L>EPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington, DC 20460
EPA/600/R-92/245
February 1993
Life-Cycle
Assessment:
Inventory Guidelines
and Principles
-------
-------
EPA/600/R-92/245
February 1993
LIFE-CYCLE ASSESSMENT:
INVENTORY GUIDELINES AND PRINCIPLES
by
B. W. Vigon, D. A. Tolle, B. W. Cornaby, and H. C Latham
Battelle
Columbus, Ohio 43201-2693
and
C. L. Harrison, T. L Boguski, R. G. Hunt and J. D. Sellers
Franklin Associates, Ltd.
Prairie Village, Kansas 66208
Contract No. 68-CO-0003
Project Officer
Mary Ann Curran
Pollution Prevention Research Branch
Risk Reduction Engineering Laboratory
Cincinnati, Ohio 45268
This study was conducted in cooperation with the
Office of Air Quality Planning and Standards,
the Office of Solid Waste,
and the Office of Pollution Prevention and Toxics.
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
Printed on Recycled Paper
-------
DISCLAIMER
The information, in this document has heen funded wholly by the United States Environmental
Protection Agency (EPA) under Contract No. 68-CO-0003 to Battelle. It has been subjected to
peer and administrative review, and it has been approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute endorsement or recommen-
dation for use. Use of this methodology does not imply EPA approval of the conclusions of any
specific life-cycle inventory.
II
-------
FOREWORD
Today's rapidly developing and changing technologies and industrial products and practices
frequently carry with them the increased generation of materials that, if improperly dealt with,
can threaten both public health and the environment. The U.S. Environmental Protection
Agency (EPA) is charged by Congress with protecting the Nation's land, air, and water
resources. Under a mandate of national environmental laws, the Agency strives to formulate
and implement actions leading to a compatible balance between human activities and the abil-
ity of natural systems to support and nurture life. These laws direct the EPA to perform
research to define our environmental problems, measure the impacts, and search for solutions.
The Risk Reduction Engineering Laboratory is responsible for planning, implementing, and
managing research, development, and demonstration programs to provide an authoritative,
defensible engineering basis in support of the policies, programs, and regulations of the EPA
with respect to drinking water, wastewater, pesticides, toxic substances, solid and hazardous
wastes, and Superfund-related activities. This publication is one of the products of that re-
search and provides a vital communication link between the researcher and the user
community.
This document is written in a manner to be useful to a broad audience. This audience includes
organizations currently conducting studies, those intending to conduct such studies, and those
interpreting studies done by other organizations. By providing a template for generalizing the
inventory development process and describing a set of rules to assist in making necessary
assumptions regarding, for example, assessment boundaries, data quality and coverage, and
equivalency of use in a consistent fashion, the guide should reduce the tendency for studies to
be published with apparently contradictory conclusions. The added methodological structure
should also aid organizations in reading, evaluating, and applying the results of inventories by
articulating desired quality milestones.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
HI
-------
ABSTRACT
This document describes the three components of a life-cycle assessment (inventory analysis,
impact analysis, and improvement analysis) as well as scoping activities, presents a brief over-
view of the development of the life-cycle assessment process, and develops guidelines and prin-
ciples for implementation of a product life-cycle assessment. The major stages in a life cycle
are raw materials acquisition, manufacturing, consumer use/reuse/maintenance, and recycle/
waste management. The basic steps of performing a life-cycle inventory (defining the goals and
system boundaries, including scoping; gathering and developing data; presenting and reviewing
data; and interpreting and communicating results) are presented along with the general issues
to be addressed. The system boundaries, assumptions, and conventions to be addressed in each
stage of the inventory are presented. Life-cycle impact analysis and life-cycle improvement
analysis will be topics of forthcoming guidance documents. ;
This report was submitted by Battelle in fulfillment of Contract No. 68-CO-0003 under the
sponsorship of the U.S. Environmental Protection Agency. Technical effort leading to this
report covers a period from August 1990 to May 1991. The draft report was completed in Sep-
tember 1991. Following a comment period, this final report was prepared between March and
November 1992. !
hr
-------
CONTENTS
Page
FOREWORD Hi
ABSTRACT jv
ACKNOWLEDGMENTS .;...,. x
SUMMARY AND INDEX OF GUIDING STATEMENTS AND KEY PRINCIPLES xi
CHAPTER ONE
INTRODUCTION 1
CHAPTER TWO
OVERVIEW. 4
Life-Cycle Assessment Concept ..4
A Brief History of Life-Cycle Inventory Analysis .....5
Overview of Life-Cycle Assessment Methodology 7
Three Components 7
Identifying and Setting Boundaries for Life-Cycle Stages 9
Issues That Apply to All Stages 10
Applications of an Inventory Analysis ^ 10
Format of this Report 12
CHAPTER THREE
PROCEDURAL FRAMEWORK FOR LIFE-CYCLE INVENTORY , 13
Introduction 13
Define the Purpose and Scope of the Inventory 14
Private Sector Uses 14
e
Public Sector Uses ; 14
Study Specificity 15
-------
Contents
maram
; Page
Define the System Boundaries T 16
Devise an Inventory Checklist 23
Institute a Peer Review Process 26
Gather Data ? 27
I i
Identify Subsystems 28
Sources of Data -.'. -29
Develop Stand-Alone Data '. » - 32
Construct a Computational Model -33
!
Present the Results 36
Interpret and Communicate the Results . ? 37
i
CHAPTER FOUR
GENERAL ISSUES IN PERFORMING A LIFE-CYCLE INVENTORY 40
Introduction 40
Using Templates in Life-Cycle Inventory Analysis -40
Inputs in the Product Life-Cycle Inventory Analysis 43
Outputs of the Product Life-Cycle Inventory Analysis.... 51
Transportation i - -54
Coproduct Allocation i : 56
Data Issues 59
Data Quality 59
Data Time Period , 64
Specific Data versus Composite Data ». 64
Geographic Specificity » ; M
t
Technology Mixes/Energy Types 65
Data Categories ......65
-------
Contents
Page
Routine/Fugitive/Accidental Releases 65
Special Case Boundary Issues.... 65
Capital Equipment . 66
Personnel Issues 66
Improper Waste Disposal 66
CHAPTER FIVE ,
ISSUES APPLICABLE TO SPECIFIC LIFE-CYCLE STAGES .67
introduction. 67
Raw Materials Acquisition Stage 67
Subsystem Boundaries 67
Specific Assumptions and Conventions 70
Manufacturing Stage , .-. . 74
Materials Manufacture Step 75
Subsystem Boundaries 75
Specific Assumptions and Conventions 75
Product Fabrication Step -. .'.77
Subsystem Boundaries 77
Specific Assumptions and Conventions 79
Filling/Packaging/Distribution Step 79
Subsystem Boundaries .79
Specific Assumptions and Conventions 80
Use/Reuse/Maintenance Stage 82
Subsystem Boundaries 82
Specific Assumptions and Conventions 82
via
-------
Contents
i
Page
Recycle/Waste Management Stage -84
Subsystem Boundaries \ -85
Specific Assumptions and Conventions J 85
REFERENCES ;. - : M
GLOSSARY 96
i
APPENDIX : - 104
\
Clean Air Act (CAA) : 104
Clean Water Act (CWA) -.105
Comprehensive Environmental Response, Compensation
and Liability Act (CERCLA) ,.- 106
Resource Conservation and Recovery Act (RCRA) ; 106
Toxic Substances Control Act (TSCA) -108
viii
-------
, Contents
LIST OF FIGURES
Page
Figure 1 Defining system boundaries ............... . [[[ ; ................. ... ....... \j
Figure 2 Steps in the manufacturing stage ................. . ...... ..... ..... . ................. . ...................... . .......... 13
Figure 3 Example system flow diagram for bar soap .. ......... . [[[ 20
Figure 4 A typical checklist with worksheet for performing a life-cycle inventory .............. 24
Figure 5 Example coproduct allocation based on relative weight ................................ . ............ ; 34
Figure 6 Life-cycle inventory template ... ............................... . [[[ . 41
Figure 7 Detailed system flow diagram for bar soap ........................ . .............................. .; ........... 42
Figures Allocating resources and environmental burdens for a product
and coproduct .......... . .................... ...................... . ................................... ..... .............. _ 55
Figure 9 The first step in a product's life cycle is the acquisition of raw materials
and energy [[[ . [[[ .......68
Figure 1 0 Materials manufacture converts raw materials into a form usable
in a finished product [[[ 75
Figure 1 1 Product fabrication converts intermediate materials into a finished product ... ...... 78
Figure 12 Filling, packaging, and distribution is the final step in manufacturing
before a product reaches the consumer ................ ..; [[[ 81
Figure 1 3 Consumer use/reuse/maintenance is the third stage in a product's life cycle ........... 83
Figure 14 Recycle/waste management is 4he last stage in a product's life cycle .............. . ........ 86
-------
ACKNOWLEDGMENTS
This document was prepared for the U.S. Environmental Protection Agency (EPA) Office of
Research and Development (ORD), Office of Air Quality Planning and Standards (OAQPS),
Office of Solid Waste (OSW), and Office of Pollution Prevention and Toxics (OPPT). Mary
Ann Curran of ORD, Risk Reduction Engineering Laboratory (RREL), Cincinnati, Ohio,
served as project officer. Additional EPA guidance, reviews, and comments were provided
by David C. Fege (OPPT), Eun-Sook Goidel (OPPT), Michael Flynn (OSW), Paul Kaldjian
(OSW), Lynda Wynn (OSW), Eugene Lee (OSW), Timothy Mohin (OAQPS), Timothy Ream
(OAQPS), and Anne Robertson (RREL). The technical work was conducted under Battelle/
EPA Contract No. 68-CO-0003 by Battelle with Bruce Vigon as work assignment leader. Karl
Nehring and Steve Pomeroy provided technical contributions. Technical review was also
provided by Sid Everett (SRI International). Vincent Brown and Lynn Copley-Graves edited
the document and Diane Holbrook coordinated publication.
Peer reviewers included Bob Berkebile, American Institute of Architects; Joel Charm,
Allied-Signal, Inc.; Frank Consoli, Scott Paper; Michelle Crew, New Jersey Department of
Environmental Protection; Gary Davis, University of Tennessee; Norman Dean, Green Seal,
Inc.; Richard Denison, Environmental Defense Fund; Michael Harrass, U.S. Food and Drug
Administration; Greg Keoleian, University of Michigan; John Kusz, Industrial Design Society
of America; Gail Mayville, Ben & Jerry's; Beth Quay, Coca-Cola U.S.A.; Derek Augood, Sci-
entific Certification Systems, Inc.; T. Michael Rothgeb and Charles Pittinger, Procter &
Gamble; Jacinthe Seguin, Environment Canada; Karen Shapiro, Tellus Institute; William W.
Walton, U.S. Consumer Product Safety Commission; Matt Weinberg, Office of Technology
and Assessment; and Jeanne Wirka, Environmental Action Foundation. ,
Views contained in this document may not necessarily reflect those of individual reviewers.
-------
SUMMARY AND INDEX OF
GUIDING STATEMENTS
AND KEY PRINCIPLES
This summary combines the key principles and recommendations from within the hody
of the report into one list for quick reference. Page numbers indicate the location of
each item in the text.
Page
CHAPTER ONE
Life-cycle assessment is a concept to evaluate the environmental effects associated with
any given activity from the initial gathering of raw material from the earth until the point
at which all residuals are returned to the earth .1
Life-cycle assessment consists of three complementary components: inventory analysis,
impact analysis, and improvement analysis 2
The newest aspects of life-cycle assessment are goal definition and scoping analysis which
serve to tailor the scope and boundaries to the study ......2
CHAPTER TWO
Life-cycle inventories may be used both internally by organizations to support decisions
in implementing improvements and externally to inform decisions, with external
applications requiring a higher standard of accountability. ..4
Life-cycle inventory analyses can be used in process analysis, material selection, product
evaluation, product comparison, and policy-making 4
Life-cycle assessment is not necessarily a linear or step wise process. Information from
any one component can complement that from the others 5
For internal applications, scoping may be an informal process undertaken by project
staff, whereas external applications may require establishment of a multi-organization
group and a formal scoping procedure g
To calculate the total results for the entire life cycle, energy and solid waste mass or
volumes can be summed. Other releases should not be summed, except in specific
circumstances of business confidentiality 10
XI
-------
SUMMARY AND INDEX OF GUIDING STATEMENTS
Page
CHAPTER THREE
Clear definitions of the purposes and boundaries of a life-cycle inventory analysis
help ensure that the results will be validly interpreted. 14
The specificity of a life-cycle inventory may range from completely generic to entirely
product-specific, with most studies falling somewhere in between. Data collection and
results interpretation are strongly influenced by study specificity j 15
In life-cycle inventory analysis, the term "system" refers to a collection of operations
that together perform some defined function i...... 16
A broad-based life-cycle inventory begins with raw materials and continues through
final disposition, accounting for every significant step in a product system 16
I
Understanding the consequences of narrowing study boundaries is important for
evaluating tradeoffs between the ability of the inventory to address the environmental
attributes and cost, time, or other factors , 19
In a comparative life-cycle inventory, the basis of comparison should be equivalent
usage. Each system should be defined so that a functionally equal amount of product
or equivalent service is delivered to the consumer , ...19
A general rule for excluding a step from a system is that in all alternatives compared,
that step is exactly the same in process, materials, and quantity '. '..22
Equivalent usage for comparative studies can often be based on volume or weight 23
The level of required detail should be evaluated in light of available funding and the
size of the system while maintaining the technical integrity of the study;. 23
After system boundaries are determined, a system flow diagram is developed which
depicts every operation contributing to the system function ;. 23
In studies intended for external application, the system flow diagram is often
incorporated into a formal scope, boundary, and data collection document ...23
An inventory checklist is used to guide data collection and validation and
to enable construction of the computational model 23
A peer review process, implemented early in a study, is recommended to address
scope/boundary methodology, data acquisition and compilation, validation of
assumptions and results, and communication of results in any study to be used in
a public forum ..' , 26
Each subsystem requires input of materials and energy; requires transportation of
product produced; and has outputs of products, coproducts/by-products, solid waste,
atmospheric emissions, waterborne wastes, and other releases 28
xil
-------
SUMMARY AND INDEX OF GUIDING STATEMENTS
Page
Trim or off-spec materials reused within a process are considered part of an
internal recycling loop and are not included in the inventory, because they do
not cross the boundaries of the subsystem ...28
Many data sources are available to use in inventories. Well-characterized data from
industry are best for production processes 29
When collecting data (and later, when reporting the results), protection of
confidential business information should be balanced with the need for full and
detailed analysis or disclosure. 30
For data on recycling rates and recycled material, consumers and processors may be
helpful. Trade associations and consumers of the recycled materials can also provide
data 32
A standard unit of output must be determined for each subsystem. All data could be
reported in terms of the number of pounds, kilograms, or tons of subsystem product 32
When the data are at a consistent level, the energy and material requirements and
the environmental releases are attributed to the production of each coproduct, using
a technique called coproduct allocation 32
Once the inputs and outputs of each subsystem have been allocated, numerical
relationships of the subsystems within the entire system, flow diagram can be
established. 33
The overall system flow diagram is important in constructing the model because it
numerically defines the relationships of the individual subsystems to each other in
the production of the final product 33
Sensitivity analyses of key elements in the system should be performed to estimate
the effect of uncertainties in the system. 35
The report should explicitly define the systems analyzed and the boundaries. All
assumptions made should be explained, the basis for comparison among systems
given, and equivalent usage ratios explained 36
Both graphical and tabular results presentation are valuable in data interpretation;
however, oversimplification is not appropriate 36
CHAPTER FOUR
Templates, or material and energy balance diagrams, are guides used to direct the
gathering and developing of data. 40
xiii
-------
SUMMARY AND INDEX OF GUIDING STATEMENTS
Page
A template may be applied directly to a subsystem without evaluating the processes
or subprocesses within the subsystem, depending on the nature of the subsystem and
the availability of data 41
Materials required in small amounts, or that use energy or resources incrementally at
less than the variability in the total system requirements, may be excluded based on
the application of defined decision rules. . 43
Designations of natural resources into renewability categories should be considered
part of impact analysis 46
Energy data should be given as specific quantities of fuel used 46
Precombustion energy is defined as the total amount of energy necessary to deliver a
usable fuel to the consumer of the fuel 47
The inventory should characterize energy requirements according to basic sources of
energy and should indicate whether precombustion energy is included or not 47
Models currently in use for analyzing the U.S. electrical energy system are based on
the fuel mix in regional or national grids.: ... 48
Because of the energy inefficiencies of the system, a conversion of 11.3 MJ per kWh
is often used to reflect the actual use of fuel to deliver electricity to the consumer 48
For products or materials that consume raw materials whose alternative use is as a
fuel or energy resource, an energy value is assigned equivalent to the fossil fuel
combustion value of petroleum and natural gas 49
Several options, including the tracking of nonfuel inherent energy as a separate category,
are available for ensuring that all energy demands are accounted for 49
Where data related to procurement of energy from a foreign source are not available,
the approach has been to apply domestic data as an initial estimate j 50
The energy value of combustible materials is credited against the system energy use
when waste-to-energy recovery is practiced ^50
Consumptive usage is the fraction of total water withdrawal from surface or *
groundwater sources which either incorporated into the product, coproducts (if any),
or wastes or evaporated .....50
Environmental releases are actual discharges (after control devices) of pollutants or
other materials from a process or operation ,
Solid waste includes solid material and certain liquids or gases that are
disposed of
from all sources within the system. These wastes are typically reported by weight
and converted to volume using representative landfill density factors.
xhr
.51
.52
-------
SUMMARY AND INDEX OF GUIDING STATEMENTS
Page
Distinction is made between industrial and postconsumer solid wastes, as they are
generally disposed of in different ways and, in some cases, at different facilities 53
Separate accounting of hazardous and nonhazardous solid, waste is useful, especially
if an impact analysis is to be conducted . 53
The efficiency of each mode of transport is used to convert the units of ton-miles or
tonne-kilometers into fuel units. The fuel units are then converted to energy units. 54
Transportation is reported only for the product of interest and not for any coproducts,
because the destination of the coproducts is not an issue .54
A coproduct must be recovered as a predominant industrial activity, not merely be
potentially recoverable , 56
Coproducts are of interest only to the point where they no longer affect the primary
product , 56
Weight-proportioned coproduct allocation is used in most instances. Where a technical
argument may be made for a different procedure, its justification should be provided 56
If industrial scrap is to be collected and used as a material input to a production system
or process, then it is credited as a coproduct at the point where it was produced ....58
When consumption of a coproduct falls within the boundaries of the analysis, it must
no longer be considered a coproduct, but a primary product carrying with it energy
requirements and environmental releases. 59
If an estimate of true variability is known, it should form the basis for high- and low-
range sensitivity analysis estimates 61
Order-of-magnitude estimates may be used to decide which input values require
specification to a higher level of accuracy and which may be left "as is." 61
The individual variability of each variable and simultaneous variability of only the few
most critical variables should be considered 61
Actual data should be used rather than estimates or regulatory limits 61
Data should be collected at as detailed a level as possible, which allows for better
analysis and reporting, and all emissions should be recorded at the same time 62
Where a variety of technologies exist, it may be more appropriate to assume the missing
data are equal to the quantity averaged over only the plants reporting, assign the value to
the missing data, and then average the total 62
When the purpose of the inventory is for internal operations improvement, it is best to
use data specific to the system that is being examined 64
xv
-------
SUMMARY AND INDEX OF GUIDING STATEMENTS
I Page
Composite or average data are preferable when inventory results are to be used for
broad application across the industry, particularly in studies performed for public use 64
Data time periods should be long enough to smooth out any deviations or variations in
normal operations. A fiscal year of production is usually sufficient , 64
When no specific geographical data exist, practices that occur in other countries are
typically assumed to be the same as their domestic counterparts . ....64
Estimates can be made for accidental emissions based on historical data, pertaining to
frequency and concentrations of accidental emissions i, 65
Energy and emissions involved with capital equipment can be excluded when the
manufacture of the unit of interest accounts for a minor fraction of the total output
produced over the life of the equipment 66
i
In many situations, personnel consequences are very small and would probably occur
whether or not the product is manufactured 1 66
CHAPTER FIVE
Resource requirements and environmental emissions are calculated for all processes
involved in acquiring raw materials and energy by tracing materials and energy back to
then: sources I .....67
When traditional fuels are used as raw materials, they are assigned an inherent energy
value equal to the heat of combustion of the raw material because the fuels have been
removed from the total fuel supply 70
Energy requirements of a system are not reduced or credited for the use of
"renewable" resources instead of "nonrenewable" resources '. 71
It may be necessary to create a separate category of nonfuel inherent energy 71
i
Overburden from mining operations and wood left in forests due to the harvesting of
trees is not considered to be a solid waste, but land use changes may be quantified 71
Environmental emissions from drilling operations are allocated between crude oil
and natural gas production based on historical production data 71
When analyzing the use of an animal product, the feed is usually considered to be the
main raw material 72
Whenever a specific fuel is used in any of the processing or transportation steps, the
appropriate quantities of precombustion energy and emissions are included in the total
energy and emissions attributed to the use of that fuel i 73
xwi
-------
SUMMARY AND INDEX OF GUIDING STATEMENTS
Page
The basic energy sources included in a life-cycle analysis are coal, petroleum, natural
gas, nuclear power, hydropower, and wood 73
The national average fuel mix for the electric utilities is representative when used for all
of the manufacturing steps in an analysis, except for activities in industries where
geographically specific sources are used 74
Material scrap from a subsystem can be reused internally, sold as industrial scrap, or
disposed of as solid waste. The inventory account for each option is handled differently.. 77
Filling and packaging products ensure that products remain intact until they are ready
for use, whereas distribution transfers the products from the manufacturer to the
consumer .80
In addition to primary packaging, some products require secondary and tertiary
packaging, all of which may need to be accounted for in a life-cycle inventory 80
Energy and emissions associated with the filling of a product (bottle or package) can
be ignored when the products being compared use the same basic rilling procedures
and equipment 80
Any special circumstances in transportation, such as refrigeration used to keep a
product fresh, should be considered in the inventory .80
An average distance for product transportation must be developed 80
Household operations are rarely allocated to an individual product 82
The collection and transportation of discarded materials for the various waste
management options should be included.. * . 85
Life-cycle inventory techniques adjust all resource requirements and emissions for
products that are recycled or contain recycled content 87
In closed-loop recycling, products are reused again and again in the same product 87
In open-loop recycling, products are recycled into new products that are eventually
disposed. > 87
When composting is considered in a life-cycle inventory, the solid waste for a given
material is reduced 91
Many products release energy when burned in an incinerator, reducing the net
reported energy requirements 93
xwil
-------
-------
Chapter One
INTRODUCTION
The concept of life-cycle assessment is to
evaluate the environmental effects associated
with any given activity from the initial gath-
ering of raw material from the earth until the
point at which all residuals are returned to
the earth. This concept, often referred to as
"cradle to grave" assessment, is not new.
While the practice of conducting life-cycle
studies has existed for more than 20 years,
there has been no comprehensive attempt to
describe the procedure in a manner that
would facilitate understanding of the overall
process, the underlying data, and the inher-
ent assumptions. The literature contains few
published assessments and even fewer peer-
reviewed publications describing the techni-
cal basis for life-cycle assessments. The
Society for Environmental Toxicology and
Chemistry (SETAC) life-cycle assessment
technical framework workshop report pub-
lished in January 1991 summarizes the cur-
rent status of the field and outlines the
technical basis for life-cycle studies. The pur-
pose of this U.S. Environmental Protection
Agency inventory guidelines and principles
document is to provide guidance on the spe-
cific details involved in the conduct of life-
cycle studies.
Some of the most promising applications of
life-cycle assessment are for internal use by
corporations and regulatory agencies. By
developing and using information regarding
environmental effects that are both
"upstream" and "downstream" of the particu-
lar activity under scrutiny, a new paradigm is
created for basing decisions in both corporate
management and regulatory policy-making.
Recently, there has been a sharp increase in
the number of groups conducting life-cycle
assessments. Often, the results of these stud-
ies have been used to support public claims
about various products or processes. Predict-
ably, the results of these studies are often in
conflict, and somewhat dependent on the
group sponsoring the study. With the increas-
ing use of this information to gain a competi-
tive advantage in the marketplace, there is a
clear need for neutral, scientifically oriented,
consensus-based guidelines on the conduct of
life-cycle assessment.
The EPA has initiated a project to develop
such guidelines. The project involves a multi-
office EPA group devoted to addressing meth-
odological issues concerning life-cycle
assessments. This core group consists of rep-
resentatives from the Office of Research and
Development, Office of Solid Waste, Office of
Air Quality Planning and Standards, and
Office of Pollution Prevention and Toxics.
This inventory guidelines document is the
first in a series on conducting life-cycle
assessment studies. Additional documents
will follow as the knowledge and under-
standing of life-cycle assessment evolves.
Near-term efforts include the preparation of
documents that provide guidance on the
-------
Introduction
impact analysis component, on data availabil-
ity, and on data quality issues for life-cycle
assessments. Improvement analysis and guid-
ance for streamlining life-cycle studies are
potential future products of this core group.
The EPA's life-cycle assessment project
includes using a consensus-building
approach and working in close coordination
with SETAC. As a scientific and professional
society, SETAC has provided infrastructure,
credibility, resources, and technical expertise
to the development of life-cycle concepts
both in the United States and internationally.
Through the organization of a series of work-
shops, SETAC has overseen the development
of an emerging technical framework for the
conduct of life-cycle assessment.
Based on discussions at the 1990 SETAC
workshop, life-cycle assessment consists of
three components: inventory analysis, impact
analysis, and improvement analysis. This
document is intended to be a practical guide
to conducting and interpreting life-cycle
inventory analysis, which consists of an
accounting of the resource usage and envi-
ronmental releases associated with a product,
process, or activity throughout each stage of
its life cycle. Recently, the SETAC model has
been expanded to include an initial step of
goal definition and subsequent scoping analy-
sis. These newer aspects of the SETAC model
serve to tailor the scope and boundaries of
life-cycle studies to be appropriate with the
stated goals of the study. To the extent practi-
cable, this document incorporates the con-
cepts of goal definition and scoping as they
apply to life-cycle inventory analysis.
Recent SETAC activity also has begun to
define a conceptual framework for life-cycle
impact analysis. Preliminary findings of this
effort suggest that certain categories of
impacts may require expanded or modified
inventory data collection. To the extent that
these requirements can be anticipated, this
document incorporates the additional scope.
This document is not a "cookbook." Given
the range of application^, it is not feasible to
provide "recipes" for every situation that
could be encountered. Instead, this guide
attempts to provide a rationale for ensuring
internal consistency of procedures for both
data acquisition and calculation used in life-
cycle inventory analyses. This document
relies heavily on practices.that have been
used by some life-cycle practitioners and that
have evolved over many years. Certain deci-
sion rules in this guide are presented as spe-
cific recommendations because they have
proven to be practical over their years of use.
In other cases, where judgment is essential
regarding an assumption in the study, the
guide presents the relevant alternatives with
some of the associated advantages and disad-
vantages. There is full recognition within this
guide that the practice of life-cycle assess-
ment continues to evolve. This guidance
should be viewed as a starting point, captur-
ing a "snapshot" of the state of the science of
life-cycle inventory assessment. As the over-
all life-cycle assessment framework continues
to evolve, it is very likely that changes to the
inventory methods presented herein will be
necessary. '
Currently there is no single correct way to
conduct a life-cycle assessment. One clear
message of this document is that when a life-
cycle practitioner makes assumptions or
defines the boundary conditions of a life-
cycle study, these decisions must be transpar-
ent to the users of that study. In other words,
it is imperative for the credibility of the study
that the goals, scope, and all assumptions
-------
Introduction
inherent in any life-cycle study are clear to the
audience for that study. It is recommended
that all groups having a stake in the continued
development and application of life-cycle
assessment adopt the recommendations con-
tained hi this guide and fully disclose all of
the assumptions used hi the conduct of their
life-cycle studies.
The guidance manual consists of five chap-
ters. Chapter Two provide a methodology
overview, including the status of current
research and the basics of the life-cycle assess-
ment methodology. Readers familiar with the
concept may wish to skip this chapter. Subse-
quent chapters assume a considerable famil-
iarity with the terms and concepts used for
life-cycle studies. Chapter Three describes a
technical framework for conducting a life-
cycle inventory. Chapter Four discusses gen-
eral issues in performing a life-cycle
inventory. Chapter Five contains descriptions
and analyses of issues pertaining to the indi-
vidual stages and steps of a life-cycle inven-
tory: raw materials acquisition; manufacturing
(including materials manufacture, product
fabrication, and fillmg/packagihg/distribu-
tion); consumer use/reuse/maihtenance; and
recycle/waste management.
-------
Chapter Two
OVERVIEW
LIFE-CYCLE ASSESSMENT CONCEPT
Over the past 20 years, environmental
issues have gained greater public recogni-
tion. The general public has become more
aware that the consumption of manufac-
tured products and marketed services, as
well as the daily activities of our society,
adversely affect supplies of natural
Major Concepts
Life-cycle assessment is a tool to evaluate
the environmental consequences of a
product or activity holistically, across its
entire life.
There is a trend in many countries toward
more environmentally benign products
and processes.
A complete life-cyde assessment consists
of three complementary components:
Inventory, Impact, and Improvement
Analyses.
Life-cyde inventories can be used both
internally to an organization and exter-
nally, with external applications requiring
a higher standard of accountability.
Life-cycle inventory analyses can be used
; in process analysis, material selection,
product evaluation, product comparison,
and policy-making.
resources and the quality of the environ-
ment. These effects occur at all stages of the
life cycle of a product, beginning with raw
material acquisition and continuing
through materials manufacture and product
fabrication. They also occur during product
consumption and a variety of waste man-
agement options such as landfilling, incin-
eration, recycling, and composting. As
public concern has increased, both govern-
ment and industry have intensified the
development and application of methods to
identify and reduce the; ad verse environ-
mental effects of these activities.
Life-cycle inventory is a "snapshot" of
inputs to and outputs from a system. It can
be used as a technical tool to identify and
evaluate opportunities to reduce the envi-
ronmental effects associated with a specific
product, production process, package,
material, or activity. This tool can also be
used to evaluate the effects of resource
management options designed to create
sustainable systems. Life-cycle inventories
may be used both internally by organiza-
tions to support decisions in implementing
product, process, or activity improvements
and externally to inform consumer or pub-
lic policy decisions. External uses are
expected to meet a higher standard of
accountability in methodology application.
Life-cycle assessment adopts a holistic
approach by analyzing the entire life cycle
-------
Overview
of a product, process, package, material, or
activity. Life-cycle stages encompass
extraction and processing of raw materials;
manufacturing, transportation, and distri-
bution; use/reuse/maintenance; recycling
and composting; and final disposition. It is
not the intent of a life-cycle assessment to
analyze economic factors. A life-cycle
assessment can be used to create scenarios
upon which a cost analysis could be
performed.
The three separate but interrelated compo-
nents of a life-cycle assessment include
(1) the identification and quantification of
energy and resource use and environmental
releases to air, water, and land (inventory
analysis); (2) the technical qualitative and
quantitative characterization and assess-
ment of the consequences on the environ-
ment (impact analysis); and (3) the
evaluation and implementation of opportu-
nities to reduce environmental burdens
(improvement analysis). Some life-cycle as-
sessment practitioners have defined a
fourth component, the scoping and goal
definition or initiation step, which serves
to tailor the analysis to its intended use.
Life-cycle assessment is not necessarily a
linear or stepwise process. Rather, informa-
tion from any of the three components can
complement information from the other
two. Environmental benefits can be realized
from each component in the process. For
example, the inventory analysis alone may
be used to identify opportunities for reduc-
ing emissions, energy consumption, and
material use. The impact analysis addresses
ecological and human health consequences
and resource depletion, as well as other
effects, such as habitat alteration, that can-
not be analyzed in the inventory. Data defi-
nition and collection to support impact
analysis may occur as part of inventory
preparation. Improvement analysis helps
ensure that any potential reduction strate-
gies are optimized and that improvement
programs do not produce additional, unan-
ticipated adverse impacts to human health
and the environment. This guidance docu-
ment is concerned primarily with inventory
analyses.
A BRIEF HISTORY OF LIFE-CYCLE
INVENTORY ANALYSIS
Life-cycle inventory analysis had its begin-
nings in the 1960s. Concerns over the limi-
tations of raw materials and energy
resources sparked interest in finding ways
to cumulatively account for energy use and
to project future resource supplies and use.
In one of the first publications of its kind,
Harold Smith reported his calculation of
cumulative energy requirements for the
production of chemical intermediates and
products at the World Energy Conference in
1963.
Later in the 1960s, global modeling studies
published in The Limits to Growth (Mead-
ows et al., 1972) and A Blueprint for
A Life-Cycle Assessment Has
Three Components
These components overlap and build on
each other in the development of a
complete life-cyde assessment.
Inventory Analysis
Impact Analysis
Improvement Analysis
Scoping is an activity that initiates an
assessment, defining its purpose, bound-
aries, and procecures.
-------
Overview
Survival (Club of Rome) resulted in predic-
tions of the effects of the world's changing
population on the demand for finite raw
materials and energy resources. The predic-
tions of rapid depletion of fossil fuels and
climatological changes resulting from
excess waste heat stimulated more detailed
calculations of energy use and output in
industrial processes. During this period,
about a dozen studies were performed to
estimate costs and environmental implica-
tions of alternative sources of energy.
In 1969 researchers initiated a study for
The Coca-Cola Company that laid the foun1
dation for the current methods of life-cycle
inventory analysis in the United States. In a
comparison of different beverage containers
to determine which container had the low-
est releases to the environment and least
affected the supply of natural resources,
this study quantified the raw materials and
fuels used and the environmental loadings
from the manufacturing processes for each
container. Other companies in both the
United States and Europe performed simi-
lar comparative life-cycle inventory analy-
ses in the early 1970s. At this time, many of
the data were derived from publicly avail-
able sources such as government docu-
ments or technical papers, as specific
industrial data were not available.
The process of quantifying the resource use
and environmental releases of products
became known as a Resource and Environ-
mental Profile Analysis (REPA), as prac-
ticed in the United States. In Europe it was
called an Ecobalance. With the formation of
public interest groups encouraging industry
to ensure the accuracy of information in the
public domain, and with the oil shortages
in the early 1970s, approximately 15 REP As
were performed between 1970 and 1975.
Through this period, a protocol or standard
research methodology for conducting these
studies was developed. This multistep
methodology involves a number of assump-
tions. During these years, the assumptions
and techniques used underwent consider-
able review by EPA and major industry rep-
resentatives, with the result that reasonable
methodologies evolved.1
From 1975 through the early 1980s, as
interest in these comprehensive studies
waned because of the fading influence of
the oil crisis, environmental concern
shifted to issues of hazardous waste man-
agement. However, throughout this time,
life-cycle inventory analyses continued to
be conducted and the methodology
improved through a slow stream of about
two studies per year, most of which
focused on energy requirements. During
this time, European interest grew with the
establishment of an Environment Director-
ate (DG XI) by the European Commission.
European life-cycle assessment practition-
ers developed approaches parallel to those
being used in the USA. Besides working to
standardize pollution regulations through-
out Europe, DG Xl issued the Liquid Food
Container Directive in 1985, which charged
member companies with monitoring the
energy and raw materials consumption and
solid waste generation of liquid food
containers. :
When solid waste became a worldwide
issue in 1988, the life-cycle inventory
analysis technique again emerged as a tool
for analyzing environmental problems. As
interest in all areas affecting resources and
the environment grows,;the methodology
for life-cycle inventory analysis is again be-
ing improved. A broad base of consultants
and research institutes in North America
-------
Overview
and Europe have been further refining and
expanding the methodology. With recent
emphasis on recycling and composting
resources found hi the solid waste stream,
approaches for incorporating these waste
management options into the life-cycle
inventory analysis have been developed.
Interest in moving beyond the inventory to
analyzing the impacts of environmental
resource requirements and emissions brings
life-cycle assessment methods to another
point of evolution.
During the past 2 years, the Society of EmnU
ronmental Toxicology and Chemistry
(SETAC) has served as a focal point for
technical developments in the life-cycle
assessment arena. Workshops on the over-
all technical framework, impact analysis,
and data quality were held to allow consen-
sus building on methodology and accept-
able professional practice. Public forums
and a newsletter have provided additional
opportunity for input from the user
community.
Over the past 20 years, most life-cycle
inventories have examined different forms
of product packaging such as beverage con-
tainers, food containers, fast-food packag-
ing, and shipping containers. Many of these
inventories have supported efforts to
reduce the amount of packaging in the
waste stream or to reduce the environ-
mental emissions of producing the
packaging.
Some studies have looked at actual con-
sumer products, such as diapers and deter-
gents, while others have compared
alternative industrial processes for the
manufacture of the same product.
OVERVIEW OF LIFE-CYCLE
ASSESSMENT METHODOLOGY
Three Components
Inventory Analysis
The inventory analysis component is a
technical, data-based process of quantifying
energy and raw material requirements,
atmospheric emissions, waterborne emis-
sions, solid wastes, and other releases for
the entire life cycle of a product, package,
process, material, or activity. Qualitative
aspects are best captured in the impact
analysis, although it could be useful during
the inventory to identify these issues, hi the
broadest sense, inventory analysis begins
with raw material extraction and continues
through final product consumption and
disposal. Some inventories may have more
restricted boundaries because of their
intended use (e.g., internal industrial prod-
uct formulation improvements where raw
materials are identified). Inventory analysis
is the only component of life-cycle analysis
that is well developed. Its methodology has
been evolving over a 20-year period.
Refinement and enhancement continue to
occur following the SETAC workshop in
1990. Chapters Three through Five present
the current framework, assumptions, and
steps in inventory analysis.
Impact Analysis
The impact analysis component is a techni-
cal, quantitative, and/or qualitative process
to characterize and assess the effects of the
resource requirements and environmental
loadings (atmospheric and waterborne
emissions and solid wastes) identified in
the inventory stage. Methods for impact
analysis are in the early stage of develop-
ment following a SET AC workshop in early
-------
Overview
1992. The analysis should address both
ecological and human health impacts,
resource depletion, and possibly social wel-
fare. Other effects, such as habitat modifica-
tion and heat and noise pollution that are
not easily amenable to the quantification
demanded in the inventory, are also part of
the impact analysis component.
The key concept in the impact analysis
component is that of stressors. The stressor
concept links the inventory and impact
analysis by associated resource consump-
tion and releases documented in the inven-
tory with potential impacts. Thus, a
stressor is a set of conditions that may lead
to an impact. For example, a typical inven-
tory will quantify the amount of SO2
released per product unit, which then may
produce acid rain and which in turn might
affect the acidification in a lake. The result-
ant acidification might change the species
composition to eventually create a loss of
biodiversity.
An important distinction exists between
life-cycle impact analysis and other types
of impact analysis. Life-cycle impact analy-
sis does not necessarily attempt to quantify
any specific actual impacts associated with
a product or process. Instead, it seeks to
establish a linkage between the product or
process life cycle and potential impacts.
The principal methodological issue is man-
aging the increased complexity as the
stressor-impact sequence is extended.
Methods for analysis of some types of
impacts exist, but research is needed for
others.
Improvement Analysis
The improvement analysis component of
the life-cycle assessment is a systematic
evaluation of the needs and opportunities
to reduce the environmental burden associ-
ated with energy and raw material use and
waste emissions throughout the life cycle of
a product, process, or activity. This analy-
sis may include both quantitative and qual-
itative measures of improvements. This
component has not been widely discussed
hi a public forum.
Scoping or Initiation
Scoping is one of the first activities in any
life-cycle assessment and is considered by
some practitioners as a fourth component.
During scoping, the product, process, or
activity is defined for the context in which
the assessment is being made. The scoping
process links the goal of the analysis with
the extent, or scope, of the study, i.e., what
will or will not be included. For some
applications, an impact analysis will be
desired or essential. In these cases, the pre-
paration of the inventory is not a stand-
alone activity. The scoping process will
need to reflect the intent to define and col-
lect the additional inventory data for the
impact analysis.
i
For internal life-cycle inventories, scoping
may be done informally by project staff.
Scoping for external studies may require
the establishment of a multi-organization
group and a formal procedure for reviewing
the study boundaries and methodology.
Although scoping is part of life-cycle analy-
sis initiation, there may be valid reasons for
reevaluating the scope periodically during
a study. As the life-cycle inventory model
is defined or as data are collected, scope
modifications may be necessary.
-------
Overview
Identifying and Setting
Boundaries for Life-Cycle Stages
The quality of a life-cycle inventory depends
on an accurate description of the system to be
analyzed. The necessary data collection and
interpretation is contingent on proper under-
standing of where each stage of a life cycle
begins and ends.
General Scope of Each Stage
Raw Materials Acquisition
This stage of the life cycle of a product
includes the removal of raw materials and
energy sources from the earth, such as the
harvesting of trees or the extraction of crude
oil. .Transport of the raw materials from the
point of acquisition to the point of raw
materials processing is also considered part
of this stage.
Manufacturing
The manufacturing stage produces the prod-
uct or package from the raw materials and
delivers it to consumers. Three substages or
steps are involved hi this transformation:
materials manufecture, product fabrication,
and filling/packaging/distribution.
Materials Manufacture. This step involves
converting a raw material into a form that can
be used to fabricate a finished product. For
example, several manufacturing activities are
required to produce a polyethylene resin from
crude oil: The crude oil must be refined;
ethylene must be produced in an olefins plant
and then polymerized to produce polyethyl-
ene; transportation between manufacturing
activities and to the point of product fabrica-
tion is considered part of materials
manufacture.
Product Fabrication. This step involves pro-
cessing the manufactured material to create a
product ready to be filled or packaged, for
example, blow molding a bottle, forming an
aluminum can, or producing a cloth diaper.
Filling/Packaging/Distribution.This step
includes all manufacturing processes and
transportation required to fill, package, and
distribute a finished product. Energy and
environmental wastes caused by transporting
the product to retail outlets or to the consumer
are accounted for in this step of a product's
life cycle.
Use/Reuse/Maintenance
This is the stage consumers are most famil-
iar with, the actual use, reuse, and mainte-
nance of the product. Energy requirements
and environmental wastes associated with
product storage and consumption are
included in this stage.
Recycle/Waste Management
Energy requirements and environmental
wastes associated with product disposition
are included in this stage, as well as post-
consumer waste management options such
as recycling, composting, and incineration.
Stages of a Life Cycle
Raw Materials Acquisition
Manufacturing
Materials Manufacture
Product Fabrication
Filling/Packaging/Distribution
Use/Reuse/Maintenance
Recycle/Waste Management
-------
Overview
Issues That Apply to All Stages
The following general issues apply across
all four life-cycle stages:
Energy and Transportation
Process and transportation energy require-
ments are determined for each stage of a
product's life cycle. Some products are
made from raw materials, such as crude oil,
which are also used as sources of fuel. Use
of these raw materials as inputs to products
represents a decision to forego their fuel
value. The energy value of such raw materi-
als that are incorporated into products typi-
cally is included as part of the energy
requirements in an inventory analysis.
Energy required to acquire and process the
fuels burned for process and transportation
use is also included.
Environmental Waste Aspects
Three categories of environmental wastes
are generated from each stage of a product's
life cycle: atmospheric emissions, water-
borne wastes, and solid wastes. These envi-
ronmental wastes are generated by both the
actual manufacturing processes and the use
of fuels in transport vehicles or process
operations.
Waste Management Practices
Depending on the nature of the product, a
variety of waste management alternatives
may be considered: landfilling, incineration,
recycling, and composting.
Allocation of Waste or Energy Among Primary
and Goproducts
Some processes in a product's life cycle
may produce more than one product, hi this
event, all energy and resources entering a
particular process and all wastes resulting
from it are allocated among the product and
coproducts. Allocation is most commonly
based on the mass ratios of the products, but
there are exceptions to this.
Summing the Results of Each Stage
To calculate the total results for the entire
life cycle of a particular [product, the energy
and certain emission values for each stage of
the product's life cycle can be summed. For
example, energy requirements for each stage
are converted from fuel units to million .
Btus or megajoules and summed to find the
total energy requirements. Solid wastes may
be summed in pounds or converted to vol-
ume and summed. The current, preferred
practice is to present the individual envi-
ronmental releases into each of the environ-
mental media on a pollutant-by-pollutant
basis. Where such specificity in an external
study would reveal confidential business
information, exceptions should be made on
a case-by-case basis. Claims for confidential-
ity should be made only when it is reason-
able to expect that release of the information
would damage the supplier's competitive
position. Even then, the data inputs to an
external use are legitimately expected to be
independently verified. A peer review pro-
cess leading to agreed-upon reporting is one
possible mechanism for dealing with this
issue.-Other approaches for independent
verification are possible.
APPLICATIONS OF
AN INVENTORY ANALYSES
An inventory conforming to the scope
defined hi this document will provide a
quantitative catalog of ekiergy and other
resource requirements, atmospheric emis-
sions, waterborne emissions, and solid
wastes for a specific product, process, pack-
age, material, or activity. Once an inventory
10
-------
Overview
has been performed and is deemed as accu-
rate as possible within the defined scope
and boundaries of the system, the results
can be used directly to identify areas of
greater or lesser environmental burden, to
support a subsequent life-cycle impact
analysis, and as part of a preliminary
improvements analysis. Life-cycle impact
assessment can be applied to quantify the
human and ecological health consequences
associated with specific pollutants identi-
fied by the inventory.
The following are possible applications for
life-cycle inventories. These are organized
according to whether the application is sup-
portable with the inventory alone or
whether some level of additional impact
analysis is appropriate. The critical issue
that users should keep in mind is that if an
application context results only in an inven-
tory, the resulting information must not be
over-interpreted. Inventories can be applied
internally to an organization or externally to
convey information outside of the sponsor-
ing organization. External uses are broadly
denned in this document to include any
study where results will be presented or
used beyond the boundaries of the sponsor-
ing organization. Most applications will
require some level of impact analysis in
addition to the inventory.
To Support Broad Environmental Assessments
The results of an inventory are valuable in
understanding the relative environmental
burdens resulting from evolutionary
changes in given processes, products, or
packaging over time; in understanding the
relative environmental burdens between
alternative processes or materials used to
make, distribute, or use the same product;
and in comparing the environmental aspects
of alternative products that serve the same
use.
To Establish Baseline Information
A key application of a life-cycle inventory is
to establish a baseline of information on an
entire system given current or predicted
practices in the manufacture, use, and dis-
position of the product or category of prod-
ucts. In some cases it may suffice to
establish a baseline for certain processes
associated with a product or package. This
baseline would consist of the energy and
resource requirements and the environ-
mental loadings from the product or process
systems analyzed. This baseline information
is valuable for initiating improvement
analysis by applying specific changes to the
baseline system.
To Rank the Relative Contribution of Individual
Steps or Processes
The inventory provides detailed data
regarding the individual contributions of
each step hi the system studied to the total
system. The data can provide direction to
efforts for change by showing which steps
require the most energy or other resources,
or which steps contribute the most pollu-
tants. This application is especially relevant
for internal industry studies to support deci-
sions on pollution prevention, resource con-
servation, and waste minimization
opportunities.
These first three applications are support-
able with the understanding that the inven-
tory data convey no information as to the
possible environmental consequences of the
resource use or releases. Any interpretation
beyond the "less is best" approach is
subjective.
-------
Overview
To Identify Data Gaps
The performance of life-cycle inventory
analyses for a particular system reveals
areas in which data for particular processes
or regarding current practices are lacking or
are of uncertain or questionable quality.
When the inventory is to he followed by an
impact analysis, this use can also identify
areas where data augmentation for the
impact analysis is appropriate.
To Support Policy
For the public policymaker, life-cycle
inventories and impact analyses can help
broaden the range of environmental issues
considered in developing regulations or set-
ting policies.
To Support Product Certification
Product certifications have tended to focus
on relatively few criteria. Life-cycle inven-
tories, only when augmented by appropriate
impact analyses, can provide information
on the individual, simultaneous effects of
many product attributes.
To Provide Education for Use in Decision-Making
Life-cycle inventories and impact analyses
can be used to educate industry, govern-
ment, and consumers on the tradeoffs of
alternative processes, products, materials,
and/or packages. The data can give industry
direction in decisions regarding production
materials and processes and create a better
informed public regarding environmental
issues and consumer choices.
These last three applications of life-cycle
inventories are the most prone to overinter-
pretation. This is partly due to their more.
probable use external to the performing
organization and partly due to their implicit
orientation towards assessing the environ-
mental consequences of a product or
process. ,
FORMAT OF THIS REPORT
The remainder of the report provides more
specific guidance and application examples
for those who perform or interpret life-cycle
inventories. Chapter Three presents the pro-
cedural framework for performing a life-
cycle inventory, defines the scope and
structure, and describes the construction of
the model, the collection and availability of
sources of data, and the presentation of
results. Chapter Four discusses issues com-
mon to all stages of life-cycle inventory
analysis. Chapter Five discusses in greater
detail issues pertinent to specific life-cycle
stages. This guide is not intended to be a
point by point prescription for the inventory
preparation process. Giyen the dynamic
nature of the science and the developing
methods for impact analysis, potential users
are provided with some degree of method-
ological flexibility while still maintaining
scientific integrity and transparency.
12
-------
Chapter Three
PROCEDURAL FRAMEWORK
FOR LIFE-CYCLE INVENTORY
INTRODUCTION
This chapter describes the procedural
framework for performing a life-cycle inven-
tory. Although there is broad scientific
agreement on the major elements of a life-
cycle inventory, procedural decisions occur
at many steps in the process. This guidance
document presents these decisions in the
form of a decision tree. The advantages and
limitations of each option are discussed.
Major Concepts
Clear definitions of the purposes and boundar-
ies of a life-cyde inventory analysis help ensure
valid interpretation of the results.
In life-cycle inventory analysis, the term "sys-
tem" refers to a collection of operations that
together perform some well-defined function.
A broad life-cyde inventory accounts for every
significant step in a product system.
System flow diagrams and calculations are used
to determine the resource requirements and
environmental emissions for a product.
Interpretation of results depends on boundary
conditions, the quality of data, and the assump-
tions used.
A peer review process for external application
inventories should be implemented early in the
study.
When possible, each option is presented as
typical or desired practice based on previ-
ous technical forums and scientific
soundness.
The inventory process begins with a con-
ceptual goal definition phase to define both
the purpose for performing the inventory
and the scope of the analysis. An inventory
procedure is then employed, and data on
the product or system are gathered. Next,
the data are incorporated into a computer
model to determine the results for the entire
system. Additional adjustments in system
boundaries and collection of data may be
necessary as a result of analyzing prelimi-
nary results. Finally, results are presented
and interpreted. Chapter Four discusses
more general issues related to performing an
inventory.
The remaining sections of this chapter
describe the steps of performing a life-cycle
inventory. They are as follows:
Define the Purpose and Scope of
the Inventory
Define the System Boundaries
Devise an Inventory Checklist
Institute a Peer Review Process
Gather Data
Develop Stand-Alone Data
-------
Procedural Framework
Construct a Computational Model
Present the Results
Interpret and Communicate the Results.
DEFINE THE PURPOSE AND
SCOPE OF THE INVENTORY
The decision to perform a life-cycle inven-
tory usually is based on one or several of the
following objectives:
To establish a baseline of information on
a system's overall resource use, energy
consumption, and environmental
loadings
To identify stages within the life-cycle of
a product or process where a reduction in
resource use and emissions might be
achieved
To compare the system inputs and out-
puts associated with alternative products,
processes, or activities
To help guide the development of new
products, processes, or activities toward a
net reduction of resource requirements
and emissions
To help identify areas to be addressed
during life-cycle impact analysis.
These objectives can be further categorized
into applications by alternative user groups
as listed below.
Private Sector Uses
Evaluation for Internal Decision Making
Compare alternative materials, products,
processes, or activities within the
organization.
Compare resource use and release inven-
tory information with'comparable infor-
mation on other manufacturers' products.
Train personnel responsible for reducing
the environmental burdens associated
with products, processes, and activities,
including product designers and
engineers. '
Provide the baseline information needed
to carry out other components of the life-
cycle assessment.
i
Evaluation for Public Disclosure of Information
Provide information to policymakers, pro-
fessional organizations, and the general
public on resource use and releases,
including appropriate disclosure and
documentation of findings.
Help substantiate product-related state-
ments of quantifiable reductions in
energy, raw materials, and environmental
releases, provided that information is not
selectively reported.
Public Sector Uses j
i
i
Evaluation and Policymakingi
Supply information for evaluating exist-
ing and prospective policies that affect
resource use and releases.
Develop policies and regulations on mate-
rials and resource use and environmental
releases when the inventory is supple-
mented by an impact analysis.
« Identify gaps in information and knowl-
edge, and help establish research priori-
ties and monitoring requirements on the
state and federal levels.
14
-------
Procedural Framework
Evaluate product statements of quantifi-
able reductions in energy, raw materials,
and environmental releases.
Public Education
Develop materials to help the public
understand resource use and release
characteristics associated with products,
processes, and activities.
Design curricula for training those
involved hi product, process, and activ-
ity design.
Key decisions in defining the scope and
boundaries of the inventory rest on the
defined goal or purpose of the inventory.
Conceptually, it is useful to distinguish
between the study boundaries and the sys-
tem boundaries. Study boundaries include
both the issues to be dealt with and the
physical system boundaries to be analyzed.
Study Specificity
At the outset of every study, the level of
specificity must be decided. In some cases,
this level will be obvious from the applica-
tion or intended use of the information. In
other instances, there may be several
options to choose among, ranging from a
completely generic study to one that is
product-specific in every detail. Most stud-
ies fall somewhere ifl between.
A life-cycle inventory can be envisioned as
a set of linked activities that describe the
creation, use, and ultimate disposition of
the product or material of interest. At each
life-cycle stage, the analyst should begin by
answering a series of questions: Is the prod-
uct or system in this life-cycle stage spe-
cific to one company or manufacturing.
operation? Or does the product or system
represent common products or systems
generally found hi the marketplace and
produced or used by a number of
companies?
Such questions help determine whether
data collected for the inventory should be
specific to one company or manufacturing
facility, or whether the data should be more
general to represent common industrial
practices.
The appropriate response to these ques-
tions often rests on whether the life-cycle
inventory is being performed for internal
organizational use or for a more public pur-
pose. Accessibility to product- or facility-
specific data may also be a factor. A
company may be more interested hi exam-
ining its own formulation and assembly
operations, whereas an industry group or
government agency may be more interested
in characterizing industry-wide practice.
Life-cycle inventories can have a mix of
product-specific and industry-average
information. For example, a cereal manu-
facturer performing an analysis of using
recycled paperboard for its cereal boxes
might apply the following logic. For opera-
tions conducted by the manufacturer, such
as box printing, setup, and filling, data spe-
cific to the product wo.uld be obtained
because average data for printing and filling
across the cereal industry or for industry in
general would not be as useful.
Stepping back one stage to package manu-
facturing, the cereal manufacturer is again
faced with the specificity decision. The
inventory could be product-specific, or
generic data for the package manufacturing
stage could be used. The product-specific
approach has these advantages: the aggre-
gated inventory data will reflect the opera-
tions of the specific papermills supplying
the recycled board, and the energy and
15
-------
Procedural Framework
resources associated with this stage can be
compared with those of similar specificity
for the filling, packaging, and distribution
stage. A limitation of this option is the addi-
tional cost and time associated with collect-
ing product-specific data from the mills and
the level of cooperation that needs to be
established with the upstream vendors.
Long-term confidentiality agreements with
vendors may also represent unacceptable
burdens compared with the value added by
the more specific data.
The alternative decision path, using indus-
trial average data for making recycled paper-
board, has a parallel mix of advantages and
limitations. Use of average, or generic, data
may be advantageous for a manufacturer
considering use of recycled board for which
no current vendors have been identified. If
the quality of these average data can be
determined and is acceptable, their use may
be preferable. The limitation is that data
from this stage may be less comparable to
that of more product-specific stages. This
limitation is especially important in studies
that mix product-specific and more general
analyses in the same life-cycle stage. For
example, comparing virgin and recycled
paperboard using product-specific data for
one material and generic data for the other
could be problematic.
Another limitation is that the generic data
may mask technologies that are more envi-
ronmentally burdensome. Even with some
measure of data variability, a decision to use
a particular material made on the basis of
generic data may misrepresent true loadings
of the actual suppliers. Opportunities to
identify specific facilities operating in a
more environmentally sound manner are
lost. Generic data do not necessarily repre-
sent industry-wide practices. The extent of
representation depends on the quality and
coverage of the available data and is impos-
sible to state as a general rule.
It is recommended that the level of specifi-
city be very clearly defined and communi-
cated so that readers are more able to
understand differences in the final results.
Before initiating data collection and periodi-
cally throughout the study, the analyst
should revisit the specificity decision to
determine if the approach selected for each
stage remains valid in view of the intended
use.
DEFINE THE SYSTEM BOUNDARIES
Once the goal or purpose for preparing a
life-cycle inventory has been determined
and the intended use is known, the system
should be specifically defined. A "system"
is a collection of operations that together
perform some clearly defined function. A
broad-based system begins with raw materi-
als acquisition and continues through
industrial or consumer use and final dispo-
sition. Great care should be taken in defin-
ing the systems to be analyzed and in
explaining the boundaries for the defini-
tions in any report of inventory results.
Clear definitions help ensure valid interpre-
tations of the results.
In defining the system, the first step is to set
the system boundaries. A complete life-
cycle inventory will set the boundaries of
the total system broadly to quantify resource
and energy use and environmental releases
throughout the entire life cycle of a product
or process, as shown in Figure 1. The life-
cycle stages used in this guidance document
differ in two respects from those presented
16
-------
Procedural Framework
in the SETAC Technical Framework (Fava
et al., 1991). These differences arise because
this guidance document is written from the
perspective of assembling information to
perform an inventory study. For this reason,
transportation activities used to move mate-
rials from one stage to the next have been
disaggregated rather than being presented as
a separate stage. Each transportation step is
associated with a specific upstream life-
cycle stage. Although it is common to pre-
sent transportation-related energy and
emissions separately in reporting results,
the transportation system type and the dis-
tances covered are defined within each
stage.
This model combines materials manufac-
ture, product fabrication, _and filling/
packaging/distribution in the manufacturing
stage. Separating these three aspects of man-
ufacturing into separate substages, or steps,
reflects the fact that different organizations
typically are involved in these activities.
The separate treatment also reflects the dif-
ferent nature of the operations and the deci-
sions discussed earlier regarding product,
material, or activity specificity. Figure 2
shows the three steps of manufacturing.
Recycling and waste management are com-
bined into one stage because, especially for
post-consumer material, recycling and
Life-Cycle Stages
Inputs
Raw
Materials
Energy
"" >""? f '
? ff
fcr
fc-
' -
Raw Materials Acquisition
V
"\
Manufacturing
i
f.
Use/Reuse/Maintenance
\
Recycle/Waste Management
, , ,
" *s
*r
1^
' 5 :
Outputs
Atmospheric
Emissions
Waterborne
Wastes
Solid
Wastes
Coproducts
Other
Releases
System Boundary
figure 1
Defining system boundaries
17
-------
Procedural Framework
Manufacturing
Materials Manufacture
~ *'.-;*,* ~
Product Fabrication
Filling/Packaging/Distribution
RgureZ
Steps in the manufacturing stage
waste management simply represent a split-
ting of the material flow between the two
streams. A life cycle, therefore, comprises
four major stages:
Raw Materials Acquisition Stage. All the
activities required to gather or obtain a
raw material or energy source from the
earth. This stage includes transportation
of the raw material to the point of mate-
rial manufacture, but does not include
material processing activities.
Manufacturing Stage. Encompasses three
steps:
- Materials Manufacture. The activities
required to process a raw material into a
form that can be used to fabricate a par-
ticular product or package. Normally, the
production of many intermediate chemi-
cals or materials is included in this cate-
gory. Transport of intermediate materials
is also included.
Product Fabrication. The process step
that uses raw or manufactured materials
to fabricate a product ready to be filled or
packaged. This step often involves a con-
sumer product that will be distributed
for retail sales, but the product could
also be distributed for use by other in-
dustries, i
Filling/Packaging/Distribution. Processes
that prepare the final products for ship-
ment and that transport the products to
retail outlets. Although these activities
may commonly require a change in the
location or physical configuration of a
product, they do not involve a transfor-
mation of materials.
Use/Reuse/Maintenance Stage. Begins
after the distribution of products or mate-
rials for intended use and includes any
activity in which the product or package
may be reconditioned, maintained, or
serviced to extend its useful life.
Recycle/Waste Management Stage.
Begins after the product, package, or
material has served its intended purpose
and either will enter a new system
through recycling or will enter the envi-
ronment through the waste management
system. ;
-------
Procedural Framework
Each step in the life cycle of a product,
package, or material can be categorized
within one and only one of these life-cycle
stages. Each step or process can be viewed
as a subsystem of the total product system.
Viewing the steps as subsystems facilitates
data gathering for the inventory of the
system as a whole. The boundaries of sub-
systems are defined by life-cycle stage cate-
gories in Chapter Five. The rest of this
chapter deals with defining the boundaries
of the whole product system. Many deci-
sions must be made in defining the specific
boundaries of each system.
Product systems are easier to define if the
sequence of operations associated with a
product or material is broken down into
primary and secondary categories. The pri-
mary, or zero-order, sequence of activities
directly contributes to making, using, or
disposing of the product or material. The
secondary category includes auxiliary
materials or processes that contribute to
making or doing something that in turn is
in the primary activity sequence. Several
tiers of auxiliary materials or processes may
extend further and further from the main
sequence. In setting system boundaries, the
analyst must decide where the analysis will
be limited and be very clear about the rea-
sons for the decision. The following ques-
tions are useful in setting and describing
specific system boundaries:
Does the analysis need to cover the entire
life cycle of the product? A theoretically
complete life-cycle system would start
with all raw materials and energy
sources in the earth and end with all
materials back in the earth or at least
somewhere in the environment but not
part of the system. Any system boundary
different from this represents a decision
by the analyst to limit it in some way.
Understanding the possible conse-
quences of such decisions is important
for evaluating tradeoffs between the abil-
ity of the resulting inventory to thor-
oughly address environmental attributes
of the product and constraints on cost,
time, or other factors that may argue in
favor of a more limited boundary. Too
limited a boundary may exclude conse-
quential activities or elements.
Depending on the goal of the study, it may
be possible to exclude certain stages or
activities and still address the issues for
which the life-cycle inventory is being per-
formed. For example, it may be possible to
exclude the acquisition of raw materials hi
a life-cycle inventory without affecting the
results. Suppose a company wishes to per-
form an internal life-cycle inventory to
evaluate alternative drying systems for
formulating a snack food product. If the
technologies are indifferent to feedstock, it
is possible to assume the raw materials
acquisition stage will be identical for all
options. If the decision will be based on
selecting a drying system with lower energy
use or environmental burdens, it may be
acceptable to analyze such a limited sys-
tem. However, with this system boundary,
the degree of absolute difference in the
overall system energy or environmental
inventory cannot be determined. The differ-
ence in the product manufacturing stage,
although significant for the manufacturer,
may represent a minor component of the
total system. Therefore, statements about
the total system should not be made.
What will be the basis of use for the
product or material? Is the study
intended to compare different product
systems? If the products or processes are
19
-------
Procedural Framework
Grain
Production
Cattle Raising
Salt
Mining
Meat Packing
Tallow Rendering
Caustic
Manufacture
Soap
Manufacturing
J
Forestry
Consumer
Postconsumer
Waste Management
Figure 3
Example system flow diagram for bar soap
1
Soap
Packaging ,
*
Paper
Production
used at different rates, packaged in vary-
ing quantities, or come in different sizes,
how can one accurately compare them?
Can equivalent use ratios be developed?
Should market shares be considered to
estimate the proportionate burden from
each product in a given category? Is the
study intended to compare service sys-
tems? Are the service functions clearly
defined so that the inputs and outputs are
properly proportioned?
What ancillary materials or
chemicals are used to make
or package the products or
run the processes? Might
these ancillary materials or
chemicals contribute more
than a minor fraction of the
energy or emissions of the
system to be analyzed?
How do they compare by
weight with other materials
and chemicals in the prod-
uct system?
In a comparative analysis,
are any extra products
required, to allow one prod-
uct to deliver equivalent or
similar performance to
another? Are any extra ma-
terials pr services required
for one service to be func-
tionally equivalent to
another or to a,comparable
product?
Figure 3 shows an example of
setting system boundaries for
a product baseline analysis of
a bar soap system. Tallow is
the major material for soap
production, and its primary raw material
source is the grain fed to cattle. Production
of paper for packaging the soap is also
included. The fate of both the soap and its
packaging end the life cycle of this system.
Minor inputs could include, for example,
the energy required to fabricate the tires on
the combine used to plant and harvest the
grain.
In a life-cycle inventory to create a baseline
for future product development or improve-
ment, the unit upon which the analysis is
20
-------
Procedural Framework
performed can be almost anything that pro-
duces internally consistent data. In the bar
soap example, one possible usage unit could
be a single bar. However, if the product
packaging were being analyzed at the same
time, it would be important for consistency
to consider packaging hi different amounts
such as single bars, three packs, and so on.
If the life-cycle inventory were intended to
analyze whether bar soap should be manu-
factured using an animal-derived or
vegetable-derived raw material source, the
system boundaries and units of analysis
would be more complicated. First, the sys-
tem flow diagram would have to be
expanded to include the growing, harvest-
ing, and processing steps for the alternative
feedstock. Then the performance of the fin-
ished product would have to be considered.
Do the options result in a bar that gets used
up at different rates "when one material or
the other is chosen? If this were the case, a
strict comparison of equal-weight bars
would not be appropriate.
Suppose an analyst wants to compare bar
soap made from tallow with a liquid hand
soap made from synthetic ingredients.
Because the two products have different raw
material sources (cattle and petroleum), the
analysis should begin with the raw materi-
als acquisition steps. Because the two prod-
ucts are packaged differently and may have
different chemical formulas, the materials
manufacture and packaging steps would
need to be included. Consumer use and
waste management options also should be
examined because the different formulas
could result in varying usage patterns. Thus
for this comparative analysis, the analyst
would have to inventory the entire life cycle
of the two products.
Again, the analyst must determine the basis
of comparison between the systems.
Because one soap is a solid and the other is
a liquid, each with different densities and
cleansing abilities per unit amount, it would
not make sense to compare them based on
equal weights or volumes. The key factor is
how much of each is used hi one hand-
washing to provide an equivalent level of
function or service. An acceptable basis for
comparison might be equal numbers of
hand-washings. Because these two products
may be used at different rates, it would be
important to find data that give an equiva-
lent use ratio. For example, a research lab
study may show that 5 mm3 of bar soap and
10 mm3 of liquid soap are used per hand-
washing. If the basis for comparison were
chosen at 1,000 hand-washings, 5,000 mm3
of bar soap would be compared to 10,000
mm3 of liquid hand soap. Thus, the equiva-
lent use ratio is 1 to 2.
When specific brands of soap are being ana-
lyzed to establish individual life-cycle
inventories, market share information need
not be obtained. However, when two or
more items perform the same function and
the inventory application is framed to
answer the question of identifying which
product type exhibits higher or lower load-
ings, market share information is important
hi comparing product types. For example,
hi an analysis comparing a typical bar soap
with a typical liquid soap, market share data
would be used to allocate the raw materials
and emissions among the specific soap
brands of each type. These data would be
used to proportion the contributions of
chemicals, raw materials, usage rates, etc.,
of the two soap types to develop average
data. On the other hand, if specific soap
brands are being compared, then, of course,
market share is also irrelevant.
21
-------
Procedural Framework
Because the two soap product types are
packaged in different quantities and materi-
als, the analyst would need to include pack-
aging in the system. Contributions of extra
ingredients, such as perfumes, might also be
considered. The analyst may or may not
find that any extra raw materials are used in
one of the two soap types to make it clean as
well as the other. Soaps typically must meet
a minimum standard performance level.
However, if the liquid hand soap also had a
skin moisturizer in its formula, the analyst
would need to include a moisturizing lotion
product in the boundary of the bar soap sys-
tem on two conditions. The first condition
would apply if the environmental issues
associated with this component were ger-
mane to the purpose of the life-cycle inven-
tory. The second condition, which is not as
clear-cut, is if there is actual value received
by the consumer from inclusion of the mois-
turizer. If market studies indicate that con-
sumers purchase the product in preference
to an identical product without a moistur-
izer, or if they subsequently use a moisturiz-
ing lotion after using a nonmoisturizing
soap, then equivalent use would entail
including the separate moisturizing lotion.
Including the moisturizing lotion would
move the comparison beyond equivalent
hand-washing to equivalent hand-washing
and skin moisturizing.
In defining system boundaries, it is impor-
tant to include every step that could affect
the overall interpretation or ability of the
analysis to address the issues for which it is
being performed. Only in certain well-
defined instances can life-cycle elements
such as raw materials acquisition or waste
management be excluded. In general, only
when a step is exactly the same in process,
materials, and quantity in all alternatives
considered, can that step be excluded from
the system, hi addition, the framework for
the comparison must be recognized as rela-
tive because the total system values exclude
certain contributions. This rule is especially
critical for inventories used in public
forums rather than for internal company
decision making. For example, a company
comparing alternative processes for pro-
ducing one petrochemical product may not
need to consider the iise and disposal of the
product if the final composition is identical.
The company may also find that each pro-
cess uses exactly the same materials in the
same amounts per unit pf product output.
Therefore, the company may consider the
materials it uses as having no impact on the
study results. Another example is a filling
operation for bottles. A company interested
in using alternative materials for its bottles
while maintaining the same size and shape
may not need to include filling the bottles
as part of the inventory; system. However, if
the original bottles were compared to boxes
of a different size and shape, the filling step
would need to be included.
When a life-cycle inventory is used to com-
pare two or more products, the basis of com-
parison should be equivalent use; i.e., each
system should be defined so that an equal
amount of product or equivalent service is
delivered to the consumer. In the hand-
washing example, if bar soap were com-
pared to liquid soap, the logical basis for
comparison would be an equal number of
hand-washings. Another example of equiva-
lent use would be in comparing cloth dia-
pers to disposable diapers. One type of
diaper' may typically be changed more fre-
quently than the other, and market/use
studies show that often cloth diapers are
doubled, whereas disposables are not. Thus,
22
-------
Procedural Framework
throughout a day, more cloth diapers will be
used. In this case, a logical basis for compar-
ison between the systems would be the total
number of diapers used over a set period of
time.
Equivalent use for comparative studies can
often be based on volume or weight, partic-
ularly when the study compares packaging
for delivery of a specific product. A bever-
age container study might consider 1,000
liters of beverage as an equivalent use basis
for comparison, because the product may be
delivered to the consumer in a variety of
different-size containers having different
life-cycle characteristics.
Resource constraints for the life-cycle
inventory may be considerations in defining
the system boundaries, but in no case
should the scientific basis of the study be
compromised. The level of detail required
to perform a thorough inventory depends on
the size of the system and the purpose of the
study. In a large system encompassing sev-
eral industries, certain details may not be
significant contributors given the defined
intent of the study. These details may be
.omitted without affecting the accuracy or
application of the results. However, if the
study has a very specific focus, such as a
manufacturer comparing alternative pro-
cesses or materials for inks used on packag-
ing, it would be important to include
chemicals used in very small amounts.
Additional areas to consider in setting
boundaries include the manufacture of capi-
tal equipment, energy and emissions associ-
ated with personnel requirements, and
precombustion impacts for fuel usage.
These are discussed in Chapter Four.
After the boundaries of each system have
been determined, a system flow diagram as
shown in Figure 3 can be developed to
depict the system. Each system step should
be represented individually in the diagram,
including the production steps for ancillary
inputs or outputs such as chemicals and
packaging. If a decision to exclude certain
items has been made, it is appropriate, for
purposes of maintaining transparency, to
explain the system flow diagram. Often in
studies intended for external application,
the system flow diagram is incorporated
into a formal scope, boundary, and data col-
lection (SBDC) document. An SBDC docu-
ment provides the basis for peer reviewers
and others .to understand how the analyst is
defining the study and how these defini-
tions and assumptions will translate into
data collection activities.
DEVISE AN INVENTORY CHECKLIST
The inventory checklist is a tool that covers
most decision areas hi the performance of
an inventory. After the inventory purpose
and boundaries have been defined, a check-
list can be prepared to guide data collection
and validation and to enable construction of
the computational model. Figure 4 shows a
generic example of an inventory checklist
and an accompanying data worksheet.
Although this checklist is an effective guid-
ance tool and enhances transparency, it is
not the sole quality control process under
which the analysis should be performed.
Analysts will want to tailor this checklist
for a given product or material. Eight gen-
eral decision areas should be addressed on
the checklist or worksheet:
Purpose of the inventory
System boundaries
Geographic scope
23
-------
: Procedural Framework
LIFE-CYCLE INVENTORY CHECKLIST PART ISCOPE AND PROCEDURES
INVENTORY OF: _j !
Purpose of Inventory: (check all that apply)
Private Sector Use
Internal Evaluation and Decision Making
D Comparison of Materials, Products, or Activities
D Resource Use and Release Comparison with Other
Manufacturer's Data
D Personnel Training for Product and Process Design
D Baseline Information for Full LCA
External Evaluation and Decision Making
D Provide Information on Resource Use and Releases
D Substantiate Statements of Reductions in Resource Use and
Releases
Public Sector Use
Evaluation and Policy-making '
D Support Information for Policy and Regulatory Evaluation
D Information Gap Identification
D Help Evaluate Statements of Reductions in Resource Use and
Releases
Public Education
D Develop Support Materials for Public Education
D Assist in Curriculum Design
Systems Analyzed
List the product/process systems analyzed in this inventory:.
Key Assumptions: (list and describe)
Define the Boundaries ' \ '
For each system analyzed, define the boundaries by life-cycle stage, geographic scope, primary processes, and ancillary inputs included in
the system boundaries. i ' - .
i
Postconsumer Solid Waste Management Options: Mark and describe the options analyzed for each system.
D Landfill D Open-loop Recycling . '
O Combustion .
D Composting .
D Closed-loop Recycling
D Other
Basis for Comparison
D This Is not a comparative study. D This is a comparative study.
State basis for comparison between systems: (Example: 1000 units, 1,000 uses) _
If products or processes are not normally used on a one-to-one basis, state how equivalent function was established.
Computational Model Construction j
O System calculations are made using computer spreadsheets that relate each system component to the total system.
D System calculations are made using another technique. Describe: j
Describe how inputs to and outputs from postconsumer solid waste management are handled. .
Quality Assurance: (state specific activities and initials of reviewer)
Review performed on: D Data Gathering Techniques :
D Coproduct Allocation '
-D Input Data :
_D Model Calculations and Formulas.
D Results and Reporting '
Peer Review: (state specific activities and initials of reviewer)
Review performed on: D Scope and Boundary
D Data Gathering Techniques .
D Coproduct Allocation
D Input Data '.
. D Model Calculations and Formulas.
. D Results and Reporting :
Results Presentation
O Methodology is fully described.
O Individual pollutants are reported.
D Emissions are reported as aggregated totals only.
Explain why:
D Report is sufficiently detailed for its defined purpose.
LJ Report may need more detail for additional use beyond
defined purpose. :
G Sensitivity analyses are included in the report.
List . : L_ :
U Sensitivity analyses have been performed but are not included
in the report. List: ! :
Figure 4
A typical diecklist of criteria with worksheet for performing a life-cycle inventory
24
-------
Procedural Framework
LIFE-CYCLE INVENTORY CHECKLIST PART II MODULE WORKSHEET
Im/Antnrynf- Pronaror-
Life-Cycle Stage D
Date.-
Rsnription:
Onalitv Acci iranrva Ar>r»rr>val-
MODULE DESCRIPTION-
Materials
Process
Otherte)
Energy
Process
Precombustion
Water Usage
Process
Fuel-related
Product
CoproductsW
Air Emissions
Process
Fuel-related
Water Effluents
Process
Fuel-related
Solid Waste
Process
Fuel-related
Capital Repl.
Transportation
Personnel
Data Value(a)
Type^
MODULE INPUTS
MODULE OUTPUTS
Data(°) Age/Scope .
Quality Measures^)
,
(a) Include units.
(b) Indicate whether data are actual measurements, engineering estimates, or theoretical or published values and whether the numbers are from a specific manufacturer
or facility, or whether they represent industry-average values. List a specific source if pertinent, e.g., "obtained from Atlanta facility vrastewater perniit monitoring data."
(c) Indicate whether emissions are all available, regulated only, or selected. Designate data as to geographic specificity, e.g., North America, and indicate the period
covered, e.g., average of monthly for 1991 .
(d) List measures of data quality available for the data item, e.g., accuracy, precision, representativeness, consistency-checked, other, or none.
(e) Include nontraditional inputs, e.g., land use, when appropriate and necessary.
(f) If coproduct allocation method was applied, indicate basis in quality measures column, e.g., weight.
Figure 4
Continued
25
-------
Procedural Framework
Types of data used
Data collection and synthesis procedures
Data quality measures
Computational model construction
Presentation of the results.
A standard checklist can be helpful in sev-
eral settings. The analyst performing the
life-cycle inventory can use the checklist to
help ensure that all important stages and
categories of information are included. The
checklist can also help clarify the issues,
boundaries, and conditions to be dealt with
in a particular study. Worksheets can be
used by the analyst to collect and qualify
data from facilities.
The checklist consists of two major compo-
nentsa summary section describing the
procedures and systems included in the
study and a set of worksheets listing and
qualifying the data collected. The checklist
is to be used throughout a study to ensure
that all of the boundaries and comparison
issues are identified and addressed during
the study. The worksheet portion has a dual
purpose: as a tool for the analyst to coordi-
nate and assimilate data and for use in
requesting data from others. In a life-cycle
inventory where there may be many steps in
each life-cycle stage, the worksheets help
ensure consistency among the various infor-
mation sources. Modules consisting of sub-
system inputs and outputs are the basis for
preparing a life-cycle inventory. Subsystem
modules represent fundamental operations
that are building blocks for aggregating data
to the life-cycle stage and overall system
level. For example, a module may be con-
structed for the production of caustics as
shown in the system flow diagram for bar
soap (Figure 3). Tb*is module would be one
of several comprising the materials manu-
facture substage. Worksheets may be pre-
pared for each facility in each subsystem
module and may be used by the analyst to
aggregate these data to the life-cycle stage
level. Additional internal quality control
and quality assurance procedures should be
in place to ensure that the inventory is com-
plete and sound.
A checklist such as the one in Figure 4 also
can be used as a communication tool. By
including a completed checklist in the
report on the results of an inventory, the
analyst can communicate to readers some of
the factors that may affect the results. The
checklist will help readers gain knowledge
and understanding of the system's bound-
aries, data quality, methodology used, and
level of detail. Standard checklists included
in reports of inventories also can help read-
ers recognize and understand differences
among various reports on the same topic.
The standard checklist also can be of use to
peer reviewers as it provides useful criteria
and information to assure the completeness
of a particular life-cycle inventory.
INSTITUTE A PEER REVIEW PROCESS
The desirability of a peer review process has
been a major focus of discussion in many
life-cycle analysis forums. The discussion
stems from concerns in four areas: lack of
understanding regarding the methodology
used or the scope of the study, desire to
verify data and the analyst's compilations of
data, questioning key assumptions and the
overall results, and communication of
results. For these reasons, it is recom-
mended that a peer review process be estab-
lished and implemented early in any study
that will be used in a public forum.
-------
Procedural Framework
SET AC is working with business, consumer,
and environmental groups, and with acade-
mia to develop a peer review process for
inventory studies (Fava et al., 1992). The
following discussion is not intended to be a
blueprint of a specific approach. Instead, it
is meant to point out issues that the practi-
tioner or sponsor should keep in mind
when establishing a peer review procedure.
Overall, a peer review process should -
address the four areas previously identified:
Scope/boundaries methodology
Data acquisition/compilation
Validity of key assumptions and results
Communication of results.
The peer review panel could participate at
several points in the study: (1) reviewing
the purpose, system boundaries, assump-
tions, and data collection approach;
(2) reviewing the compiled data and the
associated quality measures; and (3) review-
ing the draft inventory report, including the
intended communication strategy.
A checklist such as the one presented in
Figure 4 would be useful in addressing
many of the issues surrounding scope/
boundaries methodology, data/compilation
of data, and validity of assumptions,and
results. Criteria may need to be established
for communication of results. These criteria
could include showing how changes hi key
assumptions could affect the study results
and guidance on how to publish and com-
municate results without disclosing propri-
etary data.
It is generally believed that the peer review
panel should consist of a diverse group of 3
to 5 individuals representing various sec-
tors, such as federal, state, and local govern-
ments; academia; industry; environmental
or consumer groups; and LCA practitioners.
Not all sectors need be represented on every
panel. The credentials or background of
individuals should include a reputation for
objectivity, experience with the technical
framework or conduct of life-cycle studies,
and a willingness to work as part of a team.
Issues for which guidelines are still under
development include panel selection, num-
ber of reviews, using the same reviewers for
all life-cycle studies or varying the members
between studies, and having the review
open to the public prior to its release. The
issue of how the reviews should be per-
formed raises a number of questions, such
as these: Should a standard checklist be
required? Should oral as well as written
comments from the reviewers be accepted?
How much time should be allotted for
review? Who pays for the review process?
The peer review process should be flexible
to accommodate variations in the applica-
tion or scope of life-cycle studies. Peer
review should improve the conduct of these
studies, increase the understanding of the
results, and aid in further identifying and
subsequently reducing any environmental
consequences of products or materials. EPA
supports the use of peer reviews as a mecha-
nism to increase the quality and consistency
of life-cycle inventories.
GATHER DATA
The system flow diagram (Figure 3) is useful
in conjunction with the checklist and work-
sheets (Figure 4) in directing efforts to
gather data for the life-cycle inventory.
27
-------
Procedural Framework
Identify Subsystems
For data-gathering purposes it is appropriate
to view the system as a series of subsystems.
A "subsystem." is defined as an individual
step or process that is part of the defined
production system. Some steps in the
system may need to be grouped into a
subsystem due to lack of specific data for
the individual steps. For example, several
steps may be required in the production of
bar soap from tallow. However, these steps
may all occur within the same facility,
which may not be able to or need to break
data down for each individual step. The
facility could, however, provide data for all
the steps together, so the subsystem bound-
ary would be drawn around the group of
soap production steps and not around each
individual one.
Each subsystem requires inputs of materials
and energy; requires transportation of prod-
uct produced; and has outputs of products,
coproducts, atmospheric emissions, water-
borne wastes, solid wastes, and possibly
other releases. For each subsystem, the
inventory analyst should describe materials
and energy sources used and the types of
environmental releases. The actual activities
that occur should also be described. Data
should be gathered for the amounts and
kinds of material inputs and the types and
quantities of energy inputs. The environ-
mental releases to air, water, and land
should be quantified by type of pollutant.
Data collected for an inventory should
always be associated with a quality mea-
sure. Although formal data quality indica-
tors (DQIs) such as accuracy, precision,
representativeness, and completeness are
strongly preferred, a description of how the
data were generated can be useful in judging
quality. EPA is specifically addressing the
use of quantitative and qualitative DQIs in a
separate guidance document on data quality
hi life-cycle assessment.
Coproducts from the process should be
identified and quantified. Coproducts are
process outputs that have value, i.e., those
not treated as wastes. The value assigned to
a coproduct may be a market value (price)
or may be imputed. La performing coprod-
uct allocation, some means must be found
to objectively assign the resource use,
energy consumption, and emissions among
the coproducts, because there is no physical
or chemical way to separate the activities
that produce them. Advantages and disad-
vantages of specific approaches to coprod-
uct allocation are discussed later in this
chapter and in Chapter Four. Generally,
allocation should allow technically sound
inventories to be prepared for products or
materials using any particular output of a
process independently and without overlap
of the other.outputs.
In the meat packing step: of the bar soap
example shown in Figure 3, several coprod-
xucts could be identified: meat, tallow, bone
meal, blood meal, and hides. Other exam-
ples of coproducts are the trim scraps and
off-spec materials from a molded plastic
plate fabricator. If the trim scraps and off-
spec materials are used or marketed to other
manufacturers, they are considered as
coproducts. Industrial scrap is the common
name given to such materials. If the trim is
discarded into the solid waste stream to be
landfilled, it should be included in the solid
waste from the process. If the trim or off-
spec materials are reused within the pro-
cess, they are considered "home scrap,"
which is part of an internal recycling loop.
These materials are not included in the
28
-------
Procedural Framework
Sources of Data
Electronic non-bibliographic data bases (govern-
ment and industriaO
- averaged industrial data
- product specifications
Electronic bibliographic data bases
Electronic database clearinghouses
Relevant documents
- government reports
- open literature papers and books
- other life-cycle inventories
* Facility-specific industrial data
- publidy accessible
- nonpublidy accessible
Laboratory test data
Study-specific data
inventory, because they do not cross the
boundaries of the subsystem.
All transportation from one process location
to another is included in the subsystem.
Transportation is quantified in terms of dis- ,
tance and weight shipped, and identified by
the mode of transport used.
Sources of Data
A number of sources should be used in col-
lecting data. Whenever possible, it is best to
get well-characterized industry data for pro-
duction processes. Manufacturing processes
often become more efficient or change over
time, so it is important to seek current data.
Inventory data can be facility-specific or
more general and still remain current.
Several categories of data are often used in
inventories. Starting with the most disaggre-
gated, these are:
Individual process- and facility-specific:
data from a particular operation within a
given facility that are not combined in
any way
Composite: data from the same operation
or activity combined across locations
Aggregated: data combining more than
one process operation
Industry-average: data derived from a rep-
resentative sample of locations and
believed to statistically describe the typi-
cal operation across technologies
Generic: data whose representativeness
may be unknown but which are qualita-
tively descriptive of a process or
technology.
Complete and thorough inventories often
require use of data considered proprietary
by either the manufacturer of the product,
upstream suppliers or vendors, or the LCA
practitioner performing the study. Confiden-
tiality issues are not relevant for life-cycle
inventories conducted by companies using
their own facility data for internal purposes.
However, the use of proprietary data is a
critical issue in inventories conducted for
external use and whenever facility-specific
data are obtained from external suppliers for
internal studies. As a consequence, current
studies often contain insufficient source and
documentation data to permit technically
sound external review. Lack of technically
sound data adversely affects the credibility
of both the life-cycle inventories and the
method for performing them. An individual
company's trade secrets and competitive
technologies must be protected. When
29
-------
Procedural Framework
collecting data (and later when reporting the
results), the protection of confidential busi-
ness information should be weighed against
the need for a full and detailed analysis or
disclosure of information. Some form of
selective confidentiality agreements for enti-
ties performing life-cycle inventories, as
well as formalization of peer review
procedures, is often necessary for invento-
ries that will be used in a public forum.
Thus, industry data may need to undergo
intermediate confidential review prior to
becoming an aggregated data source for a
document that is to be publicly released.
The purpose, scope, and boundary of the
inventory help the analyst determine the
level or type of information that is required.
For example, even when the analyst can
obtain actual industry data, hi what form
and to what degree should the analyst show
the data (e.g., the range of values observed,
industry averages, plant-specific data, best
available control techniques)? These ques-
tions or decisions can usually be answered
if the purpose or scope has been well
defined. Typically, most publicly available
life-cycle documents present industry aver-
ages, while many internal industrial studies
use plant-specific data. Recommended prac-
tice for external life-cycle inventory studies
includes the provision of a measure of data
variability in addition to averages. Fre-
quently the measure of variability will be a
statistical parameter, such as a standard
deviation. Other options, which may be use-
ful for small data sets or where confidential-
ity may be breached by a reported standard
deviation, are discussed in more detail in
Chapter Four.
Examples of private industry data sources
include independent or internal reports,
periodic measurements, accounting or engi-
neering reports or data sets, specific mea-
surements, and machine specifications. One
particular issue of interest hi considering
industrial sources, whether or not a formal
public data set is established, is the influ-
ence of industry and related technical asso-
ciations to enhance the accuracy, represen-
tativeness, and currentness of the collected
data. Such associations-may be willing,
without providing specific data, to confirm
that certain data (about which their mem-
bers are knowledgeable) are realistic.
Government documents and data bases pro-
vide data on broad categories of processes
and are publicly available. Most government
documents are published on a periodic
basis, e.g., annually, biennially, or every
4 years. However, the data published within
them tend to be at least several years old.
Furthermore, the data fojund in these docu-
ments may be less specific and less accurate
than industry data for specific facilities or
groups of facilities. However, depending on
the purpose of the study and the specific
data objectives, these limitations may not be
critical. All studies should note the age of
the data used. Some useful government
documents include: i
U.S. Department of Commerce, Census of
Manufacturers :
i
U.S. Bureau of Mines,! Census of Mineral
Industries
U.S. Department of Energy, Monthly
Energy Review
U.S. Environmental Protection Agency,
Toxic Release Inventory (TRI) Database.
Government data bases include both non-
bibliographic types where the data items
themselves are contained in the data base
3O
-------
Procedural Framework
and bibliographic types that consist of
references where data may be found.
Technical books, reports, conference
papers, and articles published in technical
journals can also provide information and
data on processes in the system. Most of
these are publicly available. Data presented
in these sources are often older, and they
can be either too specific or not specific
enough. Many of these documents give the-
oretical data rather than real data for pro-
cesses. Such data may not be representative
of actual processes or may deal with new
technologies not commercially tested. In
using the technical data sources in the fol-
lowing list, the analyst should consider the
date, specificity, andjrelevancyof the data:
Encyclopedia of Chemical Technology,
Kirk-Othmer
Periodical technical journals such as Jour-
nal of the Water Environment Federation
Proceedings from technical conferences
Textbooks on various applied sciences.
Surveys designed to capture information on
a representative sample of end users can
provide current information on the param-
eters of product or service use. Surveys typi-
cally center around a question:
How long or how many times is a product
or service used before it is discarded (e.g.,
the number of years a television set has
been in use and is expected to be in use)?
What other materials and what quantities
of these materials are used in conjunction
with product use or maintenance (e.g.,
moisturizing lotion use after hand-
washing)?
How frequent is the need for product
repair or maintenance (e.g., how often is
an appliance repaired over its lifetime,
and who does the repair)?
What other uses does the product have
beyond its original purpose?
What does the end user do with the prod-
uct when he or she is through with it?
Frequently, the end user will not be able to
supply specific information on inputs and
outputs. However, the end user can provide
data on user practices from which inputs
and outputs can be derived. Generally, the
end user can be the source of related infor-
mation from which the energy, materials,
and pollutant release inventory can be .
derived. (An exception would be an institu-
tional or commercial end user who may
have some information on energy consump-
tion or water effluents.) Market research
firms can often provide qualitative and
quantitative usage and customer preference
data without the analyst having to perform
independent market surveys.
Recycling provides an example of some of
the strengths and limitations encountered in
gathering data. For some products,
economic-driven recycling has been prac-
ticed for many years, and an infrastructure
and markets for these materials already
exist. Data are typically available for these
products, including recycling rates, the con-
sumers of the reclaimed materials, and the
resource requirements and environmental
releases from the recycling activities (collec-
tion and reprocessing). Data for materials
currently at low recycling rates with newly
forming recycling infrastructures are more
difficult to obtain. In either case, often the
best source for data on resource require-
ments and environmental releases is the
31
-------
Procedural Framework
processors themselves. For data on recy-
cling rates and recycled material, consumers
and processors may be helpful, but trade
associations as well as the consumers of the
recycled materials can also provide data.
For materials that are recycled at low rates,
data will be more difficult to find.
Two other areas for data gathering relate to the
system as a whole and to comparisons
between and among systems. It is necessary to
obtain data on the weights of each component
in the product evaluated, either by obtaining
product specifications from the manufacturer
or by weighing each component. These data
are then used to combine the individual com-
ponents in the overall system analysis. Equiv-
alent use ratios for the products compared can
be developed by surveying retailers and con-
sumers, or by reviewing consumer or trade as-
sociation periodicals.
DEVELOP STAND-ALONE DATA
"Stand-alone data" is a term used to
describe the set of information developed to
standardize or normalize the individual
subsystem module inputs and outputs for
the specific product, process, or activity
being analyzed. Stand-alone data must be
developed for each subsystem to fit the sub-
systems into a single system. There are two
goals to achieve in this step: (1) presenting
data for each subsystem consistently by
reporting the same product output from
each subsystem; (2) developing the data in
terms of the life cycle of only the product
being examined in the inventory. First, a
standard unit of output must be determined
for each subsystem. All data could be
reported in terms of the production of a cer-
tain number of pounds, kilograms, or tons of
subsystem product. For example, the har-
vesting of trees, production of paper, and
packaging of soap are all steps in packaging
soap as seen in Figure 3. Data for these steps
could be developed on the basis of 1,000
tons of output: 1,000 tons of harvested trees,
1,000 tons of paper, and 1,000 tons of pack-
aged soap. Although historically English
units heve been used for subsystem
accounting in the USA, international prac-
tice is leaning more toward the use of metric
units of measurement. The units used for
the individual steps or subsystems do not
necessarily have to match those of the final
product.
Once the decision for the reporting basis has
been made, the data obtained for the subsys-
tem need to be adjusted to that product's
output level. For example, suppose an ana-
lyst performing a study on bar soap has
received data on the caustic manufacture
subsystem. This process yields three
coproducts: caustic, chlorine, and hydrogen.
Assume the data obtained for the process
were given in terms of the joint production
of 500 tons of caustic, 250 tons of chlorine,
and 5 tons of hydrogen. Only the caustic is
used in the bar soap production system.
Because the analyst has chosen 1,000 tons
as the reporting basis for all subsystem data,
the caustic manufacturing data need to be
presented in terms of 1,000 tons of output of
caustic (and thus also 500 tons of chlorine
and 10 tons of hydrogen). To do this, all of
the process data supplied would be multi-
plied by two. For purposes of this example,
only the chlorine coproduct is considered
further.
Now that the data are at a consistent level of
reporting, the analyst needs to determine
the energy and material! requirements and
the environmental releases to be attributed
to the production of each coproduct using a
technique called coproduct allocation. One
32
-------
Procedural Framework
commonly used allocation method is based
on relative weight. Figure 5 illustrates this
technique. In this example, the top portion
of the figure can be used to illustrate the
actual process flow diagram for hypothetical
production of caustic (labeled Product A),
with chlorine (Product B) as a coproduct of
caustic. Because the bar soap system exclu-
sive of packaging uses only caustic, arid not
chlorine, the energy and material inputs and
environmental releases of the process must
be allocated separately to the caustic and
chlorine. The lower portion of the diagram
illustrates this allocation. Product A, caus-
tic, represents two-thirds of the total pro-
duction output of the process, so two-thirds
of the energy and resource inputs and two-
thirds of the environmental releases are
attributed to caustic. Likewise, Product B,
chlorine, represents one-third of the total
rproduction*output of the process, so one-
third of the energy and resource inputs and
one-third of the environmental releases are
attributed to the chlorine. Performing data
allocation in this way allows the analyst to
isolate those inputs and outputs relevant to
the product being studied. Alternative allo-
cation methods are discussed in Chapter
Four.
Once the inputs and outputs of each subsys-
tem have been allocated, numerical relation-
ships of the subsystems within the entire
system flow diagram can be established.
This is done starting at the finished product
of the system and working backward, using
the relationships of the material inputs and
product outputs of each subsystem to com-
pute the input requirements from each of
the preceding subsystems. For example,
suppose the bar soap system were to be ana-
lyzed on the basis of 1,000 tons of packaged
bar soap. If the bar soap packaging process
requires 900 tons of bar soap to produce
1,000 tons of packaged bar soap, only 900
tons of bar soap would need to be manufac-
tured for the total system. Suppose 2,000
tons of tallow are required to produce 1,000
tons of bar soap. Only 900 tons of bar soap
are required for the total bar soap system, so
only 1,800 tons of tallow are needed for the
total system.
CONSTRUCT A COMPUTATIONAL
MODEL
The next step in a life-cycle inventory is
model construction. This step consists of
incorporating the normalized data and
material flows into a computational frame-
work using a computer spreadsheet or other
accounting technique. The systems account-
ing data that result from the computations
of the model give the total results for the
energy and resource use and environmental
releases from the overall system.
The overall system flow diagram, derived in
the previous step, is important in construct-
ing the computational model because it
numerically defines the relationships of the
individual subsystems to each other in the
production of the final product. These
numerical relationships become the source
of "proportionality factors," which are
quantitative relationships that reflect the
relative contributions of the subsystems to
the total system. For example, data for the
production of a particular ingredient X of
bar soap are developed for the production of
1,000 tons of X. To produce 1,000 tons of
bar soap, 250 tons of X are needed, account-
ing for losses and inefficiencies. Thus, to
find the contributions of X to the total sys-
tem, the data for 1,000 tons of X are multi-'
plied by 0.250.
33
-------
Procedural Framework
Actual process flow diagram
for the production of
Products 'A' and 'B'
Flow diagrams showing the
normalized resources and
environmental releases for
each coproduct
Energy
3x109 Btu
*
1,600 Ib (
Raw or »J
Intermediate I
30 Ib
Atmospheric
Emissions '
10C
Solid \
Energy
2x109 Btu
1
1,067 Ib ( ,
Raw or »|
Intermediate I
20 Ib
Atmospheric
Emissions 1
67
Solid
Water
600 gal
1
"\ ir
1 Transportation
. -^-r "'
1
10 Ib
Waterborne
f Wastes
)lb
Waste
Water
400 gal
1
j Transportation
J
7lb
Waterborne
Wastes
Ib ;
Waste
1,000lb
Product 'A'
500 Ib
Product 'B1
1,000lb
Product 'A'
a
Coproduct Allocation for Product 'A'
Energy
1 x109 Btu
. i
533 Ib f
Raw or ^4
Intermediate I
10 Ib
Atmospheric
Emissions '
33
Solid
Water
200 gal
i
| Transportation
J
3lb
Waterborne
Wastes
Ib
Waste
500 Ib
Coproduct 'B'
Coproduct Allocation for Product 'B'
Figure 5
Example coproduct allocation based on relative weight
-------
Procedural Framework
The spreadsheet can be used to make other
computations beyond weighting the contri-
butions of various subsystems. It can be
used to translate energy fuel value to a stan-
dard energy unit, such as million Btu or
gigajoule (GJ). Precombustion or resource
acquisition energy can be computed by
applying a standard factor to a unit quantity
of fuel to account for energy used to obtain
and transport the fuel. Energy sources, as
weir as types of wastes, can be categorized.
Credits or charges for incineration can be
derived. Fuel-related wastes should also be
calculated based on the fuels used through-
out the system. The model should also
incorporate waste management options,
such as recycling, composting, and land-
filling. The method for handling these
aspects is discussed in Chapter Five.
It is important that each subsystem be incor-
porated in the model with its related com-
ponents and that each be linked together in
such a way that inadvertent omissions and
double-counting do not occur; The compu-
ter spreadsheet can be organized in several
different ways to accomplish this purpose.
These can include allocating certain fields
or areas in the spreadsheet to certain types
of calculations or using one type of spread-
sheet software to actually link separate
spreadsheets in hierarchical fashion. It is
imperative, however, once a system of orga-
nization is used, that it be employed consis-
tently. Haphazard organization of data sets
and calculations generally leads to faulty
inventory results.
Many decisions must be made in every life-
cycle inventory analysis. Every inventory
consists of a mix of factual data and
assumptions. Assumptions allow the ana-
lyst to evaluate a system condition when
factual data either cannot be obtained
within the context of the study or do not
exist. Each piece of information (e.g., the
weight of papefboard used to package the
soap, type of vehicle and distance for ship-
ping the tallow, losses incurred when ren-
dering tallow, or emissions resulting from
the animals at the feedlot), falls into one or
the other category and each plays a role in
developing the overall system analysis.
Because assumptions can substantially
affect study results, a series of "what if cal-
culations or sensitivity analyses are often
performed on the results to examine the
effect of making changes in the system. A
sensitivity analysis will temporarily modify
one or more parameters and affect the calcu-
lation of the results. Observing the change
in the results will help determine how
important the assumptions are with respect
to the results. The computational model is
also used to perform these sensitivity analy-
sis calculations.
Sometimes it is helpful to think ahead about
how the results will be presented This can
direct some decisions on how the model
output is specified. The analyst must
remember the defined purpose for perform-
ing the analysis and tailor the data output to
those expressed needs. For example, the
analyst might ask: Is the purpose of the life-
cycle inventory to evaluate the overall sys-
tem results? Or is it expected that detailed
subsystem information will be analyzed in
relation to the total? Will the study be used
in a public forum? If so, how? How much
detail is required? Answers to questions
such as these will help determine the com-
plexity and the degree of generalization to
build into the model, as well as the appro-
priate presentation of results.
35
-------
Procedural Framework
PRESENT THE RESUUS
When writing a report to present the final
results of the life-cycle inventory, it is
important to thoroughly describe the metho-
dology used in the analysis, The report
should explicitly define the systems ana-
lyzed and the boundaries that were set, All
assumptions made in performing the inven«
tory should be clearly explained, The basis
for comparison among systems should be
given, and any equivalent usage ratios that
were used should be explained, Use of the
checklist and worksheet, as shown in Fig-
ure 4, supports a clear process for communi-
eating this information,
Life-cycle inventory studies generate a peat
deal of information, often of a disparate.
nature, The analyst needs to select a presen-
tation format and content that are consistent
with the purpose of the study and that do
not arbitrarily simplify the information
solely for the sake of presenting it, In think"
ing about presentation of the results, it is
useful to identify the various perspectives
embodied in life-cycle inventory informa-
tion, These dimensions include but may not
be limited to the following;
Overall product system
Relative contribution of stages to the
overall system
Relative contribution of product compo-
nents to the overall system
Data categories within and across stages,
e.g., resource use, energy consumption,
and environmental releases
Data parameter groups within a category,
e,g,, air emissions, waterborne wastes,
and solid waste types
Data parameters within a group, e,g,, sul-
fur oxides, carbon dioxide, chlorine, etc,
t Geographic regionaHzation if relevant to
the study, e,g,, national versus global
Temporal changes,
The life-cycle analyst must select among
these dimensions and develop a presenta-
tion format that increases comprehension of
the findings without oversimplifying them,
Two main types of format for presenting
results are tabular and graphical,
i
Sometimes it is useful to report total energy
results while also breaking out the contribu-
tions to the total from process energy and
energy of material resource, Solid wastes
can be separated into postconsumer solid
waste and industrial solid waste, individual
atmospheric and water pollutants should be
reported separately, Atmospheric emissions,
waterbome wastes, and industrial solid
wastes can also be categorised by process
emissions/wastes and fuel-related
emissions/wastes, Such itemized presenta-
tions can assist in identifying and subse-
quently controlling certain energy
consumption and environmental releases,
The results from the inventory can be pre-
sented most comprehensively in tabular
form, The choice of how the tables should
be created varies, based on the purpose and
scope of the study, If the inventory has been
performed to help decide which type of
package to use for a particular product,
showing the overall system results will be
the most useful way to present the data, On
the other hand, when an [analysis is per-
formed to determine how a package can be
changed to reduce its releases to the envi-
ronment, it is important to present not only
the overall results, but also the
-------
Procedural Framework
contributions made by each component of
the packaging system. For example, in ana-
lyzing a liquid delivery system that uses
plastic bottles, it may be necessary to show
how the bottle, the cap, the label, the corru-
gated shipping box, and the stretch wrap
around the boxes all contribute to the total
results. The user can thus concentrate
improvement efforts on the components that
make a substantial contribution when evalu-
ating proposed changes.
Graphical presentation of information helps
to augment tabular data and can aid in inter-
pretation. Both bar charts (either individual
bars or stacked bars) and pie charts are valu-
able in helping the reader visualize and
assimilate the information from the perspec-
tive of "gaining ownership or participation
in life-cycle assessment" (Werner, 1991).
However, the analyst should not aggregate
or sum dissimilar data when creating or
simplifying a graph.
For internal industrial use by product man-
ufacturers, pie charts showing a breakout by
raw materials, process, and use/disposal
have been found useful in identifying waste
reduction opportunities.
For external studies, the data must be pre-
sented in a format that meets one funda-
mental criterionclarity. Ensuring clarity
requires that the analyst ask and answer
questions about what each graph is
intended to convey. It may be necessary to
present a larger number of graphs and incor-
porate fewer data hi each one. Each reader
should understand the desired response
after viewing the information.
INTERPRET AND COMMUNICATE
THE RESULTS
How the results of the life-cycle inventory
will be interpreted depends on the purpose
for which the analysis was performed.
Before publishing any statements regarding
the results of the analysis, it is important to
review how the assumptions and bound-
aries of the system were defined, the quality
level of the data used, and the specificity
(e.g., were the data specific to one facility or
representative of the entire industry?). Care-
ful interpretation is required to avoid mak-
ing unsupported statements.
An important criterion in understanding or
interpreting the results is data accuracy.
Many life-cycle inventories present data
that are considered representative of the
industry or group being profiled. Data for
one particular step may be gathered from a
number of manufacturers and production
facilities. For example, for the paperboard
manufacturing step, more than 25 plants
may produce a similar product. These
plants may use different raw materials,
employ different technologies, have varying
degrees of plant age/efficiency, and operate
under different site-specific conditions (e.g.,
energy sources and state environmental reg-
ulations). These aggregated and composited
data will be subject to both random and sys-
tematic sources of error. Individual process
variations within a given facility using the
same input and technology represent a
source of true random error. This type of
variability can be described in conventional
statistical terms using the mean and stan-
dard deviation of the measurements. For
small data sets, where reporting of a mean,
range, and standard deviation may compro-
mise confidentiality, a semi-qualitative
description of variability could suffice.
37
-------
Procedural Framework
However, differences resulting from system-
atic variability due to feedstock or technol-
ogy type are not random errors in a
statistical sense. These sources of variability
may be thought of as "explainable" by the
age or operating conditions of the plants or
by other identifiable factors.
The analyst should interpret the importance
of these sources of variability for the reader.
For example, variability analysis for the
paperboard manufacturing step could indi-
cate that plants using a certain type of pulp-
ing process were consistently higher in
certain water pollutants. The analyst could
interpret this fact for the reader and still
protect confidentiality.
Although a rigorous statistical analysis of
the overall inventory may not be possible,
sensitivity analyses of key elements in the
system should be performed to estimate the
magnitude of variability in the data from the
inventory that would have to exist in order
to reflect significant differences in the
results. Sensitivity analysis is a technique
for systematically varying the inputs to a
model in order to establish whether the out-
puts are distinguishable or not (Raiffa,
1968). For life-cycle inventories, a sensitiv-
ity analysis would evaluate how large the
uncertainty in the input data can be before
the results of the inventory can no longer be
used for the intended purpose.
Typically, a difference in past inventories
has been determined to occur where the
outputs vary by more than ±10-25 percent
when data were modified to simulate low
and high ranges. The differences that must
exist to use the results will be defined for
each inventory during the scoping process
as part of the data quality objectives (DQO)
setting. The guidelines for performing a sen-
sitivity analysis in a life-cycle inventory are
presented more fully in Chapter Four. This
margin of difference should be used inde-
pendently for each category of results (e.g.,
energy, atmospheric emissions, waterborne
emissions, and solid waste) and for each
data parameter (e.g., chloroform, particu-
lates, and hydrocarbons) in analyzing the
results. j
The boundaries and data for many internal
life-cycle assessments may require the inter-
pretation of the results for use within a par-
ticular corporation. For example, data used
may be specific to a particular company
and, therefore, may not represent any typi-
cal or average product oji the market. How-
ever, because the data used in this type of
analysis are frequently highly specific, a
fairly high degree of accuracy can be
assumed in interpreting the results. Product
design and facility/process development
groups within companies often benefit from
this level of interpretation.
The public sometimes receives interpreted
information from life-cycle analyses that
have been released. In these instances, life-
cycle inventory information may be pro-
vided to consumers to support statements
about certain features or specific reduction
claims. The analyst must be careful to pro-
vide an interpretive context and not to
selectively use information. In both types of
studies, general systems are described and
the data used in the analysis may not be
specific to one producer. Instead it may be
more representative of a particular industry.
La such cases, a higher margin of difference
should be achieved before results are con-
sidered to be significantly different between
systems.
38
-------
Procedural Framework
The results of externally published studies
comparing products, practices, or materials
should be presented cautiously, and
assumptions, boundaries, and data quality
should be considered in drawing and
presenting conclusions. Studies with differ-
ent boundary conditions may have different
results, yet both may be accurate. These lim-
itations should be communicated to the
public along with all the results; it is mis-
leading to selectively report inventory
results. Final conclusions about the results
of inventory studies may involve making
value judgments regarding the relative
importance of air and water quality, solid
waste issues, resource depletion, and energy
use. Based upon each individual's locale,
background, and life style, different value
judgments will be made.
39
-------
Chapter Four
GENERAL ISSUES
IN PERFORMING
A LIFE-CYCLE INVENTORY
INTRODUCTION
This chapter discusses the general issues
encountered in every stage of a life-cycle
inventory. These issues pertain to the type
of information these studies quantify and
the decisions or assumptions that must be
made in evaluating and using the informa-
tion. One major tool in life-cycle inventory
analysis is the template, which is a pictor-
ial guide that identifies the information that
must he obtained at each step hi an inven-
Major Concepts
Templates, or material and energy balance
diagrams, are tools used to support data
gathering and development for life-cycle
inventory analyses.
Data for processes produdng more than
one product are allocated based on the rela-
tive weights of product output or another
justifiable method.
Data quality objectives are the required per-
formance specifications for information in a
life-cyde inventory. Establishment of
these specifications is determined by the
defined purpose of the life-cycle.inventory.
Data quality indicators are qualitative or
quantitative characteristics of data. These
indude accuracy, bias, representativeness,
and other attributes that measure data
goodness and applicability.
tory analysis. Issues discussed in this
chapter include: >
Using Templates in Life-Cycle Inventory
Analysis
Data Issues
Special Case Boundary Issues.
USING TEMPLATES IN LIFE-CYCLE
INVENTORY ANALYSIS
A template is a guide used by analysts to
direct the collection of data. A template
depicts the material and energy accounts to
visually describe a defined system or sub-
system. A generic version of a template,
shown in Figure 6, indicates which catego-
ries of information are necessary to con-
struct the energy and materials input-
output analysis that is at the core of tradi-
tional life-cycle inventories.
Inputs
(Requirements)
Outputs
(Environmental
releases and
products)
Raw or intermediate
materials
Energy
Water
Other inputs
Atmospheric
emissions
Waterborne wastes
Solid wastes ,
Other releases
Products
Coproducts
40
-------
General Issues
The concept of life-cycle studies has been
extended in recent years beyond merely
specifying physical quantities to a more
comprehensive characterization. To accom-
modate this extension, the life-cycle analyst
may want to augment the traditional tem-
plate with additional information. The data
and information that describe the more
extensive inputs or outputs should be
based on the study purpose, with care
taken to ensure that the life-cycle inventory
remains a data-driven accounting proce-
dure. The additional data could include
categories of information not traditionally
considered in a material balance, such as
noise, aesthetics, and odors, and more
broadly interpreted emissions such as
workplace releases and land use changes.
Figure 7 illustrates the steps or processes
included in a life-cycle inventory for the
bar soap system. Individual steps within
this overall flow diagram are referred to as
subsystems. The bar soap production step
is one example of a subsystem. All subsys-
tem steps, including those for secondary/
tertiary packaging, together form the bar
soap system. The template can be applied
at either the subsystem or system level.
As the bar soap production subsystem illus-
trates, numerous processes or subprocesses
are included such as making fatty acids,
vacuum distillation, making toilet soap,
cutting, and drying. In some cases, the tem-
plate will be used for data gathering at the
subsystem level without evaluating the pro-
cesses or subprocesses within the subsys-
tem. The decision to gather data at the
subsystem level will depend on the nature
Energy
Water
1
Raw or
Intermediate
Materials
Products
Atmospheric
Emissions
Transportation
- *~ Coproducts
I
Waterborne
Wastes
Solid Waste
Figure 6
Life-cycle inventory template
Source: Franklin Associates, Ltd.
41
-------
General Issues
Soil Preparation,
Seeds, Fertilizers,
Pesticides
Harvesting and
Processing of Silage,
Grains, and Hay
Seedlings
and Seeds
Planted Forest
Harvesting
Meat Packing
and Rendering
Chlorine
Production
"Natural" Forest
Harvesting
Tallow
Production
Sodium
Hydroxide
Production
Bar Soap
Production
Paper
Production
Soap Packaging
Cardboard
Production
Cardboard
Recycler
Postconsumer
Waste
Management
Tallow Hot Water
I I
Oils
Fragrances
Colors
Glycerine Fatty Acids
Sodium
Hydroxide
Note: Energy
acquisition and
electricity generation
are not shown on this
diagram, although they
are inputs to many of
these processes.
Bar Soap Production
Figure 7
Detailed system flow diagram for bar soap
42
-------
General Issues
of the subsystem and the availability of
data. If a specific manufacturing facility has
total energy and emissions data for the sub-
system, but not specific process data, then it
makes sense to apply the template to the
entire manufacturing facility (e.g., the bar
soap facility in Figure 7) and not to each
individual process. In other cases these
individual processes must be examined
separately, then totaled to provide the sub-
system inventory. The following sections
describe each of the template components
and identify how current life-cycle studies
apply criteria to each of these areas.
Inputs in the Product Life-Cycle
Inventory Analysis
Raw/Intermediate Materials
The input materials for each subsystem are
referred to as raw/intermediate materials.
Raw materials are materials that have been
extracted from the earth but have not been
refined or processed. Intermediate materials
are products of one refining or manufactur-
ing step that are input materials into
another process. The most complete inven-
tory will begin with all raw materials at the
input of the most upstream stage. Any
inventory that sets a different boundary
should provide the associated reasons and
justification for excluding steps.. Figure 7
shows that tallow and sodium hydroxide
are the two intermediate materials needed
for bar soap production, whereas salt is a
raw material input for sodium hydroxide
production. The raw/intermediate material
requirements for a subsystem represent the
total material inputs, including all material
present in the product and material found
in losses from emissions, scrap, and off-spec
products as well as non-emission losses
(such as moisture due to evaporation). The
template is used to guide the documentation
of material requirements in pounds or kilo-
grams per unit of product output for each
subsystem.
Raw/intermediate materials have been dis-
tinguished from other process inputs by
some practitioners because raw/
intermediate materials are present in the
finished product, although they may be
transformed chemically. For example, tal-
low, oils, fragrances, and colors are input
materials for bar soap production. Water
does not always appear as a raw/
intermediate material input. Water may be
reclaimed during a drying step even if it is
used in the process and, therefore, may not
be present in the finished product.
The decision on which raw/intermediate
material requirements to include in a life-
cycle inventory is complex, but several
options are available:
Incorporate all requirements, no matter
how minor, on the assumption that it is
not possible, a priori, to decide to exclude
anything.
Within the defined scope of the study,
exclude inputs of less than a predeter-
mined and clearly stated threshold.
Within the defined scope of the study,
exclude inputs determined likely to be
negligible, relative to the intended use of
the information, on the basis of a sensitiv-
ity analysis.
Within the defined scope, consistently
exclude certain classes or types of inputs,
such as capital equipment replacement.
The advantage of the first option is that no
assumptions are made in defining and
43
-------
General Issues
drawing the system boundary. The analyst
does not have to explain or defend what
has heen included or excluded. The disad-
vantage is that application of this approach
could be an endless exercise. The number
of inputs could be very large and could
include some systems only distantly related
to the product system of interest. Besides
the computational complexity, the interpre-
tation of the results with respect to the
single desired product, package, or activity
could be difficult.
The second option, if implemented with
full explanation of what the threshold is
and why it was selected, would have the
advantages of consistency and lower cost
and time investments. Two suboptions can
be identified, depending on the nature of
the threshold. One suboption is to specify a
percentage contribution below which the
material will be excluded, for example, 1%
of the input to a given subsystem or to the
entire system. The 1% rule historically has
been useful in limiting the extent of the
analysis in inventories where the environ-
mental consequences of quantitatively
minor materials are not considered. The
disadvantage of the 1% rule is that the pos-
sible presence of an environmentally dam-
aging activity associated with these
materials could be overlooked. Also, when
used with mixed percentages (e.g., percent
of system energy, percent of subsystem
input), the results may be confusing or
inconsistent. The scoping analysis should
provide a rationale for choosing to apply
such a rule.
The second suboption is to set a threshold
based on the number of steps that the raw/
intermediate material is removed from the
maui process sequence. Consider the bar
soap example discussed earlier. Caustic
manufacture from brine electrolysis is part
of the main process sequence and would
clearly be included. Sodium carbonate is an
input material for the production of caustic
and is therefore a secondary input. Apply-
ing a "one-step back" decision rule would
include the steps associated with sodium
carbonate production. Ammonium chloride
is an input material for the production of
sodium carbonate using the Solvay process.
Relative to caustic, ammonium chloride is a
tertiary input and would be excluded if a
"one-step back" decision rule were applied.
As in the first option, the "one-step back"
decision rule has the advantages of clarity
and consistent application. For some inputs
that are analyzable in exact mathematical
terms, the "one-step back" rule may be jus-
tifiable. If the inputs to a given process bear
a fixed relationship to the next-tier process,
one step is all that may be necessary to
obtain a sufficiently accurate value
(Boustead and Hancock, 1979).
Consider the example of a refinery. Most of
the refinery's output is sold for production
of petroleum-based materials. However, a
small portion, say 8%, is used to run the
refinery. This portion, termed the parasitic
fraction, is mathematically related to the
refinery output as:
where:
M is the output product and
f is the parasitic fraction (0.08).
For a life-cycle inventory on a petroleum-
based plastic, the primary output of the
refinery clearly would be included within
the system boundary. Suppose the data
quality indicators showed that the data
44
-------
General Issues
were accurate to ±5%. Because the first-tier
use of the material represents an 8% differ-
ence, a "one-step back" rule would include
the refinery material (fuel) output used to
run the refinery. However, to produce the
material (fuel) to run the refinery requires a
further fraction of the output two steps
back from the plastic raw material. This is
calculated as:
M(l + f+P).
Thus, the incremental contribution of the
second step back is 0.6%, which is less
than the data accuracy. That is, there is no
significant difference in the system data
after the first step. Disadvantages of this
approach include the lack of simple geome-
tric relationships for many inputs and the
increased effort to analyze more tiers as
data quality increases.
The third option, drawing boundaries
based on sensitivity analysis, adds the
advantage of being systematic rather than
arbitrary in assigning the threshold. The
disadvantages of a sensitivity analysis-
based approach are that the analyst needs
to be very clear in describing how the
analysis was used and, unless a large exist-
ing database is available to supply prelimi-
nary values that can be used in the
sensitivity analysis, the required analysis
effort may not be limited by a very large
amount. A more in-depth discussion of sen-
sitivity analysis is provided later in this
chapter.
The final option, excluding certain classes
or types of input, also has been found
through experience to apply to many sys-
tems. For example, in the bar soap inven-
tory, a decision may be made to exclude the
equipment used to cut the bars of soap. The
justification is that the allocation of inputs
and outputs from the manufacture of the
machine is miniscule when the millions of
bars of soap produced by the machine are
considered. The advantage of this option is
that many complex subsystems can often be
excluded. The disadvantages are the same
as those for the first option, namely, that a
highly significant activity may be elimi-
nated. Capital equipment is the most com-
monly excluded input type. The analyst
should perform a preliminary analysis to
characterize the basic activities in each
class or type of input to ensure that a signi-
ficant contribution is not left out.
Renewable and Nonrenewable Resources
A renewable resource is one that is being
replaced in the environment in a time
frame relevant to society. Certain species of
wood are examples of such a material. Most
minerals and metals, along with certain
biological products, are of such a stable
nature that their supply does not change
within a tune perspective of several genera-
tions, and they are thus deemed nonrenew-
able (Jorgensen and Pedersen, 1990).
Hydrocarbon fuels produced from geologi-
cal repositoriescoal, petroleum, and natu-
ral gasare nonrenewable because they are
not being created in a tune frame of rele-
vance to humans. If production of hydro-
carbon fuels from biomass were to become
a significant source of fuels, the definition
would be applied to the non-fossil-sourced
fraction as well as to that from geological
sources.
The intrinsic nature of a resource does not
dictate its renewability. A further definition
of a sustainable resource may be applied if
the rate of replacement exceeds the rate of
use. The potential for being replaced in a
relevant time frame is not itself sufficient to
qualify a material as "renewable." The
45
-------
General Issues
replacement or renewal must actually be
occurring. There is still discussion on
renewability issues and whether resource
renewability as opposed to resource con-
sumption belongs in inventory analyses or
impact analyses. It is recommended that
the inventory track and report each
resource individually. Based on the lack of
consensus among practitioners, at this tune
additional designations of natural resources
into categories should be considered part of
the impact analysis.
Energy
Energy as shown on the template (Figure 6)
represents a combination of energy require-
ments for the subsystem. Three categories
of energy are quantifiableprocess, trans-
portation, and energy of material resources
(inherent energy).
Process energy is the energy required to
operate and run the subsystem process(es),
including such items as reactors, heat
exchangers, stirrers, pumps, blowers, and
boilers. Transportation energy is the energy
required to power various modes of trans-
portation such as trucks, rail carriers,
barges, ocean vessels, and pipelines. Con-
veyors, forklifts, and other equipment that
could be considered either transportation
or process are labeled according to their
role in the subsystem. For example, the
power supplied to a conveyor used to carry
material from one point in the subsystem to
another point in the same subsystem would
be labeled process energy. On the other
hand, the power supplied to a conveyor
used to transport material from one subsys-
tem to a different subsystem would be con-
sidered transportation energy. Energy of
material resources is described separately
below.
Two alternatives exist for incorporating
energy inputs in a subsystem module. One
is to report the actual energy forms of the
inputs, e.g., kilowatt-hours of electricity or
cubic feet of natural gas. The other is to
include the specific quantities of fuels used
to generate the produced energy forms in
the module.
The advantage of the first approach is that
the specific energy mix is available for each
subsystem. For example, a company may
want to evaluate the desirability of install-
ing a natural gas-fired boiler to produce
steam compared to using its electrically
heated boiler powered by a combination of
purchased and on-site-generated electricity.
A specific fuel mix could be applied to
compute the energy and fuel resource use.
The second approach, incorporating spe-
cific fuel quantities, allows a subsystem
comparison of primary! energy fuels. For
example, "x" kWh of electricity would be
specified as "y" ft3 of natural gas and "z" Ib
of uranium.
Within each subsystem, the energy input
data should be given as specific quantities
of fuel and then converted into energy
equivalents according to the conversion
factors discussed in the following two sec-
tions. For example, the energy require-
ments attributed to a polyethylene resin
plant may be specified as 500 Ib of ethylene
for feedstock, 500 ft3 of natural gas, and
50 kWh of electricity to run the process
equipment, and 50 gal of diesel fuel to
transport the resin to consumers. In this
case, the 50 kWh would be converted to
180 MJ.
Combustion and Precontbustion Values. To
report all energy usage associated with the
subsystem of concern in the template, the
46
-------
General Issues
analyst may need to consider energy data
beyond the primary process associated with
combustion of the fuel. The energy used in
fuel combustion is only part of the total
energy associated with the use of fuel. The
amount of energy expended to acquire the
fuel also may be significant in comparison to
other energy expenditures. Energy to acquire
fuel raw materials (e.g., mining coal or drilling
for oil), process these raw materials into
usable fuels, and transport them is termed by
various practitioners as "precombustion
energy" or "energy of fuel acquisition". Pre-
combustion energy is defined as the total
amount of energy necessary to deliver a usable
fuel to the consumer of the fuel.
Including precombustion energy is analo-
gous to extending the system boundaries for
fuels to raw material inputs. For example,
suppose the combustion of fuel oil in an
industrial boiler results in the release of
about 150,000 Btu per gallon. However,
crude oil*drilling and production, refining,
and transporting the fuel oil require an
additional 20,000 Btu per gallon. This addi-
tional energy is the precombustion energy.
Thus, the total energy expended (precom-
bustion energy plus combustion energy)
when a gallon of fuel oil is consumed would
be 170,000 Btu. Generally, a complete
inventory will include precombustion
energy contributions because they represent
the true energy demand of the system.
Inclusion or exclusion of this contribution
should be clearly stated.
Energy Sources. Energy is obtained from a vari-
ety of sources, including coal, nuclear power,
hydropower, natural gas, petroleum, wind,
solar energy, solid waste, and wood biomass.
Fuels are interchangeable, to a high degree,
based on their energy content. For example,
an electric utility decides which fuel or other
energy source to use based on the cost per
energy unit. Utilities can and do use multiple
forms of energy sources, making possible an
economic decision based on the energy cost
per kilowatt-hour of electricity generated.
Manufacturing companies also choose among
energy sources on the same basis. However,
reasons other than cost, such as scarcity or
emissions to the environment, also affect the
energy source decision. For example, during
periods of petroleum shortages, finding prod-
ucts that use predominantly nonpetroleum
energy sources may be desirable. For that rea-
son, the inventory should characterize energy
requirements according to basic sources of
energy. Thus, it would consider not only elec-
tricity, but also the basic sources (such as coal,
nuclear power, hydropower, natural gas, and
petroleum) that produce the electricity.
Electricity. Considerations associated with
electricity include the source of fuel used to
generate the electricity and the efficiency of
the generating system. Power utilities typi-
cally use coal, nuclear power, hydropower,
natural gas, or oil to generate electricity.
Non-utility generation sources can include
wind power, waste-to-energy, and geo-
thermal energy. Accurately determining
electrical energy use and associated emis-
sions raises several complications, such as
relating the actual electricity use of a single
user to the actual fuel used.
Although a given company pays its bills to a
particular utility, the company is not simply
purchasing power from the nearest plant.
Once electricity is generated and fed into
power lines, it is indistinguishable from
electricity from any other source. Individual
generating stations owned by a given utility
may use different fuels. The electricity gen-
erated by these stations is "mixed" in the
transmission lines of that utility. The utility
47
-------
General Issues
is interconnected with neighboring utilities
(also using various types of fuel), to form
regional grids, which then interconnect to
form a national grid.
Computational models currently used to
perform life-cycle inventories of electricity
in the USA are based on the fuel mix in
regional grids or on a national average. In
many cases where an industry is scattered
throughout the USA, the fuel mix for the
national grid (available from the U.S.
Department of Energy) can be used, making
calculations easier without sacrificing accu-
racy. Data for 1991 are shown in Table 1.
Table 1. U.S. National Electrical Grid Fuel Mix
for 1991 w*
Gigawatthours
Fuel
Coal
Nuclear
Hydro
Natural Gas
Oil
Other10'
(GWh)
1,553,581
616,759
291,657
264,478
112,146
10,339
Percent
54.5
21.7
10.2
9.3
3.9
0.4
Total
2,848,960 100.0
(a) Sources: U.S. Department of Energy, Energy Information
Agency, 1992; Canadian Electric Utilities and Natural
Energy Board, 1991.
(b) Canadian data are estimates for 1991 based on 1990 gen-
eration and export Canadian exports are 0.9% of totals
and were equally allocated across fuel types.
(c) Includes wood and waste-to-energy sources but excludes
independent generators and minor sources
(e.g., geothormal).
One exception to the national grid assump-
tion is the electroprocess industries, which
use vast amounts of electricity. Aluminum
smelting is the primary example. It and the
other electroprocessing industries are not
distributed nationally, so a national electric-
ity grid does not give a reasonable approxi-
mation of their electricity use. They are
usually located in regions of inexpensive
electric power. Some plants have purchased
their own electric utilities. In recognition of
this fact, specific regional grids or data from
on-site facilities are commonly used for life-
cycle inventories of the electroprocessing
industries.
The energy inefficiency of the electricity-
generating and delivery system must also be
considered. The theoretical conversion from
the common energy unit of kilowatt-hours
(kWh) to common fuel units (Joules) is
3.61 MJ per kWh. Ideally, the analyst would
compute a specific efficiency based on the
electrical generation fuel mix actually used.
This value is derived by comparing the
actual fuels consumed by the electricity-
generating industry in the appropriate
regional or national grid to the actual
kilowatt-hours of electricity delivered for
useful work. The value includes boiler inef-
ficiencies and transmission line losses.
*
However, a conversion of 11.3 MJ per kWh
may be used in most cases to reflect the
actual use of fuel to deliver electricity to the
consumer from the national grid.
Nuclear Power. Nuclear power substitutes
for fossil fuels in the generation of electric-
ity. There is no measurement of nuclear
power directly equivalent to the joules of
fossil fuel, so nuclear power typically is
measured at its fossil fuel equivalency. The
precombustion energy of nuclear power is
usually added to the fuel equivalency value.
The precombustion energy includes that for
mining and processing, as well as the
increased energy requirements for power
plant shielding.
Hydropower. Most researchers traditionally
have counted hydropower at its theoretical
energy equivalence of 3.61 MJ per kWh,
48
-------
General Issues
with no precombustion impacts included.
No precombustion factors are used for
hydropower because water does not have an
inherent energy value from which line
transmission losses, etc., can be subtracted.
The contribution of the capital equipment is
small in light of the amount of hydroelectric
energy generated using the equipment. Dis-
ruption to ecosystems typically has not been
considered in the inventory. However,
quantitative inventory measures that may be
suitable for characterizing related issues,
such as habitat loss due to land use conver-
sion, potentially could be included. Factors
addressing area! damage, recovery time, and
ecosystem function are under consideration
for inclusion hi the impact analysis.
Energy Content of Material Resources. Many
materials analyzed in a life-cycle inventory
have an energy content. From athermbdy--
namic perspective, it is important to ensure
that energy balance is maintained in each sub-
system. However beyond this, further distinc-
tions can be drawn.
The energy of material resources, also
known as fuel-related inherent energy or
latent energy of materials, accounts for
those products or materials that consume
raw materials whose alternative use is as a
fuel or energy resource (e.g., oil, natural gas,
and coal). Some materials are made from en-
ergy resources and, as stated earlier, inher-
ent energy is a measure of the energy
implications of the decision to forego use of
the resource as fuel. The primary example is
plastics, which are made from petroleum
and natural gas. Because the actual plastic
materials contain energy resources, result-
ing in a reduction of the planet's finite
reserves, an energy value is assigned to the
plastic material hi addition to the other
types of basic energy forms associated with
plastic production, such as process and
transportation energy. This additional
energy value is equivalent to the fossil fuel
combustion value of petroleum and natural
gas.
In assessing the energy content issue for
other material resources, different alterna-
tives have been used by various researchers.
The first option involves asking whether or
not a given material is viewed as an energy
resource. For materials such as coal, natural
gas, and petroleum, there is no question.
However, hi the case of wood, textiles, and
biomass crops, the answer is not as clear-
cut. In the USA, wood generally is not
viewed as a primary energy resource, so it
typically has not been counted as such for
U.S.-based studies. However, in performing
studies in areas of the world where wood is
a major energy resource, or for U.S. studies
involving activities hi these areas, it would
be treated as a primary energy source.
The second option includes all raw materi-
als having an energy content. This approach
has been more widely but not universally
used hi Europe (Tillman, et al., 1991;
Lundholm and Sundstrom, 1985). A third
option, combining some features of the first
two, is to track non-fuel inherent energy as a
separate category.
Residues and Renewable Energy Sources. An
important energy issue is the use of residues
and energy sources from manufacturing
operations, particularly those from agricul-
tural or forest product operations. These
sources of energy are listed commonly in the
energy profile or characterization so that fur-
ther analysis in the impact study can be made.
However, the life-cycle inventory does not
give special credit or benefit for the use of
renewable energy resources. Renewable
49
-------
General Issues
resource definition was discussed previously
as a general raw materials issue.
Geographic Scope. Energy is an international
product. All kinds of fuels are imported and
exported, and electricity passes easily across
national boundaries. Much of the crude oil
used in the USA, for example, is produced in
Middle Eastern countries such as Saudi
Arabia. Historically, data on inputs and out-
puts associated with acquiring oil often have
not been available for non-U.S. sources.
Where data related to procurement of energy
from a foreign source are not available, the
approach has been to apply U.S. data as an
initial estimate. When this approach is used,
the analyst should clearly state the
assumptions.
Energy from Waste Combustion. When waste is
burned, energy can be recovered. The ques-
tion is how to properly include the energy in
an inventory. The energy value of combustible
waste, whether industrial or postconsumer,
historically is counted as the higher heating
value (HHV) of the materials in the combusted
waste with proper adjustment for moisture,
just as fossil fuels are counted. However, there
is no theoretical reason why the actual ther-
mal yield of a waste of known composition
cannot be determined. To calculate the
thermal yield would require offsetting the
HHV with both the moisture factor loss and
the incinerator losses. Historically, the energy
yield reduction associated with noncombusti-
ble materials introduced into an incinerator
also has not been debited. Energy must be
supplied to heat up these materials to the
incinerator operating temperature. For post-
consumer waste, proper accounting of the per-
centage sent to waste-to-energy facilities needs
to be made. This energy value is then credited
against the system energy requirements for the
primary product, resulting in net energy
requirements that are less than the total
energy requirements for the system.
Water. Water volume requirements should be
included in a life-cycle inventory analysis. In
some parts of the country, water is plentiful.
Along the coasts, seawater is usable for cool-
ing or other manufacturing purposes. How-
ever, in other places water is hi short supply
and must be allocated for specific uses. Some
parts of the country have abundant water in
some years and limited supplies in other
years. Some industrial applications reuse
water with little new or makeup water
required. In other applications, however,
tremendous amounts of new water inputs are
required. j
How should water be incorporated in an
inventory? The goal of the inventory is to
measure, per unit of product, the gallons of
water required that represent water unavail-
able for beneficial uses (such as navigation,
aquatic habitat, and drinking water). Water
withdrawn from a stream, used in a process,
treated, and replaced in essentially the same
quality and in the same location should not
be included in the water-use inventory data.
Ideally, water withdrawn from groundwater
and subsequently discharged to a surface
water body should be included, because the
groundwater is not replaced to maintain its
beneficial purposes. Data to make this dis-
tinction may be difficult to obtain in a
generic study where site-specific informa-
tion is not available.
In practice, the water quantity to be esti-
mated is net consumptive usage. Consump-
tive usage as a life-cycle inventory input is
the fraction of total water withdrawal from
surface or groundwater sources that either is
incorporated into the product, coproducts
(if any), or wastes, or is evaporated. As in
so
-------
General Issues
the general case of renewable versus nonre-
newable resources, valuation of the degree
to which the water is or is not replenishable
is best left to the impact analysis.
Outputs of the Product Life-Cycle
Inventory Analysis
A traditional inventory quantifies three cat-
egories of environmental releases or emis-
sions: atmospheric emissions, waterborne
waste, and solid waste. Products and
coproducts also are quantified. Each of
these areas is discussed in more detail in
the following sections. Most inventories
consider environmental releases to be
actual discharges (after control devices) of
pollutants or other materials from a process
or operation under evaluation. Inventory
practice historically has included only regu-
lated emissions for each process because of
data availability limitations. It is recom-
mended that analysts collect and report all
available data in the detailed tabulation of
subsystem outputs. In a study not intended
for product comparisons, all of these pollu-
tants should be included hi the summary
presentations.
A comparative study offers two options.
The first is to include in the summary pre-
sentation only data available for all alterna-
tives under consideration. The advantage of
this option is that it gives a comparable pre-
sentation of the loadings from all the alter-
natives. The disadvantage is that potentially
consequential information, which is avail-
able only for some of the alternatives, may
not be used. The second option is to report
all data whether uniformly available or not.
In using this option, the analyst should cau-
tion the user not to draw any conclusions
about relative effects for pollutants where
comparable data are not available. "Compar-
able" is used here to mean the same pollut-
ant. For example, in a summary of data on a
bleached paper versus plastic packaging
alternatives, data on dioxin emissions may
be available only for the paper product. The
second option is recommended for internal
studies and for external studies where
proper context can be provided. A discus-
sion of which pollutants are associated with
various regulations is included in the
appendix.
Atmospheric Emissions
Atmospheric emissions are reported in units
of weight and include all substances classi-
fied as pollutants per unit weight of product
output. These emissions generally have
included only those substances required by
regulatory agencies to be monitored but
should be expanded where feasible. The
amounts reported represent actual dis-
charges into the atmosphere after passing
through existing emission control devices.
Some emissions, such as fugitive emissions
from valves or storage areas, may not pass
through control devices before release to the
environment. Atmospheric emissions from
the production and combustion of fuel for
process or transportation energy (fuel-
related emissions), as well as the process
emissions, are included hi the life-cycle
inventory.
Typical atmospheric emissions are particu-
lates, nitrogen oxides, volatile organic com-
pounds (VOCs), sulfur oxides, carbon
monoxide, aldehydes, ammonia, and lead.
This list is neither all-inclusive nor is it a
standard listing of which emissions should
be included in the life-cycle inventory.
Recommended practice is to obtain and
report emissions data in the most speciated
51
-------
General Issues
form possible. Some air emissions, such as
participates and VOCs, are composites of
multiple materials whose specific makeup
can vary from process to process. All emis-
sions for which there are obtainable data
should be included in the inventory. There-
fore, the specific emissions reported for any
system, subsystem, or process will vary
depending on the range of regulated and
nonregulated chemicals.
Certain materials, such as carbon dioxide
and water vapor losses due to evaporation
(neither of which is a regulated atmospheric
emission for most processes), have not been
included in most inventory studies in the
past. Regulations for carbon dioxide are
changing as the debate surrounding the
greenhouse effect and global climate change
continues and the models used for its pre-
diction are modified. Inclusion of these
emerging emissions of concern is
recommended.
Waterbome Wastes
Waterborne wastes are reported in units of
weight and include all substances generally
regarded as pollutants per unit of product
output. These wastes typically have
included only those items required by regu-
latory agencies, but the list should be
expanded as data are available. The effluent
values include those amounts still present
in the waste stream after wastewater treat-
ment and represent actual discharges into
receiving waters. For some releases, such as
spills directly into receiving waters, treat-
ment devices do not play a role in what is
reported. For some materials, such as brine
water extracted with crude oil and rein-
jected into the formation, current regula-
tions do not define such materials as
waterborne wastes, although they may be
considered in solid waste regulations under
the Resource Conservation and Recovery
Act (RCRA). Other liquid wastes may also
be deep well injected and should be
included. In general, the broader definition
of emissions in a life-cycle inventory, in
contrast to regulations, would favor inclu-
sion of such streams. It can be argued, from
a systems analysis standpoint, that materials
such as brine should count as releases from
the subsystem because they cross the sub-
system boundary. If wastes and spills that
occur are discharged to the ocean or some
other body of water, these values are always
reported as wastes.
As with atmospheric wastes, waterborne
wastes from the production and combustion
of fuels (fuel-related emissions), as well as
process emissions, are included in the life-
cycle inventory.
Some of the most commonly reported water-
borne wastes are biological oxygen demand
(BOD), chemical oxygen demand (COD),
suspended solids, dissolved solids, oil and
grease, sulfides, iron, chromium, tin, metal
ions, cyanide, fluorides^ phenol, phos-
phates, and ammonia. Again, this listing of
emissions is not meant to be a standard for
what should be included in an inventory.
Some waterborne wastes, such as BOD and
COD, consist of multiple materials whose
composition can vary from process to pro-
cess. Actual waterborne wastes will vary for
each system depending on the range of regu-
lated and nonregulated chemicals.
Solid Waste
Solid waste includes all solid material that
is disposed from all sources within the sys-
tem. The regulations include certain liquids
and gases in the definition as well. Solid
wastes typically are reported by weight.
52
-------
General Issues
Some analysts convert the weight to volume
using representative landfill density factors.
By estimating landfill volumes, researchers
can report the space occupied by the waste
in a landfill. If volume factors are used, both
weight and volume should be reported to
enhance comparability among studies.
Distinction is made in data summaries
between industrial solid wastes and post-
consumer solid wastes, as they are generally
disposed of in different ways and, in some
cases, at different facilities. Industrial solid
waste refers to the solid waste generated
during the production of a product and its
packaging and is typically divided into two
categories: process solid waste and fuel-
related solid waste. Postconsumer solid
waste refers to the product/packaging once
it has met its intended use and is discarded
into the municipal solid waste stream.
Process solid waste is the waste generated
in the actual process, such as trim or waste
materials that are not recycled as well as
sludges and solids from emissions control
devices. Fuel-related waste is solid waste
produced from the production and combus-
tion of fuels for transportation and operating
the process. Fuel combustion residues, min-
eral extraction wastes, and solids from util-
ity air control devices are examples of
fuel-related wastes.
Mine tailings and overburden generally are
not regulated as solid waste. However, the
regulations require overburden to be
replaced in the general area from which it
was removed. Furthermore, environmental
consequences associated with the removal
of mine tailings and overburden should be
included. The regulations do not require
industrial solid waste to be handled off site.
Therefore, researchers try to report all solid
waste from industrial processes destined for
disposal, whether off site or local. Histori-
cally, no distinctions have been made
between hazardous and nonhazardous solid
waste, nor have individual wastes been spe-
cifically characterized. However, in view of
the potentially different environmental
effects, analysts will find it useful to
account for these wastes separately, espe-
cially if an impact analysis is to be
conducted.
Products
The products, as identified in the template,
are defined by the subsystem and/or system
under evaluation. In other words, each sub-
system will have a resulting product, with
respect to the entire system. This subsys-
tem product may be considered either a raw
material or intermediate material with
respect to another system, or the finished
product of the system.
Again using the bar soap system, when
examining the meat packing subsystem,
meat, tallow, hides, and blood would all be
considered product outputs. However,
because only tallow is used in the bar soap
system, tallow is considered the only prod-
uct from that subsystem. All other material
outputs (not released as wastes or emis-
sions) are considered coproducts. If the life-
cycle assessment were performed on a
product such as a leather purse, hides
would be considered the product from the
meat packing subsystem and all other out-
puts would be considered coproducts.
Although for bar soap the tallow is consid-
ered the product from the meat packing sub-
system, it is simultaneously an intermediate
material within the bar soap system. Thus,
from these examples one can see that
classifying a material as a product in a
-------
General Issues
life-cycle study depends, in part, on the
extent of the system being examined, i.e.,
the position from which the material is
viewed or the analyst's point of view. This
point of view should hecome clear when the
template is applied to each subsystem
within the total system under evaluation.
Transportation
The life-cycle inventory includes the energy
requirements and emissions generated by
the transportation requirements among sub-
systems for both distribution and disposal
of wastes. Transportation data are reported
in miles or kilometers shipped. This dis-
tance is then converted into units of ton-
miles or tonne-kilometers, which is an
expression involving the weight of the ship-
ment and the distance shipped. Materials
typically are transported by rail, truck,
barge, pipeline, and ocean transport. The
efficiency of each mode of transport is used
to convert the units of ton-miles into fuel
units (e.g., gallons of diesel fuel). The fuel
units are then converted to energy units,
and calculations are made to determine the
emissions generated from the combustion of
the fuels.
The template in Figure 8 shows that trans-
portation is evaluated for the product leav-
ing each subsystem. This method of
evaluating transportation avoids any inad-
vertent double-counting of transportation
energy or emissions. Transportation is
reported only for the product of interest
from a subsystem and not for any coprod-
ucts of the subsystem, because the destina-
tion of the copro ducts is not an issue. The
raw materials for the bar soap production
system (Figure 7), for example, include salt
from salt mining and trees from natural for-
est harvesting. Applying the template to
these two subsystems shows that the trans-
port of salt from the mining operation and
the transport of trees from the logging opera-
tion must be included in the data collected
for these subsystems. Logically, there is no
transport of raw materials into these subsys-
tems because the salt and trees were
attached to the earth prior to removal.
The salt is transported to chlorine/sodium
hydroxide plants, and the trees are trans-
ported to pulp mills. Applying the template
to these subsystems shows that the transport
of chlorine and sodium hydroxide from
those plants to pulp mills is peirt of the chlo-
rine production and sodium hydroxide sub-
systems. Likewise, the transport of pulp to
paper mills is part of the pulp mill subsys-
tem. The transport of raw materials, salt,
and trees into the subsystems (chlorine pro-
duction, sodium hydroxide production, and
pulp mills) now being evaluated has already
been accounted for in the evaluation of the
salt mining and natural forest harvesting
subsystems. Applying the template through-
out the bar soap system shows the evalua-
tion of transportation ending with the
postconsumer waste management subsys-
tem, where wastes may be transported to a
final disposal site.
Backhauling may be a situation where there
is some overlap between the transportation
associated with product distribution and the
transportation associated with recycling of
the product or a different product after con-
sumer use. A backhaul has been described
as occurring when a truck or rail carrier has
a profitable load in one direction and is
willing to accept a reduced rate for a move
in the return direction. Backhaul opportuni-
ties occur when the demand for freight
transportation in one area is relatively low
54
-------
Genera! issues
Actual product flow diagram
for the production of Products
'A' and *B'
Energy
3x109 Btu
*
1,600lb (
Raw or »J
Intermediate I
30 Ib
Atmospheric
Emissions
10(
Solid'
Water
600 gal
1
"> ja
I Transportation
1
10 Ib
Waterborne
Wastes
) Ib
Waste
1,000lb
Product 'A1
500 Ib
Product's1
Flow diagrams showing the
resources and environmental
releases allocated to the two
products.
Energy
2x109 Btu
1,067lb (
Raw or »|
Intermediate!
20 Ib
Atmospheric
Emissions '
Water
400 gal
] Transportation
'
\
. 7lb
Waterborne
Wastes
1,000 Ib
Product 'A'
67 Ib
Solid Waste
Coproduct Allocation for Product 'A'
533 Ib
Raw or
Intermediat
Materials
Energy
1 x 109 Btu
\
a*"l
\
10 Ib
Atmospheric
Emissions '
33
Solid
Water
200 gal
1
"\ Transportation 500 )b
J Coproduct 'B'
\ ' '
3lb
Waterborne
Wastes
Ib
Waste
Coproduct Allocation for Product'B'
Figure 8
Allocating resources and environmental burdens for a product and coproduct
Source: Franklin Associates, Ltd.
55
-------
General Issues
and carriers have a financial incentive to
move their vehicles, loaded or empty, to a
place where the demand for freight trans-
portation is higher. Due to the lowered
transportation rates, recycled materials,
especially paper and aluminum, are often
transported by backhauling. Thus, a carrier
may take a load of new paper from a mill to
customers in a metropolitan area and pick
up loads of scrap paper hi the same area to
bring them back to the mill. In this scenario,
backhauling may reduce the energy and
emissions associated with distribution of a
product by transferring the energy and emis-
sions that would be associated with the
empty return trip to the recycling stage.
Coproduct Allocation
Most industrial processes are physical and/
or chemical processes. The fundamentals of
life-cycle inventory are based on modeling a
system in such a way that calculated values
reasonably represent actual (measurable)
occurrences. Some processes generate mul-
tiple output streams in addition to waste
streams. Usually, only certain of these out-
put streams are of interest with respect to
the primary product being evaluated. The
term coproduct is used to define all output
streams other than the primary product that
are not waste streams and that are riot used
as raw materials elsewhere in the system
examined in the inventory. A basis for
coproduct allocation needs to be selected
with careful attention paid to the specific
items calculated. Figure 8 illustrates a com-
mon coproduct allocation scheme based on
mass, the most common allocation basis
used. However, each industrial system must
be handled on a case-by-case basis. No allo-
cation basis exists that is always applicable.
Coproducts are of interest only to the point.
where they no longer affect the primary
product. Subsequent refining of coproducts
is beyond the scope of the analysis, as is
transport of coproducts to facilities for fur-
ther refining. In effect, the boundary for the
analysis is drawn between the primary
product and coproducts, with all materials
and environmental loadings attributed to
coproducts being outside the scope of the
analysis. For example, the production of
fatty acids from tallow for soap manufacture
generates glycerine, a secondary stream that
is collected and sold. Glycerine, therefore, is
considered a coproduct, and its processing
and use would be outside the scope of the
bar soap analysis.
Basis for Coproduct Allocation
The first step is to investigate any complex
process hi detail and attempt to identify
unit subprocesses that produce the product
of interest. If sufficient detail can be found,
no coproduct allocation will be necessary.
The series of subprocesses that produce the
product can simply be summed. Many
metal manufacturing plants illustrate this
approach. In steel product manufacture, all
products are made by melting the raw mate-
rials, producing iron, and then producing
, raw steel. These steps are followed by a
series of finishing operations that are
unique to each product line. It is generally
possible to identify the particular
subprocesses in the finishing sequence of
each product and to collect sufficient data
to carry out the life-cycle inventory without
coproduct allocation. In many cases, a care-
ful analysis of unit systems will avoid the
need to make coproduct allocations. Still, in
some cases, such as a single chemical
reaction vessel that produces several differ-
ent products, there is no analytical method
56
-------
General Issues
for cleanly separating the subprocesses. In
this example, coproduct allocation is
necessary.
The analyst needs to determine the specific
resource and environmental categories
requiring study. For a given product, differ-
ent coproduct allocations may be made for
different resource and environmental cate-
gories. To find the raw materials needed to
produce a product, a simple mass balance
will help track the various input materials
into the output materials. For instance, if a
certain amount of wood is needed to pro-
duce several paper products and the analy-
sis concerns only one of the products, a
mass allocation scheme as shown hi Fig-
ure 8 will be used to determine the amount
of wood required for the target product.
If a process produces several different
chemical products, care must be taken in
the analysis. It will be necessary to write
balanced chemical equations and trace the
chemical stoichiometry from the raw mate-
rials into the products. A simple mass allo-
cation method frequently gives reasonable
results, but not always. In calculating
energy, heat of reaction may be the appro-
priate basis for allocating energy to the vari-
ous coproducts. These calculations can
become quite complex, but if the chemical
products being produced are similar, experi-
ence has shown that a simple mass alloca-
tion very closely approximates the results of
even more complex calculations.
If the various coproduct chemicals are quite
different in nature, some other allocation
method may be needed. For example, an
electrolytic cell can produce hydrogen and
oxygen from water. Each water molecule
requires 2 electrons to produce 2 hydrogen
atoms and 1 oxygen atom. On a macroscopic
basis, electricity that produces 1 mole (or
2 grams) of hydrogen only produces one-
half mole (or 16 grams) of oxygen. Thus, the
input electrical energy would be allocated
between the hydrogen and oxygen coprod-
ucts on a molar basis. That is, two-thirds of
the energy would be allocated to the hydro-
gen and one-third to the oxygen, resulting hi
an energy per unit mass for hydrogen that is
16 times that of oxygen. However, conserva-
tion of mass is used to determine the materi-
als requirements. Each mole of water
(18 grams) contains 2 grams of hydrogen
atoms and 16 grams of oxygen atoms, and
the dissociation of the water results in
2 grams of hydrogen and 16 grams of oxy-
gen. Thus, a mass allocation would be
appropriate for raw material calculations in
this example.
For environmental emissions from a
multiple-product process, allocation to dif-
ferent coproducts may not be possible. For
example, in a brine cell that produces
sodium, chlorine, and hydrogen as coprod-
ucts, it may be tempting to associate any
emissions containing chlorine with the
chlorine coproduct alone. However, because
the sodium and hydrogen are also produced
by the same cell and cannot be produced
from this cell without also producing chlo-
rine, all emissions should be considered as
joint wastes. The question arises as to how
to allocate chlorine emissions (as well as
other emissions) to all three products. There
is no entirely satisfactory solution, but com-
mon practice based on chemical engineer-
ing, chemistry, and physics experience is to
apply a mass allocation scheme as a reason-
able modeling technique.
It has been suggested that the selling price
of the coproducts could be used as a basis
for this allocation. This solution is not
57
-------
General Issues
entirely satisfactory because the selling
prices of the various coproducts vary greatly
with time and with independent competi-
tive markets for each coproduct. Although a
mass allocation basis may not be ideal, it is
a widely recognized practice and produces a
predictable and stable result.
It is necessary to carefully analyze each pro-
cess and determine a basis for coproduct
allocation based on the physical and chemi-
cal processes, and based on the resource
and environmental parameters under study.
However, when no such system can be
agreed upon, a simple mass allocation can
be used.
One final issue is the distinction between
marginal wastes and coproducts. In some
cases it is not clear whether a material is a
waste or a coproduct. A hypothetical exam-
ple might be a valuable mineral that occurs
as 0.1% of an ore. For each pound of min-
eral product, 999 Ib of unneeded material is
produced. This discarded material might
find use as a road aggregate. As such, it has
value and displaces other commercial aggre-
gates and appears to be a coproduct along
with the valuable mineral. However, its
value is so low that hi some cases it may
simply be dumped back on the ground
because of limited markets. Whether this
material is considered a waste or a coprod-
uct may have a significant effect on the
results of a product life-cycle inventory. It
does not seem reasonable to use a simple
mass allocation scheme here. It is more rea-
sonable to assume that all of the energy and
other resources and emissions associated
with this process are incurred because of
the desire for the valuable product mineral.
However, there are some cases where the
"waste" has marginal, but greater value than
the example used here. It becomes difficult
in these, instances to determine precisely
which of the coproduct allocation method-
ologies discussed above is most "correct".
One important role of ah inventory is to pro-
vide information upon which impact analy-
sis and improvement analysis can be based.
Li cases where there is no clear methodolog-
ical solution, the inventory should include
reasonable alternative calculations. It
remains at some later time to make the judg-
ments as to which of several reasonable
alternatives is the correct one. In any event,
it should be made clear what assumptions
were made and what procedures were used.
Industrial Scrap
One coproduct stream of particular interest
is industrial scrap. This term is used to spe-
cifically identify process wastes of value
(trim scraps and off-spec materials) that are
produced as an integral part of a manufac-
turing process. Further, the wastes have
been collected and used as input materials
for additional manufacturing processes. The
last criterion is that these scrap materials
have never been used as originally intended
when manufactured. For example, a com-
mon polyurethane foam product is seat
cushions for automobiles. The trim from
cutting the cushions is never incorporated
into seat cushions. Likewise, off-spec seat
cushions sold as industrial scrap are never
used as seat cushions, but are used as input
material for another process.
A careful distinction must be made between
industrial scrap and postconsumer waste for
proper allocation in the inventory. If the
industrial scrap is to be collected and used
as a material input to a production system
or process, it is credited in the life-cycle
inventory as a coproduct at the point where
it was produced. Unfortunately, systems
S3
-------
General Issues
that use material more efficiently, i.e., that
produce lesser amounts of salable coprod-
ucts, assume a higher percentage of the
upstream energy and releases using this
criterion.
When the consumption of a coproduct falls
within the boundaries of the analysis, it
must no longer be considered as a coprod-
uct, but as a primary product carrying with
it all the energy requirements and environ-
mental releases involved with producing it,
beginning with raw materials acquisition.
For example, a study of carpet underlay-
ment made from polyurethane scrap would
include the manufacturing steps for produc-
ing the polyurethane scrap. Its production
must be handled as is any other subsystem
of a life-cycle inventory. Industrial scrap
does not displace virgin raw materials,
because the consumption of the industrial
scrap redefines the system to include the
virgin materials for its production (isocya-
nates and polyalcohols in the case of poly-
urethane foam). Tallow (Figure 7) is another
example of a material that would be defined
as an industrial scrap/coproduct. Histori-
cally, the thinking has been that once a
material shifts from the waste category to
being a utilized material, or a coproduct,
then it should bear some of the burden
(energy, raw/intermediate material input,
and environmental releases) for its own
production.
DATA ISSUES
The gathering of data for each subsystem of
the inventory is one of the greatest chal-
lenges in the inventory process. The defined
purpose, goals, and objectives for the inven-
tory will in part determine how these data
issues are viewed or considered.
Data Quality
The quality of data used will significantly
influence results. Because of the importance
of data quality in life-cycle assessments,
EPA is specifically addressing this issue in a
separate guidelines document. This section
presents an overview of some of the major
data quality issues with respect to
inventories.
Because life-cycle inventories are data-
intensive and data quality can affect the out-
come of an assessment, the development of
uniform criteria is crucial for selection and
reporting of data sources and types. Some
basic objectives for data quality should be
specified by the analyst based on the
defined purpose for the study. From a data
perspective, life-cycle inventories can be
thought of as comprising two parts: a set of
process and activity measurements that are
amenable to standard statistical treatments,
and a set of assumptions and decision rules
for combining the data sets into a system.
This discussion of data issues applies pri-
marily to process and activity measure-
ments. Considerations should include age of
the data (because the technology on which
data are based can become obsolete), fre-
quency of data collection (ensuring that sea-
sonal or other variability in the system is
properly captured), and representativeness
of the data (inclusion of the mix of activities
that may contribute to an environmental
loading). In addition, more traditional indi-
ces of data quality (accuracy, precision,
detection limits, and completeness) should
be evaluated with regard to life-cycle inven-
tory applications.
The most accurate and recent data are desir-
able for performing a life-cycle inventory.
All data received must be critically
59
-------
General Issues
reviewed regarding the source and content
before the data are used. Much of the data
gathered for the performance of a life-cycle
inventory is actual industry data, either
direct facility measurements or indirect esti-
mates from published summaries. Thus,
accuracy is determined by the quality of the
measurement or estimation procedures
where all of the variables may not be known
or controlled, and by the averaging process
to obtain representative subsystems. Many
plants produce several products, often from
the same processes; thus some engineering
estimation is often involved in getting repre-
sentative data for a particular product or
process. Many plants may produce the same
product, but the processes, energy usages,
and environmental loads may differ among
plants. Thus, when data are gathered and
averaged, the resulting data may not be
characteristic of any existing plant.
Data confidentiality may also affect accu-
racy. Much of the data in a life-cycle inven-
tory are from industry, either directly or
indirectly. Ideally, companies using inven-
tories publicly would report all data. If the
inventory is to be publicly scrutinized and
traceable, the data generally are aggregated
to a more general level than if the data were
guaranteed to be confidential. However, use
of a peer review process at the most specific
and detailed data collection level can help
ensure that minimal concessions to accu-
racy loss are made and that variability mea-
sures are provided.
Specific and detailed data sources for all
steps of the life cycle of a product are not
always available, and researchers must
resort to more general, and perhaps less
accurate data sources, such as textbooks,
periodicals, and public databases, which
lack the level of detail desired for all the
steps. In particular, formal data quality cri-
teria often are not included in databases.
Lack of data can be overcome in several
ways. If the process in question is similar to
other processes for which data are available,
comparisons and estimates can be made.
Much of the time, processes for which data
are unobtainable or of uncertain quality rep-
resent a small portion of the entire system to
be analyzed. Performing and applying a sen-
sitivity analysis helps identify the relative
importance of a particular step and can
determine the amount of work necessary to
obtain data that meet the needs of the inven-
tory. At a minimum, the analyst should
identify uncertain data and, if possible, esti-
mate the degree of uncertainty.
Sensitivity Analysis
As noted in Chapter Three, sensitivity
analysis is a systematic procedure for esti-
mating the effects of data uncertainties on
the outcome of a computational model.
Applying sensitivity analyses to a life-cycle
inventory begins early during the establish-
ment of the boundaries and continues
throughout the remainder of the inventory.
It is not possible to establish a priori the sig-
nificance of any individual contribution to
the final inventory data set. The true test of
whether an element is significant to an
inventory is the sensitivity of the final result
to the element's inclusion or exclusion,
although experience allows trends to be
discerned.
Typically, a sensitivity analysis is con-
ducted by evaluating the range of uncer-
tainty in the input data and recalculating
the model's output to see the effect. Thus,
higher levels of uncertainty in strongly
influential variables will be less acceptable
if the objectives of the study are to be met.
Rules have been formulated to determine
6O
-------
General Issues
how much and hi which combinations the
inputs should be varied. As a rule if an esti-
mate of the true variability, such as a mea-
sured statistical variance, is known, it
should form the basis for the high- and low-
range uncertainty estimates. The 95% confi-
dence bounds are generally used for this
purpose. For many inputs, the variability
may not be known or may reflect variations
in the feedstock over the period of tune for
which the inventory is being prepared. It is
usual, in these instances, to vary the input
by a range around its expected mean value.
This may be 5% for some variables and an
order of magnitude or more for others.
Frequently, hi setting the boundaries of the
inventory, that is, in answering the question
of how far back to go, such order-of-
magnitude estimates may be used to decide
which input values require specification to
a higher level of accuracy and which may be
left "as is." Consider a simple, linear system
with four inputs and an output as follows:
j + (l-a1)A2] + B + C = D
where D is the total emission of an ah- pollu-
tant, and A, B, and C, are individual unit
process contributions to the total overall
amount of D. Each process may be further
divided into subprocesses. The fractions a.,
and 1 a^ determine the relative contribu-
tions from subprocesses to process A.
In setting up the boundaries, an analyst may
want to decide if a process should be
included or if order-of-magnitude estimates
are acceptable. By running the calculations
with and without a given process or by set-
ting upper and lower ranges for the contri-
bution, it is possible to decide whether a
variable has a large or small effect. The rules
for deciding simultaneous input variability
are not well defined. However, the probabil-
ity of all parameters simultaneously being at
one extreme of the uncertainty range is low.
Therefore, it is recommended that the ana-
lyst evaluate single variables at extremes of
then- estimated uncertainty range and then
consider simultaneous variability of only
the few most critical variables. These varia-
bles will either most influence the outcome
or have the greatest uncertainty. In choosing
which parameters to vary and hi what
amounts, the analyst should bear in mind
that sensitivity analysis is a descriptive pro-
cedure. That is, its purpose is to differenti-
ate more important and less important
inputs, not to quantify overall uncertainty
in the system. ,
Two purposes of sensitivity analysis are
served by including and reporting it in ev-
ery study. First, the analyst has a mecha-
nism for deciding whether to expend
additional effort to improve a critical data
value or better characterize a subsystem.
Second, decisionmakers have a map of the
study showing quantitative areas of uncer-
tainty. In many cases, the physical nature of
the system under consideration dictates
which inputs should be evaluated as a
group, because changes hi one variable may
determine the range of others.
Accuracy and Precision
Data on specific inputs and emissions are
preferable in an inventory. For example,
data from the manufacturing plants and
processes directly related to the material,
product, process, activity, or service under
consideration should be used whenever
possible. Actual data should be used rather
than estimates or regulatory limits. Assump-
tions and conventions used to gather and
report data should be consistent and
equitable.
61
-------
General Issues
Data should be collected at as detailed a
level as possible, which, allows for a more
detailed analysis and reporting, and all
emissions should be recorded at the same
time. If aggregation has taken place prior to
obtaining the data for the inventory, pre-
cluding disaggregation, it should be so
noted.
Data in existing life-cycle inventories are
sometimes reported in an aggregated form,
such as listing all VOCs together. However,
individual chemicals in such groups can
have very different environmental proper-
ties. An adequate interpretation of these
chemicals ha the context of a life-cycle
inventory requires that chemicals be listed
individually. If the available data do not
allow individual listing, or confidentiality
precludes individual listing, this condition
must be noted. Even more importantly, the
analyst should avoid the implication that
this type of categorical grouping can be
compared across materials or life-cycle
stages.
The various waste streams are characterized
by measuring the concentrations of chemi-
cals or of conventional parameters, e.g.,
COD and BOD, and the analytical methods
used are reported. The variability of the
measured data has to be taken into account.
This can be done by listing the range of con-
centrations, minimum and maximum val-
ues, or a computed statistical deviation from
the mean. For small data sets where statisti-
cal treatment would reveal the individual
data points, a more descriptive statement of
variability, e.g., less than a factor of two, is
acceptable if the data are proprietary. In a
number of cases, actual data will not be
available for the volume of some emissions
from waste treatment units. Sometimes it is
possible to estimate the partitioning of a
compound among the different media (air,
water, solid waste). When such estimates
are made, assumptions and calculations
have to be reported. The sensitivity of the
life-cycle inventory output to changes in
this estimation should be explained when
uncertainty exists about the accuracy of the
estimation.
Data Source Attribution
Currently, little consistency exists regarding
the specificity of source data used in life-
cycle inventories. The degree of specificity
needed is highly dependent on the scope of
the study, hi general, internal studies will
contain more site-specific data. However,
the analyst should recognize the difference
between the reporting of inventory data on a
non-site-specific basis and subsequent
impact analyses that may require additional
specificity. Further, no requirements have
existed regarding specification of the type of
source data used in studies. Because selec-
tion of varying qualities of source data can
materially affect the life-cycle inventory
results, the life-cycle inventoiry should doc-
ument in detail the data source, how it was
measured or calculated, and its type (e.g.,
average, worst-case, best-case, case-
specific), and should include a discussion
of its limitations, variability, and impact on
the study results.
Uncertainty
During the data collection process, some
gaps in data will likely be encountered. Sev-
eral situations may occur as a result:
Absence of data from one of several pro-
ducers of a particular product or mate-
rial. Where a variety of technologies exist,
it may be more appropriate to assume the
missing data are equal to the quantity
averaged over only the plants reporting,
-------
General Issues
assign that value to the missing data, and
then average the total (Fava et al., 1991).
For example, if only three of five produc-
ers report the data, then only three data
points will be averaged to represent the
entire process. The degree of complete-
ness should he reported in the results
presentation.
Lack of consistency on how many consti-
tuents are being recorded (i.e., not every
entity may collect the same data on a pro-
cess). The best approach is to document
the omission by making explicit that data
were missing, not that the value was zero
(i.e., during reporting, if summations are
used they should be footnoted to explain
that actual values may be higher or lower
due to missing data) (Fava et al., 1991).
Depending on conventions, data may be
reported as nondetectable or as less than
a certain value (the detection limit). If
nondetectable entries are used, the detec-
tion limits should be reported. If data are
reported as less than a certain value, that
value should be used. Total numbers
should be footnoted to indicate that
actual values may be lower.
Some analysts report that the margin of sig-
nificant difference in particular data catego-
ries for a life-cycle inventory is ±10-25%.
Sensitivity analyses examining the high and
low ranges of data points and weight ratios
support this assertion. However, indepen-
dent verification of these claims has not
been made. The nature of the overall
accounting error is predominantly system-
atic, and is thus determined by data sources
and methods rather than by measurement
randomness. In consequence, alternatives
where the solid waste volumes, for example,
differ by only 6% would not be significantly
different. The analyst should clearly com-
municate when the data do or do not sup-
port a conclusion about differences among
alternatives.
Representativeness
To ensure representativeness, characteris-
tics of the sample must correspond suffi-
ciently closely to the studied population to
represent that population. Some issues
regarding representativeness can be
resolved by reviewing the experimental
design and the way results are reported. At
a minimum, the degree to which the
reported data encompass the population
should be stated. Such a statement for the
bar soap example might be.'This study uses
data covering 83% of domestic tallow
production."
Occasionally inventory data are difficult to
obtain for certain steps of the system or
even for certain inputs or outputs of a sub-
system. It is important to evaluate critically
whether it is worth the time and effort to
obtain data for some of the minor subsys-
tems of the study. When literally thousands
of numbers are involved in an analysis,
most individual numbers contribute little to
the overall analysis, especially if they have
to do with a relatively minor component.
Sometimes data for similar processes can be
used to estimate the data for a minor com-
ponent of the system. If data are carefully
estimated in this way, potential error from
estimating the data can be minimized. The
importance of any data input on the results
can be determined by performing a sensitiv-
ity analysis.
63
-------
General Issues
Data Time Period
The time period that data represent should
he long enough to smooth out any devia-
tions or variations in the normal operations
of a facility. These variations might include
plant shutdowns for routine maintenance,
startup activities, and fluctuations in levels
of production. Often data are available for a
fiscal year of production, which is usually a
sufficient time period to cover such
variations.
Specific Data versus Composite Data
When the purpose of the inventory is to find
ways to improve internal operations, it is
hest to use data specific to the system that is
heing examined. These types of data are
usually the most accurate and also the most
helpful in analyzing potential improve-
ments to the environmental profile of a sys-
tem. However, private data typically are
guarded hy a confidentiality agreement, arid
must he protected from public use by some
means. Composite, industry-average data
are preferable when the inventory results
are to be used for broad application across
the industry, particularly in studies per-
formed for public use. Although composite
data may be less specific to a particular
company, they are generally more represen-
tative of an industry as a whole. Such com-
posite data can also be made publicly
available, are more widely usable, and are
more general in nature. Composite data can
be generated from facility-specific data in a
systematic fashion and validated using a
peer review process. Variability, representa-
tiveness, and other data quality indicators
can still be specified for composite data.
Geographic Specificity
Natural resource and environmental conse-
quences occur at specific sites, but there are
broader implications. It is important to
define the scope of interest (regional vs.
national vs. international) in an inventory.
A local community may be more interested
in direct consequences to itself than in
global concerns.
In general, most inventories done domesti-
cally relate only to that country. However, if
the analysis considers imported oil, the oil-
field brines generated in the Middle East
should be considered. It has been suggested
that the results of life-cycle inventories
indicate which energy requirements and
environmental releases (of the total environ-
mental profile of a product) are local. How-
ever, due to the fact that industries are not
evenly distributed, this segmenting can be
done only after an acceptable level of accu-
racy is agreed on. The United States,
Canada, Western Europe, and Japan have
the most accurate and most readily available
information on resource use and environ-
mental releases. Global aspects should be
considered when performing a study on a
system that includes foreign countries or
products, or when the different geographic
locations are a key difference among prod-
ucts or processes being compared. As a
compromise, when no specific geographical
data exist, practices that occur in other
countries typically are assumed to be the
same as for their domestic counterparts.
These assumptions and the inherent limita-
tions associated with their application
should be documented within the inventory
report. In view of the more stringent envi-
ronmental regulations in developed coun-
tries, this assumption, while necessary,
-------
General Issues
often is not correct. Energy use and other
consequences associated with importing
materials should also be included.
Technology Mixes/Energy Types
For inventory studies of processes using
various technology mixes, market share dis-
tribution of the technologies may be neces-
sary to accurately portray conditions for the
industry as a whole. The same is true of
energy sources. Most inventories can be
based on data involving the fuel mix in the
national grid for electricity. There are
exceptions, such as the aluminum electro-
processing industry previously discussed.
Variations of this kind must be taken into
account when applying the life-cycle inven-
tory methodology. Also, as previously men-
tioned, conditions can differ greatly across
international borders.
Data Categories
Environmental emission databases usually
cover only those items or pollutants
required by regulatory agencies to be
reported. For example, as previously men-
tioned, the question of whether to report
only regulated emissions or all emissions is
complicated by the difficulty in obtaining
data for unregulated emissions. In some
cases, emissions that are suspected health
hazards may not be required to be reported
by a regulatory agency because the process
of adding them to the list is slow. A specific
example of an unregulated emission is car-
bon dioxide, which is a greenhouse gas sus-
pected as a primary agent in global
warming. There is no current requirement
for reporting carbon dioxide emissions, and
it is difficult to obtain measured data on the
amounts released from various processes.
Thus, results for emissions reported in a
life-cycle inventory may not be viewed as
comprehensive, but they can cover a wide
range of pollutants. As a rule it is recom-
mended that data be obtained on as broad a
range as possible. Calculated or qualitative
information, although less desirable and
less consistent with the quantitative nature
of an inventory, may still be useful.
Routine/Fugitive/Accidental Releases
Whenever possible, routine, fugitive, and
accidental emissions data should be consid-
ered in developing data for a subsystem. If
data on fugitive and accidental emissions
are not available, and quantitative estimates
cannot be obtained, this deficiency should
be noted in the report of the inventory
results. Often estimates can be made for
accidental emissions based on historical
data pertaining to frequency and concentra-
tions of accidental emissions experienced at
a facility.
When deciding whether to include acci-
dents, they should be divided into two cate-
gories, based on frequency. For the
low-frequency and high-magnitude events,
e.g., major oil spills, tools other than life-
cycle inventory may be appropriate. Unus-
ual circumstances are difficult to associate
with a particular product or activity. More
frequent, lower magnitude events should be
included, with perhaps some justification
for keeping their contribution separate from
routine operations.
SPECIAL CASE BOUNDARY ISSUES
In all studies, boundary conditions limiting
the scope must be established. The areas of
65
-------
General Issues
capital equipment, personnel issues, and
improper waste disposal typically are not
included in inventory studies, because they
have heen shown to have little effect on the
results. Earlier studies did consider them in
the analysis; later studies have verified their
minimal contribution to the total system
profile. Thus, exclusion of contributions
from capital equipment manufacture, for
example, are not excluded a priori. The
decision to include or not to include them
should be clearly noted by the analyst.
Capital Equipment
The energy and resources that are required
to construct buildings and to build process
equipment should be considered. However,
for most systems, capital expenditures are
allocated to a large number of products
manufactured during the lifetime of the
equipment. Therefore, the resource use and
environmental effluents produced are usu-
ally small when attributed to the system of
interest. The energy and emissions involved
with capital equipment can be excluded
when the manufacture of the item itself
accounts for a minor fraction of the total
product output over the life of the
equipment.
Personnel Issues
Inventory studies focus on the comprehen-
sive results of product consumption, includ-
ing manufacturing. At any given site, there
are personnel-related effluents from the
manufacturing process as well as wastes
from lunchroom trash, energy use, air con-
ditioning emissions, water pollution from
sanitary facilities, and others. In addition,
inputs and outputs during transportation of
personnel from their residence to the work-
place can be significant, depending on the
purpose and scope of the inventory. In
many situations, the personnel conse-
quences are very small and would probably
occur whether or not the product were man-
ufactured. Therefore, exclusion from the
inventory may be justified. The analyst
should be explicit about including or
excluding this category. For these issues,
the goals of the study should be considered.
If the study is comparative and one option
is significantly different in personnel or
capital equipment requirements, then at
least a screening-level evaluation should be
performed to support an inclusion or exclu-
sion decision.
improper Waste Disposal!
For most studies it is assumed that wastes
are properly disposed into the municipal
solid waste stream or wastewater treatment
system. Illegal dumping, littering, and other
improper waste disposal methods typically
are not considered in life-cycle inventories
as a means of solid waste disposal. Where
improper disposal is known to be used and
where environmental effects are known or
suspected, a case may be made to include
these activities.
66
-------
Chapter Five
ISSUES APPLICABLE TO
SPECIFIC LIFE-CYCLE STAGES
INTRODUCTION
This chapter discusses the issues specific to
each of the four life-cycle stages, i.e., raw
materials and energy acquisition, manufac-
turing, use/reuse/maintenance, and
recycle/waste management. In addition, the
steps in manufacturing are discussed indi-
vidually. The subsystem boundaries, as
well as specific assumptions and conven-
tions, are discussed for each stage and step.
Major ConceptsRaw Materials
The resource requirements and environmen-
tal emissions are calculated for all of the
processes involved in acquiring raw materi-
als and energy. This analysis involves tracing
materials and energy back to their sources.
Consequences of the raw materials acquisi-
tion stage include nontraditionai inventory
outputs, such as land use changes, and non-
chemical releases, such as odor or noise. To
the extent they are quantifiable, such out-
puts may be incorporated.
When fuel sources become input materials
for a manufacturing process, an energy fac-
tor accounts for the unused energy inherent
in the fuel.
RAW MATERIALS
ACQUISITION STAGE
The life cycle of any product or material
begins with the acquisition of raw materials
and energy sources. For example, crude oil
and natural gas must be extracted from
drilled wells, and coal and uranium must
be mined before these materials can be pro-
cessed into usable fuels. All of these activi-
ties fall into the raw materials acquisition
stage.
Subsystem Boundaries
The subsystem boundaries for raw materi-
als acquisition encompass the actual pro-
cess(es) of acquiring the raw material, i.e.,
obtaining the material from the earth or
earth's surface as it naturally occurs. Raw
materials acquisition includes any energy
and water used and environmental releases.
Other consequences that are measurable,
such as land use changes, also may be
included. Transport of the raw materials to
the point of refining and processing is also
included within the boundaries of the raw
materials acquisition subsystems.
Figure 9 illustrates the manufacture and
use of bar soap on a life-cycle basis. The
shaded areas show the subsystems that are
included hi the raw materials acquisition
stage of this system. In this example, the
raw materials acquisition stage includes the
-------
Issues Applicable To Specific Life-Cycle Stages
Soil Preparation,
Seeds, Fertilizers,
Pesticides
Harvesting and
Processing of Silage,
Grains, and Hay
Seedlings
and Seeds
Cattle Raising
Planted Forest
Harvesting
Meat Packing
and Rendering
Chlorine
Production
"Natural" Forest
Harvesting
Tallow
Production
Sodium
Hydroxide
Production
Bar Soap
Production
Paper
Production
Soap Packaging
Cardboard
Production
Cardboard
Recycler
Consumer
Postconsumer
Waste
Management
Note: Energy
acquisition and
electricity generation ,
are not shown on this
diagram, although they
are inputs to many of
these processes.
Figure 9
The first step in a product's life cycle is the acquisition of raw materials and energy
68
-------
Issues Applicable To Specific Life-Cycle Stages
planting, harvesting, and processing of
silage, grains, and hay; salt mining; natural
forest harvesting; and the planting and har-
vesting of planted forests. The energy
expended and water used in acquiring the
raw materials is included in this stage. The
environmental emissions and wastes pro-
duced in the activities related to raw mate-
rials acquisition also are included in this
stage.
A typical life-cycle inventory of a product
evaluates the primary product and may
include associated primary, secondary, and
tertiary packaging. For example, the analy-
sis of bar soap includes the manufacture of
tallow and sodium hydroxide for soap mak-
ing, paper for packaging the soap (primary
packaging), and corrugated boxes for ship-
ping the soap (secondary packaging). The
manufacture of each material included in
the analysis begins with the acquisition of
the raw materials necessary to produce
each component. Any fuel, chemical, addi-
tive, or material that is significant enough
to be included in the analysis is traced back
to its raw material acquisition step. For
example, large quantities of fertilizers and
pesticides are used to produce silage, grain,
and hay, which are the raw materials for
tallow production. Thus, the energy and
emissions for the production of the fertiliz-
ers and pesticides need to be calculated in
the inventory, including the mining and
crude oil production steps to acquire the
raw materials used to produce the fertiliz-
ers and pesticides.
In a traditional inventory, a material con-
tributing less than 1% by weight of the total
system typically contributes less than 1%
to the total emissions of a particular inven-
tory item and can be omitted from the
inventory. The "less than 1%" effect is gen-
erally considered insignificant. This
approach is based on many years of con-
ducting such studies and not on statistical
or technical grounds. However, it does not
make any assumptions regarding the envi-
ronmental significance of the emission.
Thus, one problem with a blanket applica-
tion of this approach is that toxic materials
could inadvertently be eliminated from the
inventory analysis even though they pre-
sent a serious potential environmental
impact. Most practitioners examine the sys-
tem and make exceptions to a strict inter-
pretation of the "less than 1%" rule where
toxic materials are concerned.
In practice, the significance of a specific
item to the inventory may be determined by
performing a sensitivity analysis of that
item. If omission of the item does not affect
the ability to use the data to support the
purpose of the inventory, the item may be
excluded from the study. For instance, the
printing ink on a folding carton may be less
than 1% by weight of the system and prob-
ably can be eliminated from the system
with little effect on overall data accuracy.
However, the printing ink on a thin plastic
film may be 10% or more by weight of a
different system and probably should be
factored in to maintain the overall accuracy
desired. If concerns about toxic constitu-
ents in the ink are an issue, two options are
available. One is to include the ink hi the
folding carton inventory. When this is
done, however, depending on the number
of product units for which the final data are
reported, the emissions associated with the
ink can get lost in the overall data. The
other option is to conduct a separate inven-
tory on the ink.
Another example would be the cleaning
agents or routine maintenance chemicals
69
-------
Issues Applicable To Specific Life-Cycle Stages
used in a process, such as volatile organic
solvents used to clean a printing press. If
the system heing examined is large hi
scope, these chemicals will probably be
shown by sensitivity analysis to be insignif-
icant. If the study is focused on a specific
process, such as the printing operation
itself, the cleaning agents and maintenance
chemicals probably will be significant.
Therefore, if a material or chemical is a sig-
nificant part of the system, the inventory
should include its production, including
the raw material acquisition steps for pro-
ducing it.
Specific Assumptions
and Conventions
Raw Materials Acquisition
Analysis of a material or product begins
with specific data for the acquisition of raw
materials. For example, analysis of the
manufacture of the paper used to package a
bar of soap will begin with the mining of
salt and harvesting of trees, because these
are the raw materials needed to produce
paper. Specific assumptions for the acquisi-
tion of raw materials are listed below:
General 1. The acquisition of a raw material or
energy source requires a disturbance to the
environment. Ecosystems are impacted in the
harvesting of trees, in mining for minerals, in
using land to produce agricultural crops, and
in drilling for petroleum or natural gas.
Resource requirements that can be quantified
should be included in the life-cycle inven-
tory. Statistics are available that quantify
such effects as pesticide runoff from agricul-
tural activities, brine production from oil
wells, waterborne wastes from animal feed-
lots, etc. However, other consequences of raw
materials acquisition are not routinely mea-
sured and would be difficult to quantify, such
as soil erosion, damage to watersheds,
thermal pollution, and habitat destruction. If
measures of these outputs can be provided by
the analyst, they can be included in the
inventory. There is presently no consensus
among practitioners regarding the inclusion
of qualitative factors in the text of an inven-
tory report. These would be covered in an
impact analysis.
General 2. Traditional fuels such as petrol-
eum, natural gas, or coal are sometimes used
as raw materials. For example, crude oil and
natural gas are raw materials for plastic
products. When these traditional fuels are
used as raw materials, they are assigned an
inherent energy value (also called energy of
material resource or latent energy) equal to
the heat of combustion of the raw material
because the fuels have been removed from
the total fuel supply.
Because wood is not commonly used as a
fuel in the USA and the wood that is used
as a fuel in industry is typically a waste
product (e.g., bark from a pulp or paper
mill), several options for handling the
energy value of wood deserve considera-
tion. First, it is unlikely that wood waste
would be used as fuel for another industry
within the USA because of its location, so it
can be assumed that it would become a
solid waste if it were not burned at the
paper mill. Thus, one option is to avoid
giving an inherent energy value to wood
when it is used as a raw material in the.
USA. If the study were performed in a
country where wood is a primary fuel, then
an energy of material resource value equal
to the wood's heat of combustion would be
assigned to any wood used as a raw
material.
70
-------
Issues Applicable To Specifk Life-Cycle Stages
A second option for handling the wood's
energy value would be to count wood in. a
nonfuel category, especially in the USA.
This option would require the addition of a
separate category to record nonfuel inher-
ent energy, as discussed in item 4 below.
A third option that has been proposed for
considering the energy value of wood waste
is to assume that wood burned at a paper
mill to generate energy reduces the amount
of traditional fuels that are burned. For this
option, inherent energy credit would be
given to this process for reducing the
amount of traditional energy materials
required by burning the wood. Because this
option modifies the intent of a life-cycle
inventory to quantify actual consequences,
it is not recommended.
General 3. In a life-cycle inventory, energy
requirements of a system are not reduced or
credited for the use of "renewable" resources
instead of "nonrenewable" resources. For
example, a process may use wood energy
instead of coal energy. The issues or distinc-
tions between the levels of renewability are
difficult to define and even more difficult to
value in terms of energy. Furthermore, an
inventory analysis is designed to quantify
energy and raw materials requirements. The
impact analysis is the appropriate place to
characterize effects such as renewability.
General 4. To ensure that all sources of energy
are properly accounted for in the inventory
analysis, it may be necessary to create a
separate category of nonfuel inherent energy
that will permit closing the energy loop in a
thermodynamic sense without artificially
inflating the inherent energy values for fuels.
This separate category for nonfuel inherent
energy would not be reported in the summary
tables of energy, but would still permit
reviewers to account for all energy inputs and
outputs. Examples of inherent energy materi-
als that typically are not considered to be
fuels include biomass waste materials such as
sugar cane bagasse, wood residues from
logging operations, and textile fibers.
Mining Operations 1. Overburden, the material
overlying the ore or material being mined, is
not considered to be an environmental
emission (i.e., a solid waste) hi mining opera-
tions. This is because the overburden is
returned to the same land rather than being
landfilled after the mining operations are
completed. However, land use changes occur
with this operation and may be quantified in
the inventory. Habitat effects associated with
mining may occur, but these are treated in the
impact analysis, not in the inventory analy-
sis. However, other quantifiable emissions
such as particulates released into the ah* are
included in the inventory.
Petroleum and Natural Gas Acquisition 1. Brine
water is a coproduct of the production of
crude oil. A portion of the brine water is
reinjected into separate wells designed to
receive it. The waterborne wastes contained
in the part of the oil or gas well brine water
that is reinjected are not included in the
inventory because the wastes are essentially
being returned to the same location. In an
inventory, it is assumed that groundwater
flow and quality are not affected. This
assumption may be examined in an impact
analysis. Waterborne wastes discharged into
the ocean or another body of water are
included.
Petroleum and Natural Gas Acquisition 2.
Natural gas is often produced in combination
with crude oiLTherefore, environmental
emissions from drilling operations are allo-
cated between crude oil and natural gas
71
-------
Issues Applicable To Specific Ufo-Cycte Stages
production. This allocation is made based on
historical production data. For example, a
drilling operation may produce 50% crude
oil and 50% natural gas by weight. Thus, the
data for the process will be split between
them, using one of the coproduct allocation
techniques discussed in Chapters Three and
Four.
Lumbering Operations 1. Wood residue left in
forests after tree harvesting is not considered
to be a solid waste because it is not landfilled.
In most logging operations, the residue is left
to decompose where the lumber was har-
vested. However, in some operations the
harvested land may be burned off, which
generates atmospheric emissions that must be
quantified. Any assumptions made as to
waste practices in the harvest of wood must
be thoroughly documented. Other outputs
from lumbering operations that are quantifi-
able, such as land use changes, may be
included.
Agricultural Raw Materials and Animal
Products 1. Harvesting of agricultural products
often involves significant manual labor. This
is especially true in developing countries.
Energy requirements and environmental
emissions related to sustaining human life
(e.g., producing food, clothing, or shelter)
typically are not included hi the life-cycle
inventory.
Agricultural Raw Materials and Animal
Products 2. In a life-cycle inventory analyzing
title use of an animal product, the feed to
produce the animal product usually is con-
sidered to be the main raw material for the
system. For example, the analysis of tallow
for bar soap may have corn as a raw material
for the system. Corn is fed to the cattle that
produce the tallow. Acquiring the raw mate-
rial (corn) in this system requires energy for
planting, harvesting, and transporting the
corn. Emissions from pesticides and herbi-
cides are associated with producing corn as a
raw material. Therefore, the raw material
acquisition steps for those chemicals should
be included in the system.
Calculating Resource Requirement and Emis-
sions for Raw Material Acquisition. Using the
bar soap example (Figure 9), the raw material
for sodium hydroxide is salt. The energy
requirements and environmental emissions
associated with the mining of salt and its
transport to a caustic production facility are
the energy and emissions of raw materials
acquisition. Assume that the hypothetical
data below represent the energy and emis-
sions for muiing 1,000 pounds of salt.
Process Energy
Electricity
Coal
Residual oil
50 kilowatt-hours
50 pounds
50 gallons
Transportation Energy
Ocean transport
Diesel
Atmospheric Emissions
Particulates
50 ton-miles
5 gallons
5 pounds
Assume further that 5 pounds of salt must
be mined for every pound of sodium
hydroxide manufactured and used. To
determine the energy and emissions associ-
ated with raw materials acquisition for the
manufacture of 1,000 pounds of sodium
hydroxide, the above data would be multi-
plied by five in the computer spreadsheet.
Energy Acquisition
The energy requirements and environmen-
tal emissions attributed to the acquisition,
transportation, and processing of fuels to a
usable form are labeled as precombustion
energy or emissions. Whenever a specific
72
-------
Issues Applicable To Specific Life-Cycle Stages
fuel is used in any of the processing or
transportation steps, the appropriate quanti-
ties of precombustion energy and emissions
are included in the total energy and emis-
sions attributed to the use of that fuel.
Therefore, when gasoline is used as a fuel,
the energy and emissions must account not
only for those from the combustion of the
gasoline, but also for those attributable to
the raw material extraction, refining, and
processing required to produce the gasoline
before it is burned. The assumptions listed
for the acquisition of raw materials also gen-
erally apply to the acquisition of energy raw
materials. The five basic energy sources
included in a life-cycle analysis are coal,
petroleum, natural gas, nuclear power, and
hydropower. In certain cases, wood may
also be included. Minor energy sources such
as biomass, solar, and geothermal energy are
generally ignored unless significant to a spe-
cific process.
Specific Steps for Calculating Energy Acquisition.
Residual oil is one fuel used in several of the
individual processes in the bar soap manufac-
turing system. Assume that every use of
residual oil has been accounted for in this
system and that the total used is 50 gallons.
Assume that the hypothetical data listed
below represent the total energy and emis-
sions for producing 1 gallon of residual oil,
including extracting crude oil, transporting it,
and processing it into residual oil.
Process Energy
Electricity
Coal
Natural gas
Atmospheric Emissions
Particulates 0.25 pound
Hydrocarbons 0.25 pound
10 kilowatt-hours
10 pounds
10 cubic feet
Solid Wastes
0.25 pound
Transportation Energy
Pipeline 50 ton-miles
Natural gas 5 cubic feet
The hypothetical data above must be multi-
plied by 50 to calculate the precombustion
energy and emissions associated with the
50 gallons of residual oil used hi the bar
soap system. These are the total energy
requirements and emissions associated with
the acquisition of 50 gallons of residual oil
to be used as process energy in the system.
A computer spreadsheet is used to make
similar calculations for every other fuel
used in the system.
Electrical Energy Acquisition
The production of electrical energy is more
complex than simply refining a fuel. There-
fore, this section addresses electrical energy
acquisition separately. Utility power plants
generate electricity from five basic energy
sources: coal, petroleum, natural gas,
nuclear power, and hydropower. A small
percentage of electricity is also generated by
unconventional sources such as biomass,
solid waste, solar, and geothermal energy.
The acquisition of electrical energy includes
extracting the basic energy sources from the
earth, processing these energy sources into
usable fuels, and converting the fuels into
electrical energy. As noted previously, the
manufacturing processes for the production
of any given product are sufficiently scat-
tered throughout the country that the
national average fuel mix for the electric
utilities is representative when used for all
of the manufacturing steps in an analysis.
An exception to this assumption is the elec-
troprocess industries, with aluminum
-------
Issues Applicable To Specific Life-Cycle Stages
smelting as the primary example. These
industries locate in areas of inexpensive
electrical power because they require vast
amounts of electricity. The fuel mix for
regional utility grids may be used in these
specific situations. Historical data provide
the basis for determining the efficiency of
converting the different fuels to electricity.
Historical data are also available for deter-
mining the transmission line losses (i.e., dif-
ference between amounts of electricity
generated and sold) for delivery of electrical
energy to the consumer.
Specific Steps for Calculating Electrical Energy
Acquisition. Three steps are required to deter-
mine the total energy and emissions associ-
ated with the use of electrical power:
extraction of fuel sources, processing of fuels,
and converting the fuels to electrical energy.
Before any of these steps may begin, the fuel
mix for the electricity being used should be
calculated. As expressed above, the national
average fuel mix for the electric utilities is
usually adequate for most industries. Where
or when possible, electricity for a specific
industry should be traced back to its specific
fuel source to develop a more accurate profile
for each process. For example, assume that the
utilities are using 50% coal and 50% residual
oil. Thus, the energy and emissions associ-
ated with coal mining and crude oil extrac-
tion, as well as transport of these materials,
must be quantified. Any processing necessary
to make these fuels usable must be incorpo-
rated into the analysis. After processing, the
fuels are burned to produce steam for generat-
ing electrical energy. Both the emissions from
the combustion of the fuels and the efficiency
of the boiler system must be included in the
analysis. Finally, the transmission line losses
(the difference between the electricity gener-
ated and the electricity delivered) must be
accounted for. All of these processes fell into
the raw materials acquisition phase of the life-
cycle inventory.
MANUFACTURING STAG!:
The second stage of a product's life cycle is
manufacturing. Manufacturing results in the
transformation of raw materials into prod-
ucts delivered to end users. The manufac-
turing stage is further divided into three
steps: materials manufacture, product fabri-
cation, and filling/packaging/distribution.
Each of these is discussed in the following
sections.
Major ConceptsMaterials
Manufacture
Materials manufacture converts raw materi-
als into the intermediate products from
which the finished product will be
fabricated.
Material scrap from a subsystem can be
reused internally, sold as industrial scrap, or
disposed of as solid waste. The inventory
account for each option is handled
differently.
No credits or debits are applied to the sub-
system for internally recycled material
because no material crosses a subsystem
boundary.
Industrial scrap as a coproduct carries with it
the energy and wastes to produce it. This
ensures consistency with operations that use
the scrap in house.
74
-------
issues Applicable To Specific Life-Cycle Stages
Materials Manufacture Step
The first step in manufacturing is the manu-
facture of materials. This step includes all
manufacturing processes required to process
raw materials into the intermediate materi-
als from which the finished product will be
fabricated. In an inventory examining bar
soap, this step would include all operations
required to produce tallow and sodium
hydroxide from which bar soap is made.
Similarly for paper packaging, this step
would include all operations required to
transform wood into paper.
Subsystem Boundaries
The boundaries for material manufacture sub-
systems encompass the actual processes) of
manufacturing an intermediate material,
either from raw materials or from other inter-
mediate materials. This step includes any
energy, material, and water input and the
environmental releases. Transport of the inter-
mediate material produced to the site of the
next manufacturing process or to the point of
product fabrication is also included in the
boundaries of a material manufacturing sub-
system. Depending on the locations of facili-
ties where material manufacture occurs,
nontraditional inventory outputs such as land
use and odors may be relevant. Any number
of material manufacture subsystems may be
required to convert a raw material into the
intermediate material required to fabricate a
product. For example, to convert crude oil
into a polyethylene milk bottle, three subsys-
tems are required: crude oil refining, ethylene
production, and polyethylene production.
Each subsystem has its own boundaries as
described above to facilitate data gathering
and to eliminate double-counting or omis-
sions. The boundaries of the material manu-
facture step are illustrated for bar soap by the
shading in Figure 10. Each of the material
manufacturing operations can be viewed as a
subsystem within the product system. Data for
each manufacturing operation or subsystem
are gathered separately. Thus, the material
manufacturing step is not necessarily linear,
but may be a complex arrangement of pro-
cesses and multiple raw materials.
For each subsystem, the materials and energy
inputs for processing are analyzed. The air
and water emissions and the solid wastes
resulting from each subsystem are also
reported. la other words, a material and
energy balance must be performed on each
operation within the system. Energy and envi-
ronmental wastes resulting from the transpor-
tation required from one process operation to
another, or to the point of product fabrication,
are also included.
Specific Assumptions and Conventions
The following assumptions and conventions
generally apply to this step:
Coproduct Allocation. Manufacturing pro-
cesses, particularly chemical processes, often
produce more than one product. In most
cases, only one of those products is used in
the system being evaluated. Thus, some
allocation of material and energy inputs and
waste and emission outputs is necessary for
marketed output materials. Chapter Four
presents several methods for allocating
requirements and emissions to various
coproducts.
Industrial Scrap. Many manufacturing pro-
cesses generate scrap material. This material is
often reused within the same manufacturing
process, reducing the new material input into
the system. In such a case, the system is
75
-------
Issues Applicable To Specific Life-Cycle Stages
Soil Preparation,
Seeds. Fertilizers,
Pesticides
Haivesting and
Processing of Silage,
Grains, and Hay
Seedlings
and Seeds
Cattle Raising
Planted Forest
Harvesting
Meat Packing
and Rendering
Natural" Forest
Harvesting
Chlorine
Production
Tallow
Production
Sodium
Hydroxide
Production
Bar Soap
Production
Paper
Production
Soap Packaging
Cardboard
Production
Cardboard
Recycler
Consumer
Postconsumer
Waste
Management
Note: Energy
acquisition and
electricity generation
are not shown on this
diagram, although they
are inputs to many of
these processes,
Figure 10
Materials manufacture converts raw materials into a form usable in a finished product
76
-------
Issues Applicable To Specif k LJfe-Cvcle Staoes
considered to have a continual inner loop of
material exiting the system and re-entering the
system as an input material. No credits or
dehits are applied to the processes because the
material does not cross subsystem boundaries.
For example, when the trim from foam poly-
styrene extrusion and thermoforming is
collected, reground, and used as an input into
the subsystem, the amount of raw or interme-
diate material input from outside the process
is reduced. Energy and emissions are calcu-
lated based only on the raw or intermediate
material input from outside the subsystem.
Material scrap from one manufacturing pro-
cess also is frequently marketed as a mate-
rial input in another process. This marketed
scrap is commonly referred to as industrial
scrap. For example, trim scrap from flexible
polyurethane foam often is sold for use as
carpet backing. Also, vegetable hulls and
peels from food processing operations often
are used as raw materials for animal feed.
Industrial scrap is viewed as a coproduct
from the manufacturing process and carries
with it the energy requirements and envi-
ronmental wastes required to produce that
material. Therefore, when the coproduct
allocation is applied to the process generat-
ing the scrap, the coproduct has the same
energy and emissions per pound of output
as the operation that uses the scrap in
house. Some scrap generated from manufac-
turing processes is discarded to the munici-
pal solid waste stream along with other
wastes from the manufacturing facility. In
this case, the scrap is reported as solid
waste from the manufacturing process and
no allocation is applied.
Product Fabrication Step
The second step in manufacturing is the fab-
rication of the product. For the manufacture
of bar soap, this step is represented by the
subsystem that includes the production of
fatty acids from tallow, vacuum distillation,
manufacture of neat soap and, finally, the
cutting and drying of the bar soap.
Subsystem Boundaries
The product fabrication step usually has a
more narrow focus than the materials manu-
facture step and may sometimes include
only one manufacturing operation. If more
than one fabrication operation is necessary,
data may be gathered individually or
collectively depending on availability. A
product fabrication subsystem will include
any energy, material, and water input and
the environmental releases. Transport of the
product to the point of filling, packaging,
and/or distribution is also included. Trans-
portation may occur within the facility or
between facilities. The boundaries of the
product fabrication step of the life-cycle
inventory for bar soap are highlighted in
Figure 11. A number of individual processes
Major ConceptsProduct
Fabrication
Product fabrication converts intermediate
materials into products ready for their
intended use by consumers.
Facilities for which data are reported on a
plant-wide basis will require allocation of
the inputs and outputs to the product of
interest.
-------
Issues Applicable To Specific Life-Cycle Stages
Soil Preparation,
Seeds, Fertilizers,
Pesticides
Harvesting and
Processing of Silage,
Grains, and Hay
Meat Packing
and Rendering
Sodium
Hydroxide
Production
Postconsumer
Waste
Management
Tallow Hot Water -#- Gas
I
Oils
Fragrances
Colors
Neat _
Soap
-Glycerine Fatty Acids
Sodium
Hydroxide
Note: Energy
acquisition and
electricity generation
are not shown on this
diagram, although they
are inputs to many of
these processes.
Bar Soap Production
Figure 11 *..._ j- .*
Product fabrication converts intermediate materials into a finished product
78
-------
Issues Applicable To Specific Life-cycle Stages
are involved, as illustrated by the expanded
view in Figure 11. Depending on data avail-
ability, the operations can be viewed sepa-
rately or as comprising a single subsystem.
For example, a soap manufacturing plant
may account for energy consumption and
emissions with plant-wide data rather than
process-specific data. In this situation, it
would be best to analyze the overall plant
operations and then allocate the inputs and
outputs. Another soap manufacturing plant
may collect and report energy consumption
and emissions on each separate process. In
this case, each process should be analyzed
for its contribution to the entire system or
subsystem under examination.
Specific Assumptions and Conventions
The following specific assumptions and
^conventions generally apply to the product
fabrication step:
Coproduct Allocation. Manufacturing processes
often produce more than one product. In most
cases, only one of those products is evaluated
at a time. Thus, some allocation of material
and energy inputs and waste and emission
outputs is necessary for all marketed output
materials. Chapter Four presents several
methods for allocating requirements and
emissions to various coproducts.
Industrial Scrap. As in the materials manufac-
ture step, scrap material is generated by many
product fabrication processes, and the same
principles of resource and emissions alloca-
tions apply. This marketed scrap is commonly
referred to as industrial scrap. For example,
trim scrap from the production of computers
is sold for use in building products.
Filling/Packaging/Distribution Step
The third and final step in the manufactur-
ing stage is filling, packaging, and distribu-
tion. This step includes all manufacturing
processes and transportation required
between product fabrication and delivery of
the product to the consumer. Thus, in an
inventory that examines bar soap, this step
would include all operations required to
package the soap in paper wrappers, place
the packaged soap into corrugated boxes,
and transport the boxes to the retailer and
then to the consumer.
Subsystem Boundaries
The subsystem boundaries of the filling,
packaging, and distribution step begin with
the filling and packaging operations once
Major ConceptsFilling/
Packaging/Distribution
Filling and packaging products ensure
that the products remain intact until
they are ready for use. whereas distribu-
tion transfers the products from the
manufacturer to the consumer.
In addition to primary packaging, some
products require secondary and tertiary
packaging, all of which should be
accounted for in a life-cyde inventory.
Any special circumstances in transporta-
tion, such as refrigeration used to keep
a product fresh, should be considered
in the inventory.
79
-------
Issues Applicable To Specific Life-Cycle Stages
the materials have reached the facilities
where these operations occur. This step also
includes distribution to the consumer. Some
life-cycle inventories evaluate a package
rather than the substance that is put into the
package. Therefore, the analysis may not
include the contents of the packages. The
boundaries of this step of the life-cycle
inventory for bar soap are shaded in Fig-
ure 12. In this illustration the consumer
product, soap, is the focus of the inventory.
For each activity or subsystem in this step,
the materials and energy inputs are
required. Atmospheric and water emissions
and the solid wastes resulting from each
operation also should be reported. Energy
and environmental wastes resulting from
transport of the consumer product from the
manufacturer and distributor to the retail
outlet also are included in this step.
Specific Assumptions and Conventions
The assumptions and conventions dis-
cussed below are commonly used for this
step of the life-cycle inventory.
filling. Often the purpose of a life-cycle inven-
tory is to quantify relative differences between
the products, processes, or materials being
compared. Thus, in filling, certain identical
factors between systems often can be ignored
because they affect each system in the same
way. Two examples of identical factors could
be the actual contents of the packages and the
filling operation. An example of identical
effects from the actual contents of packages
would be found in an analysis comparing
packages for delivery of 1,000 vitamin tablets.
The energy requirements and emissions of the
vitamin tablets themselves will always be the
same, whereas the energy requirements and
emissions of the various package systems
being compared may not be the same. Thus
the analyst could ignore the tablets and
concentrate on the packaging. The energy and
emissions associated with filling a product
bottle or package can be ignored when the
products being compared use the same filling
procedures and equipment. For example, a
comparison between aluminum soft drink
cans and steel soft drink cans will probably
have identical filling requirements. However,
a comparison between 2-liter plastic soft drink
bottles and 12-ounce aluminum cans will
have different filling requirements that should
be investigated separately in the analysis.
Packaging. Once the bottle or primary package
is filled, secondary packaging is applied to
ensure the integrity of the product during
shipment. The amounts and types of second-
ary packaging vary with the type of product
being shipped. Two-liter plastic soft drink
bottles and aluminum cans require different
secondary packaging. Bars of soap and liquid
soap also require different secondary packag-
ing. When different, the specific amounts and
types of secondary packaging should be
quantified.
Distribution. Distribution involves transporting
the packaged product to warehouses, retail
establishments, and consumers. An average
distance for product transport must be devel-
oped. The normal mode or modes of transpor-
tation, e.g., truck, rail, or barge, must be
established. Special care should be taken to
include the analysis of any unusual situations.
For example, a study comparing the delivery
of frozen concentrated juice to the delivery of
ready-made juice would include the energy
and emissions associated with the refrigera-
tion of the frozen product during delivery.
-------
issues Applicable To Specific Life-Cycle Stages
Soil Preparation,
Seeds, Fertilizers,
Pesticides
.
r
Harvesting and
Processing of Silage,
Grains, and Hay
\
r
Cattle Raising
1
Meat Packing
and Rendering
"
<
Tallow
Production
|
Bar Soap I
Production
V _ _J
fioan Pa
ckaging
«:IAV
Cons
umer
Salt Mining
i
Sod
Hydr
Prodi
r
ium
Dxide
jction
Seedlings
and Seeds
*
Planted Forest
Harvesting
*
Chlorine "Natural" Forest
Production ^ Harvesting
\ *
Paper
Production
^
r
Cardboard
Production
\
L
Cardboard
Recycler
Postconsumer
Waste
Management
. Note: Energy
acquisition and
electricity generation
are not shown on this
diagram, although they
are inputs to many of
these processes.
Figure 12
Filling, packaging, and distribution is the final step in manufacturing before a product reaches the
consumer
-------
Issues Applicable To Specific Life-Cycle Stages
USE/REUSE/MAINTENANCE STAGE
The third stage of a product's life cycle is
the use/reuse/maintenance (U/R/M) stage.
This stage consists of a discrete set of activi-
ties that begins after distribution of finished
products or materials to the consumer and
ends when these products or materials are
either recycled or discarded into a waste
management system.
Subsystem Boundaries
The U/R/M stage begins when a finished
product arrives in the possession of a con-
sumer, including individual consumers,
commercial businesses, and institutions.
This stage of a life-cycle inventory for bar
soap is illustrated by the shading in Fig-
ure 13. Transport of the product to the con-
sumer is considered part of the filling/
packaging/distribution step, whether it
arrives by mail, by purchase from a service
Major ConceptsUse/Reuse/
Maintenance
This stage includes all of the activities
undertaken by the user of the product or
service as well as any maintenance that
may be performed by the user or obtained
elsewhere.
Household operations, such as refrigera-
tion, are rarely associated with a single
product. Either the allocation of the capital
and operating energy and environmental
releases to a particular item are too small
to affect the results or they can be propor-
tionately included.
outlet; or by transport from a retail store by
the consumer, and therefore is not part of
the U/R/M stage. The packaging that accom-
panies the product to the consumer is part
of the U/R/M stage, but the shipping boxes
used to transport a load of products to the
retail store are part of the previous filling/
packaging/distribution step. The U/R/M
stage ends when the consumer is done using
the material or product and delivers it to a
collection system for recycling or waste
management.
Specif ic Assumptions
and Conventions
Household operations performed during the
U/R/M stage are rarely allocated to an indi-
vidual product. For example, a household
rarely operates a refrigerator to cool only
one product; household refrigerators usually
contain a variety of products. Another
example is the variety of clothes that are
usually washed or dried in a single load-
The historical option for handling the
energy requirements and emissions from
refrigeration or other similar multiple-
product actions has been to omit them from
a life-cycle inventory. Sensitivity analysis
has shown that once the energy
requirements and environmental emissions
for these types of multiple-product actions
are spread over all products, the values per
product are minuscule and do not signifi-
cantly change the results. A second option
for handling multiple-product actions is to
proportionally allocate the energy require-
ments and environmental emissions based
on the weight percent, heat capacity, or
other justifiable property of each individual
product. For example, on an annual basis,
"x" loads of laundry are washed, of which
82
-------
Issues Applicable To Specific Life-Cycle Staaes
Soil Preparation,
Seeds, Fertilizers,
Pesticides
Harvesting and
Processing of Silage,
Grains, and Hay
Cattle Raising
Seedlings
and Seeds
Meat Packing
and Rendering
*
Tallow
Production
i
Bar Soap
1 Production
^^^jjIB^ j _ ..IBJPI^IPIBI., ._-._ -_,- ."fa
Soap Packaging
\
Retailer
L
1
|N!
Salt Mining
i
Planted Forest
Harvesting
*' .
Chlorine "Natural" Forest
Production Harvesting
\ 1
Sodium
Hydroxide
Production
Paper
Production
I
Cardboard
Production
1
Cardboard
Recycler
-
Postconsumer
Waste
Management
Note: Energy
acquisition and
electricity generation
are not shown on this
diagram, although they
are inputs to many of
these processes.
Figure 13
Consumer use/reuse/maintenance is the third stage in a product's life cycle
83
-------
Issues Applicable To Specif k Life-Cycle Stages
"y" percent is cloth diapers. The washer-
operating emissions allocated to diapers are
"z". Over the lifetime of the washer, manu-
facturing and other emissions are an insigni-
ficant percentage of the diaper system life
cycle.
The decision to use proportional allocation
for consumer products needs to be carefully
considered because it can have a significant
effect on results. The decision generally
relates to the basic purpose of the inventory.
In many cases, the purpose of an inventory
is to determine the incremental effects of
substituting one product for another. For
example, if two different bar soap formula-
tions are being compared by a manufacturer,
the trip by the consumer in a car to a store
to purchase the soap as well as other house-
hold goods at the same time would not be
included because the effects are identical
for either bar soap formulation. If one form-
ulation were to replace the other in the mar-
ketplace, no incremental changes in
resource depletion or environmental emis-
sions would result.
If, however, the purpose of the inventory
were to learn in an absolute sense where
resources are used and emissions are gener-
ated, the car trip would be included because
car trips contribute to global resource deple-
tion and environmental emissions. It is
necessary to determine the purpose of the
trip because, if the trip is only to purchase
one product, the entire trip must be charged
to that product. If the trip has multiple uses,
proportional allocation would be used.
If the purpose of the inventory were to dis-
cover incremental changes by comparing
two or more systems, identical consumer
actions would be excluded. In absolute
studies that include all impacts, consumer
activities must be inventoried unless found
to be insignificant.
Reuse of a product or package is also
included in this stage. A product or package
may be reused once or many times. Energy
or emissions from activities, such as clean-
ing, needed to ready the item for reuse
should be included in the inventory. If they
occur at the point of use, they should be
kept in the data for this stage. Sometimes
they occur twice, as when a recycled glass
container is cleaned by both the user before
recycle and by the manufacturer. Some
examples of reuse would be refillable glass
soft drink bottles and reusable shipping
crates. Cleaning or refurbishment energy
and emissions should be included with the
manufacturing stage data if these operations
are not performed by the user.
RECYCLE/WASTE MANAGEMENT
STAGE
The fourth and final life-cycle stage is
recycle/waste management. Normally, after
a product and its packaging have been used
by a consumer and the product has fulfilled
its intended purpose, it is either recycled,
composted, or discarded as waste. Recycling
begins when a discarded product or package
is delivered to a collection system for
recycling. Composting is the controlled, bio-
logical decomposition of organic materials
into a relatively stable, humus-like material.
The waste from the U/R/M life-cycle stage is
commonly referred to as postconsumer
waste. Waste management refers to the fate
of both industrial and postconsumer solid
waste that is discarded and picked up, and
includes incineration and landfilling. This
stage also includes postconsumer wastewa-
ter treatment.
84
-------
Issues Applicable To Specifk Life-Cycle Stages
Subsystem Boundaries
The subsystem boundaries for recycle/waste
management encompass recycling, compost-
ing, and the two waste management options
of incineration or landfilling for solid waste
and wastewater treatment. A variety of steps
or subsystems may be included in this stage.
A subsystem in this life-cycle stage will
Include any energy, material, and water
inputs and the environmental releases.
Transportation from one subsystem to a sub-
sequent one would be included.
The recycle/waste management stage of a
life-cycle inventory for bar soap is illus-
trated by shading in Figure 14. This stage
begins with the final disposition of the
product and its related packaging. The con-
sumption or use of the product may or may
not alter the product itself. The use of soap,
for example, reduces the size of the product
and thus alters the primary product. The
use of disposable diapers also alters the
Major ConceptsRecycle/Waste
Management
Recycle/Waste management is the last
stage in a product's life cycle.
In open-loop recycling, products are
recycled into new products that are even-
tually disposed of.
In closed-loop recyding, products are
recycled again and again into the same
product.
Formulas can be used to determine the
credits that should be assigned to recycled
products analyzed in a life-cyde inventory.
product. Discarded diapers contain added
weight and material from human wastes and
moisture. Some products are not altered by
use. For example, the configuration of a
glass beverage container is not changed
when the container has fulfilled its purpose
of holding a liquid.
The collection and transportation of dis-
carded materials for the various recycling/
waste management options should be
included in the life-cycle inventory of a
product, although typically these steps are
minor components. The alternatives most
often used in the disposition of discarded
products are, in order of EPA preference,
reuse, recycling, composting, and incinera-
tion/landfilling. The life-cycle inventory
should include data for the processing of
materials in the recycling and composting
processes, i.e., both the energy requirements
for these processes and the wastes emitted
from them. Incineration converts organic
products to carbon dioxide, water, and resi-
duals, but depending on the nature of the
product and incinerator control devices,
incineration may release atmospheric emis-
sions and/or leave solid and liquid wastes
in the forms of ash or scrubber blowdown.
The landfilling option ends with the burial
of products and requires the quantification
of the solid waste buried. Atmospheric and
waterborne emissions are also associated
with landfilling. Allocation options for
these are discussed below.
Specific Assumptions
and Conventions
This section discusses the basic assump-
tions and common conventions generally
used when performing a life-cycle inventory
for the recycle/waste management stage.
-------
Issues Applicable To Specific Life-Cycle Stages
Soil Preparation,
Seeds, Fertilizers,
Pesticides
*
Harvesting and
Processing of Silage,
Grains, and Hay
t
Cattle Raising
t
Meat Packing
and Rendering
1
Tallow
Production
t
Bar Soap
Production 1
j- uuaL
Soap Packaging
1 r
Retailer
^
Consumer
'v
S
Salt Mining
Seedlings
and Seeds
*
Planted Forest
Harvesting
\
Chlorine "Natural" Forest
Production Harvesting
\ »
Sodium
Hydroxide
Production
Paper
Production
*
Cardboard
Production
4
SSSSJ^SftS&Sfc^SS^JSSiSSSSS^SHSHWHi.
Cardboard
Recycler
Postconsumer
Waste
Management
X
,
Note: Energy
electricity generation
are not shown on this
diagram, although they
are inputs to many of
these processes.
Figure 14
Recycle/waste management is the last stage in a product's life cycle
86
-------
Issues Applicable To Specif k Life-Cycle Stages
Recycling
Recycling decreases the amount of solid
waste entering landfills and reduces the
production requirements of virgin or raw
materials. Therefore, life-cycle inventory
techniques adjust all resource requirements
and emissions for products that are recycled
or contain recycled content. Two recycling
systems could be considered in the life-
cycle inventory. These are the closed-loop
and open-loop systems.
Closed-Loop Recycling. Closed-loop recycling
occurs when a product is recycled into a
product that can be recycled over and over
again, theoretically endlessly (part a of Fig-
ure 15). Aluminum cans are a good example
of closed-loop recycling, because they are
recycled over and over again into aluminum
cans.
Consider an example where virgin materials
produced in operation 1 of Figure 15a are
augmented with a portion of recycled mate-
rials to yield a total mass flow, m, into the
production and use stages. If a fraction, F, is
recycled through recycling process 4, then
(l-F)m is collected from postconsumer use
Examples of Processes in
Postconsumer Recycling
Plastic milk jugs
Grinding
Washing
Remelting
Paper products
Repulping and
deinking/bleaching
for disposal. A fraction, f, of the material
collected for recycling may be rejected,
either due to technical reasons (i.e., contam-
ination), or for economic reasons (i.e., low
demand creating low recycled material
prices), and sent to disposal. Thus, the total
disposal mass amounts to (l-F(l-f))m. Sub-
tracting the recycling rejects from the recy-
cling input leaves the amount F(l-f)m,
which is recycled back to the input to aug-
ment the new raw material.
If no recycling takes place, then both the raw
materials input mass and the waste disposal
mass are equal to the value m. It can be seen
that as the recycled fraction, F, increases
toward unity, the need for virgin raw materi-
als and the solid waste generation rate
asymptotically approach the recycling reject
rate of fin at 100% recycling. Ultimately, raw
material resource use and solid waste dispo-
sal reach zero as the efficiency of the recy-
cling process approaches 100%. Thus, for
closed-loop recycling, allocation of inputs
and outputs is straightforward once the recy-
cling rate and the recycling process reject
rate are known.
Open-Loop Recycling. In the basic open-loop
recycling system, a product made from virgin
material is recycled into another product that
is not recycled, but disposed of, possibly after
a long-term diversion. An example of open-
loop recycling would be a plastic milk bottle
being recycled into plastic lumber or flower
pots, which currently are not recycled. In
open-loop recycling, energy and environmen-
tal emissions related to the production, recy-
cling, and final disposition of the plastic fesin
itself are divided between the two products by
one of several possible methods. In this way,
each of the two products made from the same
resin shares in the energy, landfill, and air and
87
-------
Virgin
(1) Materials
m
(2)
Production
and Use
m
Collection
(3)for Disposal
(5)
Disposal
Issues Applicable To Specific Life-Cycle Stages
(a) Closed-Loop Recycling
F(1-f)m
Recycling
Process
Fm
fFm
m1 (b) Open-Loop Recycling
T
(l-fjm.
Virgin Materials
d)for Product 1
i
mi
Production
and Use of
(2) Product 1
i
mi
- (1-Om,
Disposal of
,3) Product 1
fm,
Recycling
(7) Process
J
fm.|
f
Virgin Materials
(4)for Product 2
\
m2 frn
1 m2
Production
and Use of
(5) Product 2
i
1 m2
Disposal of
/g\ Products
Note: Product 1 is a virgin product.
Product 2 contains recycled material
but is not further recycled.
See text for further explanation.
Figure 15
Recycling flow diagrams
88
-------
Issues Applicable To Specific Life-Cycle Stages
water emissions savings achieved through
recycling. This basic open-loop system is
illustrated in part (b) of Figure 15.
Analysis of an open-loop system is only
slightly more complicated from a mathema-
tical perspective than that for a closed-loop
system. In the simplest open-loop system,
Product 1's production sequence produces
an output from operation (2) that is diverted
from waste disposal by some fraction f.
Thus, the mass flow to the recycling opera-
tion is foij. Again, although not shown, a
fraction can be diverted from collection to
disposal to account for poor quality or eco-
nomics just as in the closed-loop case. Prod-
uct 2 has its virgin raw materials input
reduced from m2 to m2-fm2. After use, Prod-
uct 2 is disposed of, with no further
recycling, at mass rate m2. As the recycling
rate of Product 1 approaches 100%, the
solid waste generation tends toward zero
and the raw materials input to Product 2
approaches n^m,.
Correct allocation of the inputs and outputs
requires analyzing the systems for both
Product 1 and Product 2 together. This is
the preferred option for practitioners to pur-
sue. However, in situations where this
option is not feasible, there are several pos-
sible methods for allocating all energy and
water requirements and environmental
releases attributable to the products (except
those for fabrication, packaging, and distri-
bution). Three possible methods for allocat-
ing impacts between the two products are
discussed below.
Because all of the following three allocation
methods for open-loop recycling are arbi-
trary, they are listed in decreasing order of
complexity to analyze: (1) the system
impacts can be allocated between the two
products based on the percentage of the two
products produced; (2) the disposal credits
can be allocated to the product being recy-
cled; and (3) the impacts added to the sys-
tem because of recycling can be equally
divided. No matter which allocation method
is used, it is important for the analyst to
state clearly the method being used, to
explain why it was chosen, and to use it
consistently. For each of these allocation
methods, the net inputs and outputs (N^)
can be described generically by the follow-
ing formula:
= D + E - B(a) - A(b) + G(a) + F
where:
A = all inputs and outputs associated with
production of the virgin material from
raw material to primary material pro-
duction for Product 2 (the product
that uses the recycled material).
B = all inputs and outputs associated with
disposal of Product 1 (the product that
is recycled), including transportation,
solid waste, incineration emissions,
and all other system impacts associ-
ated with waste disposal.
C = all inputs and outputs associated with
recycling of Product 1, including
transportation (to a drop-off site, a
materials recovery system, or a pro-
cessor); energy to reprocess back into
a primary material (to clean, shred, or
repulp); water; and wastes associated
with this reprocessing.
D = all inputs and outputs associated with
a no-recycling system for Product 1.
E = all inputs and outputs associated with
a virgin system for Product 2.
-------
Issues Applicable To Specific life-Cycle Stages
F = any converting inputs and outputs
incurred as a result of Product 2 using
recycled materials instead of virgin
materials.
a = the recycling rate of Product 1.
b = the recycled content level of
Product 2.
The inputs and outputs represented by D, E,
B, A, C, and F in this equation are specific
quantities associated with a given mass of
either Product 1 or Product 2. For example,
if 200 pounds of Product 1 and 100 pounds
of Product 2 are the result of the system,
then D, B, and C quantities are for 200
pounds of Product 1 andE, C, and F quanti-
ties are for 100 pounds of Product 2. Inputs
and outputs represented by N^are for the
entire system or 300 pounds of products.
First Allocation Method. In the first allocation
method, the system loadings can be allo-
cated between the two products based on
the number of units of each product needed
to produce a combined total of 100 units of
both products. This allocation method
requires that both product systems be ana-
lyzed to determine the percent of Product 1
that is recycled and the recycled content
level (from Product 1) in Product 2. The net
inputs and outputs can be calculated by the
following formulas for each product:
N^o (Product 1) = D - P, [B(a) + A(b) - C(a) - F]
N (Product 2) = E - P2 [B(a) + A(b) - C(a) - F]
'I/O
where:
p _
the percent of Product 1 out of 100
total units for both products.
the percent of Product 2 out of 100
total units for both products.
The limitations of this method are that
Product 2 may not be fully credited for recy-
cle content. For example, in a system where
Product 1 is being recycled at 50% into
Product 2, which has 100% postconsumer
recycle content from Product 1, Product 1
will be credited with higher "savings" from
recycling than will Product 2 because Prod-
uct 1 makes up a larger percentage of the
system. Although Product 2 is made from
100% recycled material, it is penalized
because Product 1 is being recycled at only
50%.
Second Allocation Method. In the second
allocation method, the disposal credits are
allocated to the product that gets recycled
(Product 1). Product 2 treats Product 1 as a
raw material that needs to be processed;
thus, the recycling inputs and outputs are
all allocated to Product 1. By treating Prod-
uct 1 as a raw material, Product 2 has
avoided virgin material production. The net
inputs and outputs for this allocation
method can be calculated by the following
formulas for each product:
Nyo (Product 1) = D - B(a)
Nvo (Product 2) = E - A(b) + C(a) + F
This method has the following limitations:
Product 2 is penalized with all of the solid
waste from the system, when actually the
production of Product 2 using recycled
material has decreased the system's total
solid waste by the quantity of material being
recycled. Product 1 receives no credit for
virgin materials savings even though the
availability of Product 1 for recycling
reduces virgin material manufacture for
making Product 2. However, this allocation
method is very useful when primary materi-
als and their productions are different or for
composted materials.
9O
-------
Issues Applicable To Specific Life-Cycle Stages
Third Allocation Method. In the third alloca-
tion method, the loadings added to the sys-
tem because of recycling are divided equally
between the two products. The loadings
associated with recycling are reduced dispo-
sal of Product 1, reduced virgin material
prqduct for Product 2, inputs and outputs
associated with recycling, and any convert-
ing net inputs and outputs incurred as a
result of using recycled materials over vir-
gin materials in Product 2. The net inputs
and outputs for this 50/50 allocation
method can be calculated by the following
formulas for each product:
NTO (Product 1) = D - l/2[B(a) + A(b) - C(a) - F]
Nwo {Product 2) = E - l/2[B(a) + A(b) - C(a) - F]
This allocation method has the following
advantages: It eliminates the possibility of
double-allocating recycling impacts, mini-
mizes arguments over which product
should receive the recycling impacts, and
enables independent evaluation of a
product. The limitation of the 50/50 alloca-
tion method is that it does not give credit to
the party (e.g., manufacturer, recycler,
municipality) that made the effort to imple-
ment the recycling change.
For all three methods typically, the impacts
of virgin material on both products must be
.assumed to be the1 same, because a manufac-
turer may have information on only the vir-
gin material for the product made in house.
The assumption of virgin material similarity
between the two products may be true for
some cases of recycling, such as recycling
milk bottles into detergent bottles, but may
not be true for other recycling situations,
such as recycling of some paper products.
Composting
Composting is an alternative to disposal by
which organic materials are removed from
the solid waste stream destined for the land-
fill. Composting is the controlled, biologi-
cal decomposition of organic materials into
a relatively stable humus-like product,
which can be handled, stored, and/or
applied to the land without adversely affect-
ing the environment (BioCycle, 1991). The
compost produced from the process has
only modest nutrient value, but is consid-
ered an excellent soil amendment. Com-
posted materials can replace peatmoss,
reduce the use of topsoil, agrichemicals,
water, and sometimes fertilizer, be used as
animal bedding, replace landfill cover mate-v
rial, etc. Thus, composting can be viewed as
a form of open-loop recycling. The process
decreases the volume of discarded material
occupying landfills; thus, when composting
is considered in a life-cycle inventory, the
solid waste for a given material is reduced.
Composted products may contain noncom-
postable materials; thus, noncompostable
solid waste resulting from the composting of
a product is attributed to that product in a
life-cycle inventory. For example, foil
found in some paperboard containers is not
compostable. It is, therefore, part of the
solid waste screened out of the composted
product and would be attributed to the solid
waste component of the paperboard
containers.
Emissions from Solid Waste Management
Options
The incineration and/or landfilling of mixed
municipal solid waste (MSW) creates prob-
lems in determining the atmospheric or
waterborne emissions from such processes
for particular products. For example, how
does one quantify the contribution a milk
91
-------
Issues Applicable To Specific Ufa-Cycle Stages
bottle would make to the leachate collected
from a landfill compared to the contribution
of a newspaper? Or the amount of incinera-
tor emissions resulting from the combustion
of a plastic film compared to the combus-
tion of a paperboard box? Little research on
specific products or even specific materials
has been done. The primary historical
option has been to omit reporting the emis-
sions from solid waste management pro-
cesses, due to the difficulty in accurately
reporting the emissions from the.combus-
tion of MSW or from leachate and air emis-
sions from mixed solid waste in a landfill.
Whenever a product disposition option spe-
cific to a particular product is used, emis-
sions from the disposition process are
quantifiable for that product and can be
reported. However, currently no accurate
method exists to allocate incinerator or
landfill emissions to a particular product
once it has been combined with other mixed
solid waste. For example, composting pro-
grams for yard waste only and diapers only
are able to collect and report emissions for
the composting of those particular materi-
als. Recycling processes also are nearly
always specific to a particular material. Alu-
minum, paper, plastics, and glass require
separate processes to convert discarded
products into usable materials.
Nevertheless, an attempt should be made to
account for MSW emissions in a life-cycle
inventory whenever possible. Several poten-
tial approaches exist for estimating these
emissions, but additional research is needed
to establish the accuracy and utility of these
methods. For almost any of these
approaches, it is important to know the per-
cent of a particular waste that is recycled,
incinerated, or landfilled. These waste dis-
posal percentages are available on a national
basis for typical MSW, but they may not be
relevant for specific products. For example,
some products have a high recycling rate
although the generic materials may not, and
other products are not recycled at all
although the general material type may be.
Also, emissions data from atypical MSW
incinerator or landfill are available, but the
materials in the mixed waste causing these
emissions are not specified. One option for
using both the waste disposal percentages
and the typical emissions from MSW incin-
erators or landfills is to proportionally allo-
cate the emissions based on a broad input
material composition. The range of selected
landfill leachate descriptors has been
reported in Chian et al. (1986). Landfill
emissions associated with a few specific
types of materials also are available, e.g.,
dry cell batteries (Jones et al.,, 1977) and
plastics (Wilson et al., 1982),
Lacking even this information, emissions
from a "typical" facility could be allocated
based on the weight percentage of MSW that
the product comprises. Similarly, incinera-
tor emissions associated with a few specific
materials are available, e.g., polyvinylchlo-
ride (PVC) (Carroll, 1988). The major groups
of incinerator air emissions listed by SET AC
(Fava et al., 1991) include CO2 and H2O, cri-
teria pollutants (NOX, SOX» pjirticulates,
VOCs, lead, and CO), products of incom-
plete combustion (PICs) and particulate
organic chemicals (POCs), heavy metals,
dioxans and furans, and waste heat.
A second option is to make idealized esti-
mates of emissions associated with waste
management methods by simulation model-
ing. Although the modeling methods have
not been validated by trial burns for very
many materials, incinerator emissions may
-------
Issues Applicable To Specific Life-Cycle stages
be estimated by equilibrium or stoichiome-
tric methods.
A third option for estimating emissions
associated with the incineration or landfill-
ing of a particular product is to analyze the
leachate or air emissions composition com-
ing from a mixed waste and assign emis-
sions to the product based on degradation
product fingerprinting. This option would
be highly resource intensive.
Energy from Incineration
Many products release energy when burned
in an incinerator. This energy can and has
been credited to a product, reducing the net
reported energy requirements for the life
cycle of the product. The energy credit tra-
ditionally is determined using the higher
heating value (HHV) of the material, taking
into account the moisture content of the
waste as disposed. The value of the recov-
ered energy for products manufactured from
fossil fuels, e.g., plastics, will not equal the
energy of the raw material resource attrib-
uted to the product, because the change in
chemical structure changes the recoverable
amount of energy.
In principle, it should be possible to employ
a second option if the actual delivered
energy is calculated by adjusting for moist-
ure as well as system thermal losses, e.g.,
heat transfer to walls and thermodynamic
conversion losses.
93
-------
REFERENCES
BioCycle, 1991. The Art and Science of Composting. Edited by the staff of BioCycle, Journal
of Waste Recycling. J.G. Press, Inc., Emmaus, PA.
Boustead, I. and G. F. Hancock, 1979. Handbook of Industrial Enarpv Analysis. Chichester:
Ellis Horwood and New York: John Wiley, ISBN 0-470-26492-6, Chapter 3, "Real Industrial
Systems," p. 76.
Boustead, I, undated. "The Relevance of Re-Use and Recycling Activities for the LCA Profile
of Products," 10 p.
Brown, H.L., B.B. Hamel, B.A. Hedman, et al., 1985. Energy Analysis of 108 Industrial Pro-
cesses. Drexel University, Philadelphia, PA. Prepared for U.S. Department of Energy.
Fairmont Press, 314 p.
Canadian Electric Utilities and National Energy Board, 1992. Personal Communication
between Raffaele DiGirolamo, Energy, Mines, and Resources Canada, and Ilia Amerson,
Battelle.
Carroll, W.F., Jr., 1988. TVC and Incineration." T. Vinvl Technol. 10(2):90-94.
Chian, E.S.K., S.B. Ghosh, B. Kahn, M. Giabbi, -and F.G. Pohland, 1986. Cndisposal of Low
Level Radioactive Waste and Sanitary Waste. SCEGIT Report No. 86-01.
DOE (U.S. Department of Energy), 1992, "Monthly Power Plant Report," Energy Information
Administration, EIA-759.
Fava, J.A., R. Denison, B. Jones, M.A. Curran, B. Vigon, S. Selke, and J. Barnum (Editors),
1991. A Technical Framework for Life-Cvrle Assessments. Society of Environmental Toxi-
cology and Chemistry and SETAC Foundation .for Environmental Education, Inc., Washing-
ton, D.C.
Fava, J.A., R. Denison, T. Mohin, and R. Parrish, 1992. "Life-Cycle Assessment Peer Review
Framework." Society of Environmental Toxicology and Chemistry, Life-Cycles Assessment
Advisory Group, 4 p.
Jones, C.J., P.J., McGugan, and P.F. Lawrence, 1977. "An Investigation of the Degradation of
Some Dry Cell Batteries Under Domestic Waste Landfill Conditions." T. Hazard. Mater..
2:259-289.
94
-------
. ' . . - References
Jorgensen, M.S. and B. Pedersen, 1990. "Quality Concepts for Processed Organic Food." Let-
ter Attachment to Timothy Mohin, U.S. Environmental Protection Agency from Bo
Pedersen, Interdisciplinary Centre, Technical University of Denmark, Lyngby, Denmark,
October 17,1990.
Lundholm, M.P. and G. Sundstrom, 1985. "Tetra Brik Aseptic Environmental Profile," 174 p.
Meadows, D. H. et-al., 1972. The Limits to Growth; a Report for the Club of Rome's Project
on the Predicament of Mankind Universe Books, New York. 205 p.
Raiffa, H., 1968. Decision AnalysisIntroductory Lectures on Choices under Uncertainty.
Addison-Wesley, Reading, MA.
Sauer, B.J., R.G. Hunt, and M.A. Franklin, 1990. "Background Document on Clean Prod-
uctsResearch and Implementation." U.S. Environmental Protection Agency, Risk Reduc-
tion Engineering Laboratory, Cincinnati, OH. EPA/600/2-90/048. 75 p.
Tillman, A.M., H. Baumann, E. Eriksson, and T. Rydberg, 1991. "Life-cycle Analyses of
Selected Packaging Materials: Quantification of Environmental Loadings." Report from
Chalmers Industriteknik to the Swedish Commission on Packaging, 206 p.
Werner, A.F., 1991. "Product Lifecycle Assessment: A Strategic Approach." Proceedings of
the Global Pollution Prevention '91 Conference. Washington D.C.
Wilson, D.C., P.J. Young, B.C. Audson, and G. Baldwin, 1982. "Leaching Cadmium from Pig-
mented Plastics in a Landfill Site." Environ. Sci. Tech. lfif9):Hftn.
95
-------
GLOSSARY
Accidental emission:
Atmospheric emissions:
Biomass:
Brines (oilfield):
Btu (British thermal unit):
By-product
Closed-loop recycling:
Composite data:
Composting:
Consumer use:
An unintended environmental release. Examples: crude oil
spills resulting from tanker accidents, venting of reactors due
to mechanical failure or human error. ,
Residual discharges of emissions to the air (usually expressed
in pounds or kilograms per unit output) following emission
control devices. Includes point sources such as stacks and
vents as well as area sources such as storage piles.
The weight of living material, occasionally used as an energy
source.
Wastewater produced along with crude oil and natural gas
from oilfield operations.
The quantity of heat energy required to raise the temperature
of 1 pound of water (air-free) from 60° to 61° Fsahrenheit at a
constant pressure of 1 standard atmosphere. Experimentally
equal to 1,054.5 joules.
A useful product that is not the primary product being pro-
duced. In life-cycle analysis by-products are treated as
coproducts.
A recycling system in which a particular mass of material is
remanufactured into the same product (e.g., glass bottle into
glass bottle).
Data from multiple facilities performing the same operation
that have been combined or averaged in some manner.
A waste management option involving the controlled biologi-
cal decomposition of organic materials into a relatively stable
humus-like product that can be handled, stored, and/or
applied to the land without adversely affecting the
environment.
The intended end use of a product. The use for which a
product was designed.
96
-------
Glossary
Coproduct:
Coproduct allocation:
Energy characterization:
Energy of material resource:
Energy profile:
Environmental loadings:
Environmental release:
Equivalent usage ratio:
Error analysis:
Finished product:
Fuel-related emissions:
Fuel-related wastes:
Fuel unit:
Fugitive emissions:
A marketable by-product from a process. This includes mate-
rials that may be traditionally defined as wastes such as
industrial scrap that is subsequently used as a raw material
in a different manufacturing process.
Adjustment of material inputs, energy requirements, and
environmental emissions from a process to allocate those
impacts attributable to the output product being considered.
Classification of energy according to primary fuel source:
wood, natural gas, petroleum, coal, nuclear, hydropower.
The fuel value of the raw materials used to make a product.
The inherent energy in a product made from a raw material
used as a fuel supply. Also known as latent energy.
A listing of the energy usage for a system by stage and/or by
source. See also energy characterization.
The releases of pollutants to the environment.
Emissions or wastes discharged to the air, land, or water.
Contaminants that cross a system boundary into the
environment.
A method of comparing two or more different products oil an
equivalent basis. For example, a comparison of beverage con-
tainers based upon the quantity of beverage delivered to the
consumer.
A systematic method for analyzing differences between a
measured or estimated quantity and the true value.
The product produced by the system being evaluated for con-
sumer purchase.
See fuel-related wastes.
Those materials or emissions generated during the combus-
tion of fuels for the production of heat, steam, electricity, or
energy to power processes and transportation equipment that
are not a component of the useable product or coproducts.
Weight or volume of fuel such as gallons of fuel oil, pounds
of coal, or cubic feet of natural gas.
Emissions from valves or leaks in process equipment or mate-
rial storage areas that are difficult to measure and do not flow
through pollution control devices.
97
-------
Glossary
Gigajoules (GJ):
Global warming:
Greenhouse effect:
Greenhouse gas:
Higher heating value (HHV):
Impact analysis:
Improvement analysis:
Incineration credit:
Industrial scrap:
Industrial solid waste:
Inherent energy:
Intermediate materials:
Inventory analysis:
Joule:
Leachate:
Life cycle:
1,000,000,000 or 109 joules.
The theory that elevated concentrations of certaiin atmo-
spheric constituents are causing an increase in the earth's
average temperature.
The theory that certain atmospheric constituents trap heat in
the earth's atmosphere leading to global warming.
An atmospheric constituent, such as carbon dioxide, that is
thought to contribute to global warming.
The gross heat of combustion for a material.
The assessment of the environmental consequences of energy
and natural resource consumption and waste releases associ-
ated with an actual or proposed action.
The component of a life-cycle assessment that iis concerned
with the evaluation of opportunities to effect reductions in
environmental releases and resource use.
The energy credit given in a life-cycle inventory for the burn-
ing of material hi a waste-to-energy incinerator.
The waste materials of value produced by a msmufacturing
process. The material is often reused within the same pro-
cess. It may also be sold to another operation as a raw
material.
Industrial solid waste includes wastewater treatment sludges,
solids from air pollution control devices, trim or scrap mate-
rials that are not recycled, fuel combustion residues (such as
the ash generated by burning wood or coal), and mineral
extraction residues.
See Energy of Material Resource.
The materials made from raw materials and from which fin-
ished products are made.
See Life-cycle Inventory.
SI (metric) unit of energy, equal to 9.48x10-* Bf u.
The solution that is produced by the action of percolating
water through a permeable solid, as in a landfill.
The stages of a product, process, or package's life, beginning
with raw materials acquisition, continuing through
-------
Glossary
Life-cycle assessment:
Life-cycle inventory:
Life-cycle stages:
MJ value:
Model (computational):
Municipal Solid
Waste (MSW):
National electricity grid:
Nonrenewable resource:
processing, materials manufacture, product fabrication, and
use, and concluding with any of a variety of waste manage-
ment options.
A concept and a methodology to evaluate the environmental
effects of a product or activity holistically, by analyzing the
entire life cycle of a particular product, process, or activity.
The life-cycle assessment consists of three complementary
componentsinventory, impact, and improvementand an
integrative procedure known as scoping.
The identification and quantification of energy, resource us-
age, and environmental emissions for a particular product,
process, or activity.
The stages of any process, including raw materials and en-
ergy acquisition; manufacturing (including materials manu-
facture, product fabrication, and filling/packaging/
distribution steps); use/reuse/maintenance; and recycle/
waste management.
Megajoule, 1,000,000 joules, equal to 948 Btu. See Joule.
A computational framework, usually a computer spreadsheet
or other such tool, that incorporates the stand-alone data and
materials flows into the total results for the energy and
resource use and environmental releases from the overall
system.
MSW includes wastes such as durable goods, nondurable
goods, containers and packaging, food wastes, yard wastes,
and miscellaneous inorganic wastes from residential, com-
mercial, institutional, and industrial sources. MSW does not
include wastes from other sources, such as municipal
sludges, combustion ash, and industrial nonhazardous pro-
cess wastes that might also be disposed of in municipal waste
landfills or incinerators.
The electricity generated by individual generators nationally
that are interconnected to form regional grids and also a
national grid.
A resource that cannot be replaced in the environment as fast
as it is being consumed.
99
-------
Glossary
Open-loop recycling:
Overburden:
Packaging, primary:
Packaging, secondary:
Packaging, tertiary:
Postconsumer solid waste:
Precombustion energy:
Predominant industrial
practice:
Process:
Process emissions:
Process energy:
Process-related wastes:
A recycling system in which a product made from one type
of material is recycled into a different type of product. The
product receiving recycled material itself may or may not be
recycled.
The material to be removed or displaced that is overlying the
ore or material being mined. ^
The level of packaging that is in contact with the product.
For certain beverages, this might be the 12-ounce aluminum
can.
The second level of packaging for a product that contains one
or more primary packages. For 12-ounce beverage cans, this
might be the plastic rings to hold the 6-pack together.
The third level of packaging for a product that contains one
or more secondary packages. For 6-packs of 12-ounce bever-
ages cans, this might be the corrugated trays and stretch wrap
over the pallet that are used in transporting the product.
A material that has served its intended use and has become a
part of the waste stream.
Energy required to extract, transport, and process the fuels
used for power generation. Includes adjustment for ineffi-
ciencies in power generation and for transmission losses.
Also known as energy of fuel acquisition.
A practice generally acknowledged to be widely used as a
significant percentage either of companies in the industry or
of the total material flow.
The operations that make up subsystems.
Waste materials generated or produced from the raw materi-
als, reactions, processes, or related equipment inherent to the
process.
The energy required for each subsystem for process require-
ments. These are quantified in terms of fuel or power units
such as gallons of distillate oil, cubic feet of natural gas, or
kilowatt-hours (kWh) of electricity.
The waste materials generated or produced from the raw
materials, reactions, processes, or related equipment inherent
to the process.
too
-------
Glossary
Random error:
Raw materials:
Recycled content:
Regional electricity grid:
Regulated emissions:
Renewable resource:
REPA:
Representativeness:
Residual oil:
Residues:
Resource requirements:
Risk assessment:
Error that does not have any definite or systematic pattern or
bias.
The total inputs for a subsystem including all material pres-
ent in the product and material found in losses due to emis-
sions, scrap and off-spec products, and no-emission losses
(such as moisture due to evaporation). Water is not always a
raw material input because it is often removed during a dry-
ing step.
The amount of recovered material, either pre- or postcon-
sumer, in a finished product that was derived from materials
diverted from the waste management system. Usually
expressed as a percent by weight.
The mix of fuel sources used to generate electricity for a
given region. A regional electricity grid is occasionally used
in a life-cycle inventory when an energy-intensive industry is
sited in a certain region to take advantage of inexpensive
electric power.
Those emissions regulated by government to limit amounts
or concentrations of waste.
A resource that can be replaced in the environment faster
than it is being consumed.
Resource and Environmental Profile Analysis. Also com-
monly called cradle-to-grave analysis or life-cycle analysis.
The state of being a sample that is characteristic of a group or
population of operations or processes.
The heavier oils that remain after the distillate fuel oils and
lighter hydrocarbons are removed in refinery operations.
Included are No. 5 and No. 6 oils.
Process wastes (such as wood bark), typically but not always
solids.
The amounts of raw materials or natural inputs and energy
used in a system.
An evaluation of potential consequences to humans, wildlife,
or the environment caused by a process, product, or activity
and including both the likelihood and the effects of an event.
1O1
-------
Glossary
Routine emissions:
Sensitivity analysis:
SI (Systeme Internationale):
Soil amendment:
Solid waste:
Specific data:
Stand-alone data:
Subprocess:
Subsystem:
System:
Systematic error:
Systems analysis:
Template:
Ton-mile:
Those releases that normally occur from a process, as
opposed to accidental releases that proceed from abnormal
process conditions.
A systematic evaluation process for describing the effect of
variations of inputs to a system on the output.
Internationally used system of standards for units and
dimensions.
The product of composting of organic materials that is
applied to the soil as a conditioning agent.
Solid products or materials disposed of in landfills, inciner-
ated, or composted. Can be expressed in weight or volume
terms.
Data that are characteristic of a particular subsystem or
process.
Normalized data consistently defining the system by report-
ing the same product output from each subsystem (e.g., on
the basis of 1,000 pounds of output).
An individual step that is a part of a defined process.
An individual process that is a part of the defined system.
A collection of operations that perform a desired function. In
a life-cycle inventory, the scope of the system is defined by
the boundary conditions.
Error that is not random. Variation caused, for example, by
differences in age of equipment, advances in technology, or
local conditions.
A stepwise evaluation of the inputs and outputs of a defined
system.
A guide used by analysts for collecting and organizing data.
The template describes a material and energy balance for a
defined system or subsystem. It includes resource require-
ments, transportation requirements, and emissions and
wastes for that system or subsystem.
A measure of the movement of 1 ton (2,000 pounds) of freight
for the distance of 1 mile. For example, 100 ton-miles is the
measure for moving 100 tons of freight 1 mile. It could also
represent moving 1 ton of freight 100 miles.
102
-------
Transmission line loss:
Transportation energy:
Waterborne wastes:
Glossary
The difference between electricity generated and electricity
delivered.
Energy required to transport materials and products through-
out the process and to final distribution to the consumer.
This is converted from the conventional units of "ton-miles"
by each transport mode (e.g., truck, rail, barge, airfreight,
pipeline, etc.) using on the average efficiency of each mode.
Discharges to water of regulated pollutants (usually
expressed hi kilograms per unit output) after existing
v treatment processes.
103
-------
APPENDIX
CI!eanAirAct(CAA)
The Clean Air Act (CAA) is a piece of
national legislation designed to identify and
control pollutants and sources of emissions
that may reduce the quality of the nation's
air. The CAA has been amended twice
since its inception in an effort to adjust it to
our changing perception and knowledge of
the environment. The most recent amend-
ments in 1990 identify 189 toxic substances
and pollutants as well as sources of emis-
sions that enter the nation's air. The objec-
tive of the CAA is to restore and maintain
the chemical, physical, and biological integ-
rity of the nation's air.
Under CAA Title IE, industrial facilities are
subject to new source performance stan-
dards for new facilities to be constructed or
notification of changes to existing ones after
the date the EPA proposes new source per-
formance standards (40 CFRPart 60).
There are 174 categories of sources listed
pursuant to CAA Title m. These include
steam generators (both fossil fuel and petrol-
eum), incinerators, cement plants, chemical
acid plants, petroleum refineries, sewage
treatment plants, metal smelting facilities,
fertilizer production plants, steel plants,
paper mills, glass manufacturing plants,
synthetic fiber production facilities, syn-
thetic organic chemical manufacturing, and
nonmetallic mineral processing plants.
In CAA subpart A, section 61.01, a list of
pollutants deemed hazardous and their
applicability has been published pursuant
to CAA section 112 (CAA-5).
In CAA Part 61, a list of National Standards
has been created to control those emissions
deemed hazardous air pollutants. Examples
of the emissions are radon from uranium
mines, beryllium, mercury, vinyl chloride,
radionuclides other than radon, benzene
leaks, radionuclide emissions from phos-
phorus plants, asbestos, and inorganic ar-
senic emissions from glass manufacturing
and primary copper smelting.
A priority list of major source categories is
given in CAA section 60.16. Among the
largest sources of hazardous pollutants are
synthetic organic chemical manufacturing,
petroleum refineries, dry cleaning, graphic
arts, stationary combustion engines, and in-
dustrial surface coating of fabric. In all there
are 59 sources; however, several of these
sources are no longer considered priority
sources, i.e., mineral wool, secondary cop-
per, and ceramic clay manufacturing.
Title I of the CAA Amendments of 1990
extends and revises the existing require-
ments for attaining and maintaining the
National Ambient Air Quality Standards
(NAAQS) for the following six criteria pol-
lutants: ozone, carbon monoxide (CO), par-
ticulate matter (PM10), lead (Pb), sulfur
dioxide (SO2), and nitrogen oxides (NOX).
Title I also addresses permit requirements
and emissions inventories for existing
stationary sources. The primsiry industry
segments impacted by Title I include
104
-------
APPENDIX
chemicals, pulp and paper, petrochemicals,
Pharmaceuticals, iron and steel, and most
manufacturing industries. Title I also
changed the definition of major stationary
source relative to volatile organic carbon
emissions (VOCs).
Title H of the CAA Amendments of 1990
covers mobile emission sources. The
requirements affect tailpipe emission stan-
dards for CO2, hydrocarbons, and CO.
Title in of the CAA Amendments of 1990
established emission standards for 189 air
toxics, or hazardous air pollutants (HAPs),
such as acrylonitrile or chlorine. The list of
HAPs is divided into industry groups, such
as polymer and resin production, and then
identifies individual source categories, such
as cellophane or polystyrene production.
Title El also provides for comprehensive
regulation of solid waste incinerators and
for development of a list of at least 100
HAPs which, if released accidentally, could
seriously threaten human health or the envi-
ronment.
Title IV of the CAA Amendments of 1990
seeks to reduce annual emissions of SO2 and
NOX, because they are principal components
hi the formation of acid rain.
Title VI of the CAA Amendments of 1990
seeks to reduce threats to the stratospheric
ozone layer by phasing out production and
use of chlorofluorocarbons (CFCs), halons,
and other widely used chemicals believed
to contribute to global warming.
Clean Water Act (CWA)
The Clean Water Act (CWA) is a Federal
statute that addresses the quality needs of
the nation's waterbodies, with regard to
both human and environmental concerns. It
is written to control known, possible, and
unknown toxins and pollutants through
proper use and disposal. The objective of
the CWA is to restore and maintain the
chemical, physical, and biological integrity
of the nation's water.
Appendix B-65, Toxic Pollutants, is one of
the earliest written and most important
tables for identifying toxic pollutants. These
toxic pollutants are composed of organics,
inorganics, heavy metals, cyanogens, halo-
gens, and PCB compounds. The amended
list is found in 53 FR 46015, October 17,
1988, and is incorporated by reference in
CWA section 307(a)(l).
In addition, the CWA designates under sec-
tion 301(2)(C)(f) categories of pollutant
emissions for organic chemicals, plastics,
synthetic fibers, and pesticides. Besides list-
ing toxic pollutants, the CWA lists 27 cate-
gories of sources that discharge toxic waste,
e.g., pulp and paper mills, dairy product
processing, textile mills, feedlots, electro-
plating industries, plastic and synthetic ma-
terials manufacturing, and petroleum
refining.
The term TTO, "total toxic organics," is
used to describe a group of chemicals with
quantifiable amount greater than 0.01 milli-
gram per liter. This expanded list of 126
toxic organics was set up to establish efflu-
ent limitations under CWA section
301(b)(2)(C).
CWA subpart D, section 414.40 has a list of
Standard Industrial Code (SIC) 28213
thermoplastic resins and thermoplastic
resin groups that are applicable to the pro-
cess wastewater discharges resulting from
the manufacture of thermoplastic resins.
CWA subpart G, section 414.70 lists bulk
organic chemicals that are associated with
105
-------
APPENDIX
wastewater resulting from the manufacture
of SIC 2865 and 2869 bulk organic chemi-
cals and organic chemical groups. The list is
broken down into five groups: aliphatic,
amine and amide, aromatic, halogenated,
and other organic chemicals.
Subpart F, section 414.60, Commodity
Organic Chemicals lists the applicable com-
pounds in process wastewater resulting
from the manufacture of SIC 2865 and 2869
commodity organic chemical groups.
Comprehensive Environmental
Response, Compensation and
Liability Act (CERCLA)
The Comprehensive Environmental
Response, Compensation and Liability Act
(CERCLA) is a national law designed to
regulate the cleaning up of environmental
contaminations made before and not cov-
ered under the Resource Conservation and
Recovery Act (RCRA). CERCLA has been
termed "Superfund" because of the large
amounts of money appropriated by Con-
gress to clean up the nation's environment.
As an amendment to CERCLA, the Super-
fund Amendments and Reauthorization Act
(SARA) gives CERCLA new strength by pro-
viding new cleanup standards, schedules,
and provisions aimed at federal facilities
and increased settlement, liability, and
enforcement powers for the EPA and
citizens.
In 40 CFR 302.4 is a list of hazardous sub-
stances and reportable quantities that are
the chemical compounds regulated by
CERCLA; the Chemical Abstract Services
Reference Number (CASRN) of each com-
pound; its synonym; the statutory source for
designation of the hazardous substance
under CERCLA-CWA, CAA, and RCRA; and
the reportable quantities. Each hazardous
substance is listed in alphabetical order and
grouped with its respective family, i.e., anti-
mony and compounds, arsenic and com-
pounds, polychlorinated biphenyls (PCBs),
spent halogenated solvents used in degreas-
ing, wastewater treatment sludges from elec-
troplating operations, and spent catalyst
from the hydrochlorinator reactor in pro-
duction of 1,1,1-trichloroethame.
CERCLA also regulates radionuclides, and
in Appendix B of 40 CFR 302.4 is a table
with an alphabetic listing of the hazardous
substances along with their reportable quan-
tities specified in curies.
Also in 40 CFR 302.4, Appendix B of
CERCLA is the "List of Extremely Hazard-
ous Substances and Their Threshold Plan-
ning Quantities." This table is listed
according to CASRN and has a column for
notes that can include such information as
Threshold Planning Quantity (TPQ), the
statutory reportable quantity,, and toxicity
criteria.
Resource Conservation and
Recovery Act (RCRA)
The Resource Conservation and Recovery
Act (RCRA) was enacted toji.ll the regula-
tory pollution control gap between the
Clean Air Act and the Clean Water Act.
Despite its name, RCRA's primary purpose
has been to control the disposal of hazard-
ous and solid wastes generated by various
manufacturing processes and the air and
water pollution control devices installed for
those processes. Waste management catego-
ries within RCRA include the following sub-
titles: Subtitle CHazardous, Subtitle
DNonhazardous, and Subtitle JMedical
1O6
-------
APPENDIX
Waste. All hazardous wastes must first meet
the definition of a solid waste. A solid
waste is any garbage, refuse, sludge, and
other discarded material, including solid,
liquid, semi-solid, or contained gaseous
material resulting from industrial, commer-
cial, mining, and agricultural operations,
and from community activities, except
domestic sewage, irrigation return flows, or
industrial discharges controlled as point
sources under the Federal Water Pollution
Control Act or materials controlled under
the Atomic Energy Act. Discarded solid
wastes include abandoned materials, recy-
cled materials, and inherently waste-like
(dioxin-containing) materials.
A hazardous waste is defined as any solid
waste, or combination of solid wastes
which, because of its quantity; concentra-
tion; or physical, chemical, or infectious
characteristics, may (1) cause, or signifi-
cantly contribute to an Increase in mortality
or an increase in serious irreversible, or in-
capacitating reversible, illness; or (2) pose a
substantial present or potential hazard to
human health or the environment when
improperly treated, stored, transported, or
disposed of, or otherwise managed. Hazard-
ous wastes include those that either exhibit
a hazardous waste characteristics, as
defined in subpart C of 40 CFR Part 261 or
are a hazardous waste listed in subpart D.
Characteristic hazardous wastes, as defined
in subpart C of 40 CFR Part 261, include
those which exhibit one of the following
characteristics:
Ignitability, as defined in section 261.21,
applies to solid wastes that are capable of
causing fires during routine handling
and/or significantly increasing the
dangers of a fire once one is started.
Corrosivity, as defined in section 261.22,
applies to liquid wastes with a pH of less
than 2 or more than 12.5 and solid wastes
with the ability to corrode steel.
Reactivity, as defined in section 261.23,
applies to the capability of a waste to
explode, undergo violent chemical
change in a variety of situations, or react
violently with water to produce toxic
fumes or vapors.
Toxicity, as defined in section 261.24,
applies to the capability of a solid waste
to release into water any of 40 toxic con-
stituents in concentrations above regula-
tory levels established by the EPA. The
standard test method for this characteris-
tic is known as the Toxicity Characteristic
Leaching Procedure (TCLP).
Specific wastes have been identified as haz-
ardous by the EPA because of known haz-
ardous characteristics. These types of
wastes and the locations of their lists in
40 CFR are given below:
Manufacturing wastes from nonspecific
sources (F code wastes) are listed in
section 261.31.
Manufacturing wastes from specific
industrial processes (K code wastes) are
listed in section 261.32.
Discarded chemical products or interme-
diates that are acutely toxic wastes (i.e., if
the LDgo is less than 50 mg/kg) (P code
wastes) are listed in section 261.3 3 (e).
Discarded chemical products or interme-
diates that present risks of chronic toxic-
ity from exposure (U code wastes) are
listed in section 261.33(f).
107
-------
APPENDIX
TOXIC SUBSTANCES CONTROL ACT
CTSCA)
The Toxic Substances Control Act (TSCA)
includes regulations and testing require-
ments for every chemical substance that is
manufactured for commercial purposes in
the United States or imported for commer-
cial purposes. The following chemicals are
regulated under TSCA with respect to pro-
cessing, use, and disposal, as well as warn-
ings and instructions that must accompany
the substance when distributed: polychlori-
natedbiphenyls (PCBs), fully halogenated
chlorofluoroalkanes (CFCs), and asbestos.
TSCA Part 761 establishes prohibitions of,
and requirements for, the manufacture, pro-
cessing, distribution in commerce, use, dis-
posal, storage, and markings of PCBs and
PCB items. Substances with PCBs that are
regulated by this rule include, but are not
limited to, dielectric fluids, contaminated
solvents, oils, waste oils, heat transfer
fluids, hydraulic fluids, paints, sludges,
slurries, dredge spoils, soils, materials con-
taminated as a result of spills, and other
chemical substances or combinations of
substances.
TSCA Part 746 prohibits the manufacture,
processing, and distribution of CFCs as
aerosol propellants, except for export. Two
other classes of exemptions for CFG propel-
lants are (1) for use in an article which is a
food, food additive, drug, cosmetic, or
exempted device, and (2) for essential and
exempted uses listed in sections 762.58 and
762.59.
TSCA Part 763, subpart D requires reporting
by persons who manufacture, import, or
process asbestos. Part 763, subpart I prohib-
its the manufacture, importation, process-
ing, and distribution in commerce of the
asbestos-containing products identified and
at the dates indicated. This subpart requires
that products subject to this rule's bans, but
not yet subject to a bah on distribution in
commerce, be labeled.
One of the major goals of TSCA is to
develop test data which are necessary to
determine whether chemical substances and
mixtures present an unreasonable risk to
health or the environment. Under Section 4
of TSCA, the EPA can require chemical
manufacturers, importers, and processors to
conduct and pay for those tests. TSCA Part
799 identifies the chemical substances, mix-
tures, and categories of substances and mix-
tures for which data are to be developed,
specifies the persons required to test, speci-
fies the test substance(s) in each case, pre-
scribes the tests that are required, and
provides deadlines for submission of reports
and data to EPA. Part 766 identifies testing
requirements to ascertain whether certain
specified chemical substances may be con-
taminated with halogenated dibenzodioxins
(HDDs)/dibenzofurans (HFDu), as well as
requirements for reporting these analyses.
108
Jf.V.S. GOVERNMENT PRINTING OFFICE: 1995 - 650-005/00230
-------
-------
-------
-------
8 1 «:
-« CO
TJ 3"
S'
------- |