb
Where

BASE Off-Cycle Effect are the average NOx off-cycle emission increases on a per vehicle (discussed
in Section 3.4).

Fleet Penetration Frac are the fraction of the fleet by vehicle class that is capable of producing off-
cycle NOX emissions. (Section 3.5)

Time Spent in Mode(iroad) is the time that a given vehicle spends in Off-cycle mode for a given
roadway type (iroad).  Section  3.5 discusses this concept.
3.10   Rebuild Program Inputs to the Excess NOX Emission Spreadsheet Model

       Most  heavy-duty diesel engines are rebuilt after their  initial useful  life has elapsed.
Typically, this occurs between the  500,000 mile and 1,000,000  mile points, and  often includes
replacing engine parts that affect NOX emissions. The practice of rebuilding is considerably more
cost-effective to the truck owner than scrapping the engine.  Because the rebuild practice is so
widespread, the engine manufacturing industry agreed as part of  the Consent Decree to rebuild a
portion of the fleet in such a way as to reduce the effect of the off-cycle operation on rebuilt engines.
This NOX excess spreadsheet model will reflect  the effects of the rebuild program.

       The following parameters are defined and used in the MOBILE6 and MOBILES spreadsheet
models to account for rebuilds.

Rebuild Fraction  - This is the fraction of the eligible fleet that will be rebuilt.  Currently this
fraction is assumed to be 0.90.

Model Year Range - Two options were available in the consent decree. The one which is modeled
in the spreadsheet model includes only the 1994 through 1998 model years. The other rebuild option
(not modeled) includes the 1993 model year.

-------
After Rebuild Emission Level - This is the NOX emission level that the rebuilt engines will emit
in units of g/bhp-hr, and is set equal to 6.00 g/bhp-hr. These were conservative values because the
consent decree set these standards as "not-to-exceed" levels.

Years of Use Before Rebuild - This is the number of years prior to the first rebuild.
       12 years for a medium duty
       5 years for a 8 a truck
       5 years for a 8b truck
Also, no rebuild credit is given before calendar year 2000. Thus, a 1994 model year 8b engine will
not be lowered to the 6.00 g/bhp-hr standard upon its first rebuild in 1999.

       The effects of rebuild are implemented in the  model by replacing  the With Off-cycle
operation Emission levels with the rebuild emission level of 6.00 g/bhp-hr on the fraction of the fleet
that is rebuilt. The default value for the fraction of the fleet that is rebuilt is 90% of the 1994 through
1998 model years.

-------
4.0    Results from MOBILE6
       The MOBILE6 model was run after adding the HDDV off-cycle emission factors and
parameters. The results are presented in this section as a series of five graphs for HDD Class 8a and
Class 8b vehicles. The complete results including the underlying data and figures can be found in
the Excel Spreadsheet Offcycleresults.xls. All of the results are shown in terms of emission factors
in units of grams per mile.  Thus, they should not be confused with inventory type presentations
which are often in units of 'tons' and reflect assumptions about total vehicle miles traveled.

       Each of the five figures shows a different aspect of HDDV Off-Cycle NOX emissions as a
function of Calendar Year. The Y-axis of the figures is in terms of average NOX emissions for a
given calendar year for either the 8 A or 8B truck class. Because it is an average calendar year result,
it contains a weighted  distribution of 25 individual vehicle model years. Thus, the 2020 calendar
year result contains vehicle emission results from model years  1995 through 2020.

       Figures 4.1 and 4.2 show the HDDV NOX emissions versus calendar year for class 8a and
class 8b, respectively.  These figures include the total NOX emissions from these vehicles, not just
the off-cycle emission portion. The "No Offcycle" line shows what the average NOX emissions of
a heavy-duty diesel vehicle would be if the HDDV Off-Cycle Nox emissions had never existed.  The
"Default M6" line shows the contribution of the Off-Cycle emissions and the effects of the Pull
Ahead and Rebuild Programs at lowering the Off-cycle NOX emission impact. The remaining two
lines "w/o Pull Ahead" and "w/o Pull or Rebuild" show what the total Nox emissions would be if
Pull Ahead was not implemented, and if neither Pull Ahead or Rebuild were implemented.

       Comparing the "No Offcycle" line with the line  entitled "Default M6" shows a sizeable
increase in NOX emissions during calendar years 1990 through 2002. After that time, the off-cycle
emissions begin to be phased out, and the "No Offcycle" line and the "Default M6" line begin to
converge together. The figures also show the moderately higher emission levels that result if Pull
Ahead and Rebuild are not done.  This is particularly true in the time frame of calendar year 2000
through 2006.

       Figure 4.3 shows the HDDV NOX off-cycle emissions as a function of calendar year and
vehicle class 8a or 8b.  The figure shows that the bulk of the off-cycle emissions are produced from
calendar year 1990 through 2005. It also shows a slightly higher emission factor for the 8b vehicle
class versus  the 8a vehicle class. Figures 4.4 and 4.5 show the HDDV NOx off-cycle emission
reductions due to Pull Ahead and Rebuild, respectively.  As can be seen from  the figures, the Nox
emission  reductions for the two programs (Pull Ahead and Rebuild) are similar in magnitude, and
during their peak years reduce the NOX HDDV emission factors by about 2.0 grams per mile versus
a peak Off-cycle emission factor of about 9 grams per mile.

       Figures  were not generated for medium  duty diesel  vehicles although some off-cycle
emissions are generated by these vehicles.  This was done because the off-cycle NOx emission
effects are  generally  small.   The  necessary  data  do  exist  in the  attached spreadsheet
Offcycleresults.xls to generate such figures. This is left to the reader.

-------
35.0
                                   Figure 4.1
        MOBILES HDDV Class 8A NOx Emissions versus Calendar Year
 0.0  I	
                                                                       •No Offcycle
                                                                        Default M6
                                                                        w/o Pull Ahead
                                                                       •w/o Pull or Rebuild
     CY    1990   1994   1998   2002   2006   2010   2014   2018   2022   2026
                              Calendar Year
                                    Figure 4.2
         MOBILE6 HDDV Class 8B NOx Emissions versus Calendar Year
                                                                         No Offcycle
                                                                         Default M6
                                                                         w/o Pull Ahead
                                                                         w/o Pull or Rebuild
     CY  1990  1994  1998 2002  2006  2010  2014 2018  2022  2026
                               Calendar Year

-------
                                   Figure 4.3
               Total Off-Cycle NOx Emission Effect by Calendar Year
                            for Two Vehicle Classes
             1990   1994
                          1998
                                2002   2006   2010
                                    Calendar Year
                                                  2014   2018   2022   2026
                                    Figure 4.4
         Effect of HDDV PULL-AHEAD on NOx Emissions by Calendar Year
                             for Two Vehicle Classes
E
5
x
O
o
!
I
<
10.0

 9.0

 8.0

 7.0

 6.0

 5.0

 4.0

 3.0
             1990
                   1994
                         1998
                               2002    2006   2010

                                    Calendar Year
                                                  2014   2018
                                                               2022
                                                                     2026

-------
   10.

   9.

   8.
 )
~  6i
O
                                   Figure 4.5
           Effect of HDDV REBUILD on NOx Emissions by Calendar Year
                            for Two Vehicle Classes
O
^
O
   4.0
            1990   1994
                        1998
                              2002   2006   2010
                                   Calendar Year
                                                2014   2018  2022  2026

-------
                                    Appendix A
                     Changes From Document M6.HDE.003
       The EPA document DRAFT M6.HDE.003 discusses the methodology for modeling the
heavy-duty diesel off-cycle  emissions.   The methodology and parameters mentioned  in that
document are very similar to those presented in this current document FINAL M6.FtDE.003.
Although the previous sections of M6.FtDE.003 contain the new parameters and methodology, they
are not explicitly presented. This Appendix re-presents these new parameters in a brief, but more
explicit fashion.

A.I    The DRAFT M6.FtDE.003  document and the associated Excel  spreadsheets discussed
       spreadsheet models which could be used to model the effect of the heavy-duty diesel vehicle
       Off-Cycle NOx emissions. This current document FFNALM6.FIDE.003 focuses on this task
       through  the context of the MOBILE6  model,  and  thereby  supercedes the previous
       spreadsheet models.

A.2    The  previous spreadsheet  model  DRAFT  M6.FtDE.003  allowed alternative VMT
       distributions by vehicle class and roadway to be used. The current default MOBILE6 uses
       the same roadway VMT distribution for all classes. However, the user is allowed to enter
       alternate distributions.

A.3    The data parameters for Pull Ahead have been modified as a result of this new version. The
       vehicles are now assumed to be at a 6.00 g/bhp-hr standard for 2002 and 2003 model years.

A.4    Rebuild was previously assumed to reduce an engine from off-cycle emission levels down
       to 7.00 g/bhp-hr. This limit has now been lowered from 7.00 g/bhp-hr to 6.00 g/bhp-hr.

A. 5    Off-cycle operation was assumed to end in the 2000 calendar year. It has now been extended
       to the 2001 calendar year.

A.6    8B engines were originally assumed to be rebuilt after 3  years of use. This has now been
       raised to 5 years of use.

-------
                                     Appendix B
                        Response of Stakeholder Comments

       Only one set of significant stakeholder comments were received for this document.  These
comments were provided by David Piech of Navistar Corporation. In this section, the stakeholder
comment will be paraphrased and a brief reply or explanation provided. For a complete record of
the comment please see the MOBILE6 stakeholder comments archive.

1.      The stakeholder claims that the draft version of this document incorrectly states that the
       excess emissions are from  'defeat devices'.  Instead he recommends the phrase 'alleged
       defeat devices' to describe the source of the excess emissions.

       EPA agrees that our original nomenclature is in error. Consequently, the term' defeat device'
       was replace with the term 'off-cycle operation' throughout the document.  If time permits,
       it may also be eliminated from the MOBILE6 input command structure.

2.      Navistar knows of no 'official' EPA heavy-duty defeat device inventory emission results
       prepared by OECA.

       EPA agrees and the word 'official' has been dropped.

3.      The report fails to identify the source of the alleged defeat device activation estimates.
       Neither the report nor associated spreadsheets provide any background information as to the
       magnitude of the emissions associated with the disputed device nor the time or miles spent
       in such modes.

       EPA agrees that this information is the cornerstone of the NOX off-cycle emission factor
       preparation,  and that this  information is not  available in  the  report.   However,   the
       information in question is proprietary to specific companies (contains potential trade secrets),
       and it cannot be published by EPA in this document.

4.      The report inaccurately incorporates the defeat device prohibition, and provides an incorrect
       definition of a defeat device.

       EPA agrees that the definition as stated is not stated accurately. It has been removed from
       the text, and a new one has been substituted.
5.      EPA incorrectly  assumed that the disputed device had widespread utilization on most
       medium and light duty engines.

       This statement has been changed. A more accurate statement would read 'virtually no light
       heavy duty engines and few medium duty engines contained the disputed device.

-------
6.      EPA failed to incorporate reductions in correction factors (CF) since 1995.

       The correction factors which were used  are  the  standard correction  factors  used in
       MOBILE6.  These correction factors can be found in EPA report M6.HDE.004  'Update
       Heavy-Duty Engine Emission Conversion Factors for MOBILE6 Analysis of BSFCs and
       Calculation of Heavy-Duty Engine Emission Conversion Factors'

7.      EPA's disputed device penetration for medium duty engines in 1998, 1999, and 2000 is
       wrong.

       The 100% value  for disputed device penetration in these years in the spreadsheet is to
       activate the pull ahead requirements. EPA is aware that the fleet penetration of the disputed
       devices is not 100% and this has been corrected.

8.      The report's use of VMT data is unclear

       The MOBILE6 model contains its own values for VMT distribution, and allows specific
       user input data. The built-in data have been updated for MOBILE6, and may not be exactly
       the same as those used in the report

9.      The report fails to account for the speed relationship between the disputed devices and
       emissions.

       No good data were available on which to develop a speed relationship between the disputed
       devices  Thus, it was assumed that the off-cycle emissions already contained the effects of
       speed.

12.     The reports assumptions on the rebuild program are wrong.

       This is probably correct. An assumption that 90% of fleet will be rebuilt within a few years
       is probably not realistic. However, this number was provided by OECA.  The MOBILE6
       model also allows for a user supplied value that can range from 0 to 100%. The report also
       assumes that consent degree emission factor will be a NTE standard of 6.0 bhp-hr, and that
       the composite vehicle will emit on average at 6.0 bhp-hr. This is fairly pessimistic because
       if the NTE is 6.0 then the average may be lower.

-------
c
c
c
c
c
c
c
c
c
c
c
c
c.
c
c
c
c
c
c
c
c
c
c
c
c

c

c
     SUBROUTINE DEFEAT(MY,ICY,IV,IROAD,DDNOX)


 Changes: (Last change first)


13 Feb 01 @EPA-elg    Added bus category to IVV determination.
08 Feb 01 @EPA-elg
01 May 00 @EPA- BG
22 Dec 99 @Dyntel-nh
07 Dec 99 @DynTel-NH
25 Oct 99 @DynTel-NH
                   Changed logic for IMY mapping for the NO PULL AHEAD
                   case and removed RETURN. Added logic to include
                   IRBLD,  IRBSTRT and IRBEND to rebuild calculation.
                   Removed use of DATA statement initializations
                   1-004 removed diagnostic parameters and references.
                   1-004  added subroutine DEFEAT_STAT
                   1-004  New for the HDDV defeat task
.Rebuild modeling options using variable IBUILD.
   IBUILD =0  -  Default option - Rebuild effects are included at a
                                   90% penetration for MY 1994-98
   IBUILD = 1  -  No Rebuild effects are included.
   IBUILD =2  -  Rebuild effects are included for MY 1994-98 at a
                  user specified level  (RBLEV).


                  Default option - Pull ahead effects are included for
                  MY 2002 and 2003. Negative emission increases.
                  Pull ahead effects are not included for MY 2002 and
                  2003.  Defeat device effects are equal to MY 2001 effects.
IPULL = 0


IPULL = 1





 USE DDDATA


 IMPLICIT NONE
      INTEGER,  INTENT(IN)

      INTEGER,  INTENT(IN)

      INTEGER,  INTENT(IN)

      INTEGER,  INTENT(IN)
                           MY

                           ICY

                           IV

                           I ROAD
C


C


C

C
    INTEGER ::  IMY, IVV


    REAL,  INTENT(OUT)  :: DDNOX


    REAL ::  DDNOXX
                                               !  Local variables
                                          !  Local variables
      DDNOX=0.0
      IF(MY.LT.1988.OR.MY.GT.2003.OR.ICY.LT.1988.OR.
     *   IROAD.EQ.4)  THEN

        DDNOX =0.0                             ! Out of Range: No defeat device
effects
        RETURN
      END IF
C
      IF(MY.GE.2004)  THEN

         IMY = 1

      ELSEIF(MY.LE.1987)  THEN

         IMY =18

      ELSE

         IMY = 2004 - MY + 1

      END IF


      IF((MY.EQ.2002.OR.MY.EQ.2003).AND.IPULL.EQ.1) THEN

        IMY = 2004 -  2001  +  1       !  Always takes 2001  value  for no pull ahead.

      END IF
      IF(IV.GE.16.AND.IV.LE.19)  THEN !was OR

         IVV = 1

-------
      ELSEIF(IV.EQ.20.OR.IV.EQ.21) THEN
         IVV = 2
      ELSEIF(IV.EQ.22)  THEN
         IVV = 3
      ELSEIF(IV.EQ.23)  THEN
         IVV = 4
      ELSEIF(IV.EQ.26.OR.IV.EQ.27) THEN
         IVV = 5
      END IF
C
C     The base NOx excess does not contain rebuild  effects.
C
      DDNOXX=NEFFCT(IMY,IVV)
C
C     Effect of rebuilds:
C     No rebuilds can occur before calendar year  2000  because
C     the defeat device litigation had not yet been concluded.
C
      IF(ICY.GT.2000) THEN
        IF(MY.GE.IRBSTRT.OR.MY.LE.IRBEND) THEN
          IF(ICY.GT.MY+IRBLD(IVV)) THEN
            IF(IBUILD.EQ.2)  THEN
              DDNOXX = RBLEV *
     *                ((RBLD(IMY,IVV)-0.10*NEFFCT(IMY,IVV))/O.90)
     *                +  (1-RBLEV)*NEFFCT(IMY,IVV)
            ELSEIF(IBUILD.EQ.1) THEN
              DDNOXX = NEFFCT(IMY,IVV)
            ELSE
              DDNOXX = RBLD(IMY,IVV)
            END IF
          END IF
        END IF
      END IF
C
C    Effects are adjusted for defeat device activation by roadway and percentage
C     of defeat device engines in the fleet.
C
      IF(IROAD.EQ.1)  THEN
         DDNOX = DDNOXX * FLEET(IMY,IVV)  * URBAN(IMY,IVV)
      ELSEIF(IROAD.EQ.2)  THEN
         DDNOX = DDNOXX * FLEET(IMY,IVV)  * ARTERL(IMY,IVV)
      ELSEIF(IROAD.EQ.3)  THEN
         DDNOX = DDNOXX * FLEET(IMY,IVV)  * EXPRES(IMY,IVV)
      END IF
C
C
      RETURN
      END

      SUBROUTINE RBEFFCTS(INPREC,INERR,RC)
C
C Changes last first
C
C 18 Oct  1999  @Duntel-nh. New subroutine added  to  set DDDATA module  variable
IBUILD
C                         when  "REBUILD EFFECTS"  label  or  "NO REBUILD"  label is
C                         found in input file.
C
C Called by DSPTCH
C
C Calls NXTTOK
C
C Argument List
C

-------
  CHARACTER*(*)  INPREC   record from input file
  INTEGER       INERR
  INTEGER       RC
  MODULE DDDATA variables
  INTEGER
  REAL
IBUILD
RBLEV
  Local variables

  CHARACTER*10  STR1  Return string from NXTTOK
  INTEGER       PTR1  Returned index from NXTTOK
  INTEGER       PTR2  INDEX for search in NXTTOK
  REAL          DELTA Used for REAL xero comparisons
      USE DDDATA
      IMPLICIT NONE
      CHARACTER*(*) INPREC
      INTEGER INERR
      INTEGER RC
      CHARACTER*10 STR1
      INTEGER PTR1,PTR2
      REAL, PARAMETER  :
         DELTA=0.000001
      IF(INDEX(INPREC,'REBUILD EFFECTS')>0) THEN
        IBUILD=2
        PTR2=21
        CALL NXTTOK(INPREC,STR1,PTR1,PTR2)
        READ(STR1,'(F4.0)')RBLEV
        IF ((ABS(RBLEV).LT.DELTA).OR.(RBLEV.GT.0.90)) THEN
           CALL QUITER(0.0,0,191,INERR)
        END IF
      ELSEIF(INDEX(INPREC,'NO REBUILD')>0)THEN
        IBUILD=1
      ELSEIF(INDEX(INPREC,'NO DEFEAT')>0)THEN
        DEFEAT_FLAG=.FALSE.
      END IF
      RC = 0
      RETURN
      END SUBROUTINE

      SUBROUTINE NOXPUL(RC)
C
C CHANGES LAST FIRST
C
C 07 Jul 00 AIR Task 05: Removed  unused INPREC and  INERR  from parameter  list.
C 18 OCT  1999  ODYNTEL-NH. NEW SUBROUTINE  ADDED  TO  SET DDDATA MODULE  VARIABLE
I PULL
C
C CALLED BY DSPTCH
C
C ARGUMENT LIST
C
C INTEGER       RC
C
C MODULE DDDATA VARIABLES
C
C INTEGER       IPULL
C REAL          RBLEV
C
      USE DDDATA
      IMPLICIT NONE
      INTEGER RC
      IPULL=1
          WHEN "NO NOX PULL AHEAD" LABEL FOUND  IN  INPUT  FILE

-------
      RC=0
      RETURN
      END SUBROUTINE

      SUBROUTINE INIDEFEAT()
C
C CALLED BY REINIT TO RESET DEFAULT VALUES  FOR  EACH  RUN.
C MODULE DDDATA VARIABLES
C
C INTEGER       IPULL
C INTEGER       IBUILD
C REAL          RBLEV
      USE DDDATA
      IMPLICIT NONE
      IPULL=0
      IBUILD=0
      RBLEV=0.90
      DEFEAT_FLAG=.TRUE.
      FIRST_DEFEAT=.TRUE.
      RETURN
      END SUBROUTINE

      LOGICAL FUNCTION DEFEAT_DEVICE()
      USE DDDATA
      IMPLICIT NONE
      DEFEAT_DEVICE=DEFEAT_FLAG
      END

      SUBROUTINE DEFEAT_STAT(IOUREP)
C
C
C CHANGES LAST FIRST
C
C 22 DEC 99 ODYNTEL-NH. INCREASED VERBOSITY OF  MESSAGES  TO  REPORT FILE.
C 08-DEC-1999 ODYNTEL-NH.  NEW SUBROUTINE ADDED  TO  REPORT DEFEAT OPTIONS.
C
C CALLED BY HRLOOP TO REPORT DEFEAT OPTIONS TO  REPORT  FILE
C
C MODULE DDDATA VARIABLES
C
C INTEGER       IPULL
C INTEGER       IBUILD
C REAL          RBLEV
C
C ARGUMENT LIST
C
C INTEGER IOUREP LOGICAL DEVICE NUMBER FOR  REPORT  FILE
C
C OUTPUTS
C
C STATUS TO REPORT FILE
C
      USE DDDATA
      IMPLICIT NONE
      INTEGER IOUREP
      IF (IPULL.EQ.l) THEN
         WRITE(IOUREP,
     C    '(" NO HDDV DEFEAT DEVICE PULL AHEAD  EFFECTS ARE  PRESENT.")')
      END IF
      SELECT CASE (IBUILD)
          CASE (1)
            WRITE(IOUREP,
     C       '( " NO HDDV DEFEAT DEVICE REBUILD PROGRAM  IN  EFFECT."  )')
          CASE (2)

-------
       WRITE(IOUREP,'(  " HDDV DEFEAT  DEVICE  EFFECTS ARE PRESENT.
C             "THE REBUILD  FRACTION  IS  ",F4.2,"."  )')RBLEV
     CASE DEFAULT
 END SELECT
 IF (.NOT. DEFEAT_FLAG) THEN
    WRITE (IOUREP,
C    '(" NO HDDV DEFEAT DEVICE  EFFECTS  ARE PRESENT.")')
 END IF
 END SUBROUTINE

-------
      MODULE DDDATA
C
C
C  Changes: (Last change first)
C
C 13 Feb 01 @EPA-elg    Added bus category to each array.
C 13 Feb 01 @EPA-elg    Changed Pull Ahead Emission Delta to reflect
C                       Changes in the underlying emission factors.
C 09 Feb 01 @EPA-djb    Updated IRBLD, NEFFCT,  RBLD & FLEET values.
C                       Added IRBSTRT and IRBEND variables.
C 01 May 00 @EPA- BG    Removed use of DATA statement initializations
C 28 Apr 00 @EPA-elg    Corrected 8A Vehicle FLEET block data.
C 06 Dec 99 @DynTel-bg  1-004  defined logical  variable FIRST_DEFEAT
C 19 Oct 99 @DynTel-NH  1-004  New for the HDDV defeat task
C
C  Dimension 18 reflects the MY's and 4 reflects the vehicle classes.
C  In RBLD - 7 reflects the MY's and 4 reflects the vehicle classes.
C
C  Values for NEFFCT are the increase in NOx emission levels
C  as the result of the heavy-duty NOx defeat device.
C
C  Light duty includes all diesel except classes 6,7,8a and 8b.
C
C..First entry is for model year 2004.  Last entry is
C  model year 1987 and earlier.   Negatives reflect Pull ahead effects.
C  Value of Zero is assigned if no Pull Ahead or effect.  There are no
C  effects for model years 2004+ or 1987 and earlier.
C
C..URBAN is the fraction of the time that the defeat device operates
C  when an equipped vehicle is operating in Urban type driving.
C
C..ARTERL is the fraction of the time that the  defeat device operates
C  when an equipped vehicle is operating in Arterial type driving.
C
C..EXPRES is the fraction of the time that the  defeat device operates
C  when an equipped vehicle is operating in Expressway type driving.
C
C..FLEET is the fraction of the fleet that is equipped with a defeat
C  device.
C
C..First year is 1999+ and last is 1993 and earlier.
C  Rebuilds only in 1994 through 1998 MYs for Plan A.
C..7.00 g/bhp-hr is the rebuild emission target for HDDSa and HDDSb.
C..6.00 g/bhp-hr is the rebuild emission target for MDDV.
C  Rebuild effect is the difference between the value and 7.00  or the
C  value and 6.00.
C
C..IRBLD is the age of a vehicle when the first rebuild is done.
C  parameters are available for the four vehicle classes.
C  These parameters are currently implicit in the results;  thus,
C  the array IRBLD is not currently used in Defeat.
C
C..IRBSTRT is the first model year where defeat device effects  are
C  rebuilt. Rebuilds are affected by IRBLD parameters.
C
C..IRBEND is the last model year where defeat device effects are rebuilt.
C
C
      IMPLICIT NONE
      SAVE
C
      INTEGER, PARAMETER ::  NCY =18
      INTEGER, PARAMETER ::  NVCLS = 5

-------
C
C
C




C




C




C




C




C
C

C




C




C




C




C




REAL
INTEGER
INTEGER
LOGICAL
LOGICAL
INTEGER, DIMENSION (NVCLS)
INTEGER
INTEGER
REAL, DIMENSION(18, 5) ::
LIGHT DIESELS
*0.0000, -1.1400,
*0.0000, 0.0000,
*0.0000, 0.0000,
*0.0000, 0.0000,
MEDIUM DIESELS (CLS
* 0.0000, -1.5700
* 1.3533, 3.3623
* 2.9136, 2.4900
* 0.0680, 0.7890
8A HEAVY-DUTIES
* 0.0000, -1.5600
* 2.3200, 8.2377
* 7.1260, 6.9118
* 8.8713, 8.3125
8B HEAVY-DUTIES
* 0.0000, -1.5600
* 2.3200, 8.2617
* 7.1022, 6.8980
* 8.8179, 7.8961
DIESEL BUSES
*0.0000, -1.1400,
*0.0000, 0.0000,
*0.0000, 0.0000,
*0.0000, 0.0000,


REAL, DIMENSION (
LIGHT DIESEL
*0.0000, -1.6300,
*0.0000, 0.0000,
*0.0000, 0.0000,
*0.0000, 0.0000,
MEDIUM DIESELS (CLS
* 0.0000, -1.8500
* 1.3533, 2.3152
* 1.4424, 1.4000
* 0.0680, 0.7890
8A HEAVY-DUTIES
* 0.0000, -1.8400,
* 2.3200, 3.8118
* 2.8636, 2.8422
* 8.8713, 8.3125
8B HEAVY-DUTIES
* 0.0000, -1.8400
* 2.3200, 3.8142
* 2.8612, 2.8408
* 8.8179, 7.8961
BUSES
*0.0000, -1.6300,
*0.0000, 0.0000,
*0.0000, 0.0000,
*0.0000, 0.0000,

-0
0
0
0
6
,-o
, 2
, 2
, 1

,-o
, 7
, 7
, 8

,-o
, 7
, 7
, 8

-0
0
0
0


18,

-0
0
0
0
6
,-o
, 1
, 2
, 1

-0
, 2
, 7
, 8

,-o
, 2
, 7
, 8

-0
0
0
0

.2850,
.0000,
.0000,
.0000,
& 7)
.3925,
.7163,
.5700,
.5000,

.3900,
.1143,
.4985,
.9200,

.3900,
.2067,
.3978,
.9200,

.2850,
.0000,
.0000,
.0000,


5) : :

.4075,
.0000,
.0000,
.0000,
& 7)
.4625,
.4226,
.5700,
.5000,

.4600,
.8624,
.4985,
.9200,

.4600,
.8717,
.3978,
.9200,

.4075,
.0000,
.0000,
.0000,
RBLEV=0 . 90
IBUILD=0
IPULL=0
DEFEAT FLAG
FIRST DEFEAT
IRBLD=(/ 99, 12, 5, 5, 99
IRBSTRT=1994
IRBEND=1998
NEFFCT=RESHAPE ( (/

0
0
0
0

0
2
0
1

2
6
7
8

2
6
7
8

0
0
0
0



0000,
0000,
0000,
0000,

0000,
7408,
0680,
5000,

3200,
9407,
9025,
9200,

3200,
9501,
7489,
9200,

0000,
0000,
0000,
0000,



0


0

0


0

2


0

2


0

0


0



000,


000,

0000,


0000,

3200,


0000,

3200,


0000,

000,


OOO/) , (/18,5/) )


RBLD=RESHAPE ( (/

0
0
0
0

0
1
0
1

2
2
7
8

2
2
7
8

0
0
0
0

0000,
0000,
0000,
0000,

0000,
4251,
0680,
5000,

3200,
8451,
9025,
9200,

3200,
8460,
7489,
9200,

0000,
0000,
0000,
0000,

0


0

0


0

2


0

2


0

0


0

000,


000,

0000,


0000,

3200,


0000,

3200,


0000,

000,


OOO/) , (/18,5/) )

-------
c
c
c
c
c
c
      REAL, DIMENSION(18
   LIGHT DIESEL
     *0.0000,  1.0000,
     *0.0000,  0.0000,
     *0.0000,  0.0000,
     *0.0000,  0.0000,
   MEDIUM DIESELS  (CLS 6
     * 0.0000,  1.0000,
     * 0.0009,  0.0009,
     * 0.0692,  0.1200,
     * 0.0000,  0.0000,
   8A HEAVY-DUTIES
     * 0.0000,  1.0000,
     * 0.0210,  0.0210,
     * 0.0468,  0.0341,
     * 0.0295,  0.0612,
   8B HEAVY-DUTIES
     * 0.0000,  1.0000,
     * 0.0267,  0.0267,
     * 0.0487,  0.0362,
     * 0.0363,  0.0753,
   DIESEL BUSES
     *0.0000,  1.0000,
     *0.0000,  0.0000,
     *0.0000,  0.0000,
     *0.0000,  0.0000,
                     ,5)
       URBAN=RESHAPE((/
                     1.0000, 0.0000, 0.000,
                     0.0000, 0.0000,
                     0.0000, 0.0000,
                     0.0000, 0.0000, 0.000,
                      & 7)
                      1.0000, 0.0009, 0.0009,
                     0.0042, 0.0049,
                     0.1200, 0.0000,
                     0.0000, 0.0000,  0.0000,

                      1.0000, 0.0210, 0.0210,
                     0.0083, 0.0107,
                     0.0271, 0.0188,
                     0.0000, 0.0000,  0.0000,

                      1.0000, 0.0267, 0.0267,
                     0.0116, 0.0115,
                     0.0299, 0.0216,
                     0.0000, 0.0000,  0.0000,

                     1.0000, 0.0000, 0.000,
                     0.0000, 0.0000,
                     0.0000, 0.0000,
                     0.0000, 0.0000, O.OOO/),(/18,5/))
      REAL, DIMENSION
   LIGHT DIESEL
     * 0.0000,
     * 0.0000,
     * 0.0000,
     * 0.0000,
   MEDIUM DIESELS
     * 0.0000,
     * 0.2000,  0.2000
     * 0.3516,  0.6100
     * 0.0000,  0.0000
   8A HEAVY-DUTIES
     * 0.0000,  1.0000
     * 0.3651,  0.3651
     * 0.4022,  0.3704
     * 0.5095,  0.5093
   8B HEAVY-DUTIES
     * 0.0000,  1.0000
     * 0.3949,
     * 0.4130,
     * 0.5188,
   DIESEL BUSES
     * 0.0000,
     * 0.0000,
     * 0.0000,
     * 0.0000,
1.0000,
0.0000,
0.0000,
0.0000,
S (CLS 6
1.0000,
1.0000,
0.0000,
0.0000,
0.0000,
& 7)
1.0000,
0.0000
0.0000
0.0000
0.0000

0.2000
             0.3949,
             0.3825,
             0.4810,

             1.0000,
             0.0000,
             0.0000,
             0.0000,
                   18,5)  :: ARTERL=RESHAPE((/
                                      0.000,
               ,  0.000,

               ,  0.2000,
0.0385,  0.0247,
0.6100,  0.0000,
0.0000,  0.0000,  0.0000,

 1.0000,  0.3651,  0.3651,
0.3592,  0.3399,
0.4080,  0.4392,
0.6600,  0.6600,  0.0000,

 1.0000,  0.3949,  0.3949,
0.3881,  0.3511,
0.4103,  0.4384,
0.6600,  0.6600,  0.0000,

 1.0000,  0.0000,  0.000,
 0.0000,  0.0000,
 0.0000,  0.0000,
 0.0000,  0.0000,  O.OOO/),(/18,5/)
   REAL, DIMENSION(18,5)
       EXPRES=RESHAPE((/
LIGHT DIESEL
  * 0.0000,  1.0000,
  * 0.0000,  0.0000,
  * 0.0000,  0.0000,
  * 0.0000,  0.0000,
                         1.0000,
                         0.0000,
                         0.0000,
         0.0000,
         0.0000,
         0.0000,
0.000,
                         0.0000, 0.0000,  0.000,
C  MEDIUM DIESELS  (CLS 6 & 7)

-------
     * 0.0000,  1.0000,
     * 0.8220,  0.8220,
     * 0.9249,  0.9800,
     * 0.0000,  0.0000,
C  8A HEAVY-DUTIES
     * 0.0000,  1.0000,
     * 0.9225,  0.9225,
     * 0.9439,  0.9357,
     * 0.9023,  0.8925,
C  8B HEAVY-DUTIES
     * 0.0000,  1.0000,
     * 0.9377,  0.9377,
     * 0.9533,  0.9478,
     * 0.9064,  0.8756,
C  DIESEL BUSES
     * 0.0000,  1.0000,
     * 0.0000,  0.0000,
     * 0.0000,  0.0000,
     * 0.0000,  0.0000,
C
C
      REAL, DIMENSION(18
C  LIGHT DIESELS
     * 0.0000,  0.0000,
     * 0.0000,  0.0000,
     * 0.0000,  0.0000,
     * 0.0000,  0.0000,
C  MEDIUM DIESELS  (CLS 6
     * 0.0000,  0.4000,
     * 1.0000,  1.0000,
     * 0.0592,  0.0120,
     * 0.0000,  0.0000,
C  8A HEAVY-DUTIES
     * 0.0000,  1.0000,
     * 1.0000,  1.0000,
     * 1.0000,  1.0000,
     * 0.2900,  0.2110,
C  8B HEAVY-DUTIES
     * 0.0000,  1.0000,
     * 1.0000,  1.0000,
     * 1.0000,  1.0000,
     * 0.4680,  0.3400,
C  DIESEL BUSES
     * 0.0000,  0.7500,
     * 0.0000,  0.0000,
     * 0.0000,  0.0000,
     * 0.0000,  0.0000,
C
      END MODULE DDDATA
 1.0000,  0.8220, 0.8220,
0.5026,  0.5396,
0.9800,  0.0000,
0.0000,  0.0000,  0.0000,

 1.0000,  0.9225, 0.9225,
0.9154,  0.9153,
0.9510,  0.9568,
0.9800,  0.9800,  0.0000,

 1.0000,  0.9377, 0.9377,
0.9373,  0.9285,
0.9625,  0.9697,
0.9800,  0.9800,  0.0000,

 1.0000,  0.0000, 0.000,
 0.0000,  0.0000,
 0.0000,  0.0000,
 0.0000,  0.0000, O.OOO/) ,  (/18,5/)
,5)  ::  FLEET=RESHAPE((/

 0.0000,  0.0000, 0.0000,
 0.0000,  0.0000,
 0.0000,  0.0000,
 0.0000,  0.0000, 0.0000,
 & 7)
 0.4000,  1.0000, 1.0000,
 0.4950,  0.2470,
 0.0060,  0.0000,
 0.0000,  0.0000, 0.0000,

 1.0000,  1.0000, 1.0000,
 1.0000,  1.0000,
 0.7590,  0.5780,
 0.1860,  0.1300, 0.0000,

 1.0000,  1.0000, 1.0000,
 1.0000,  1.0000,
 1.0000,  0.9940,
 0.1510,  0.0880, 0.0000,

 0.7500,  0.0000, 0.0000,
 0.0000,  0.0000,
 0.0000,  0.0000,
 0.0000,  0.0000, O.OOOO/),(/18,5/))

-------