A COMPARATIVE ANALYSIS OF COMMERCIALLY AVAILABLE
LIFE CYCLE ASSESSMENT SOFTWARE
By
Ricky Lee Speck

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Packaging – Doctor of Philosophy
2014

ABSTRACT
A COMPARATIVE ANALYSIS OF COMMERCIALLY AVAILABLE
LIFE CYCLE ASSESSMENT SOFTWARE
By
Ricky Lee Speck
Against a backdrop of increasing demand for products, the packaging industry is
continually searching for ways to meet the needs of consumers while simultaneously
improving the environmentally sustainability of packaging. This requires making
informed decisions based at least in part on knowledge of the environmental impacts
attributable to the different packaging options being considered. Many companies have
turned to using Life Cycle Assessment (LCA) to obtain the necessary environmental
impact information. To support this effort, several software programs have been
developed to aid in doing an LCA. These programs vary considerably in complexity and
focus, leading to questions of how similar are the results they provide.
To study the consistency of results across LCA software programs, nine common
packaging systems were modeled in each of five different programs: GaBi, SimaPro,
openLCA, COMPASS, and Package Modeling. Comparison of the LCA information
provided by these programs for the packaging systems showed several significant
differences. To better understand why the differences occurred, four simplified systems
were created with the intent of minimizing the number of variables involved in the
system models. Each of these systems consisted of obtaining and disposing of a single
basic packaging material: aluminum, glass, corrugated board, and polyethylene
terephthalate (PET). Once again a comparison of results showed significant differences.

Using information from these basic material comparisons, and making use of the
transparency in LCA calculations provided by GaBi and SimaPro, a study was done that
traced the causes of the inconsistencies in results back to how implementations of
assessment methods differed between the two programs. A comparison of 14
combinations of basic materials and impact categories involving 3 different assessment
methods found 98 instances of characterization factors differing between GaBi and
SimaPro. These differences in characterization factors accounted for all the
discrepancies in results between the two programs. No errors in calculations were found
with either program, and there were no significant differences in the type or amount of
inputs and emissions in the life cycle inventories from the two programs.

ACKNOWLEDGMENTS

I wish to thank the Center For Packaging Innovation and Sustainability for
funding the study that forms the basis of this paper, the Sustainable Packaging Coalition
for providing the COMPASS software, and Thumbprint LTD for providing the Packaging
Modeling 3 software. I would like to thank Dr. Susan Selke, Dr. Rafael Auras, and Dr.
Diana Twede of the School Of Packaging at Michigan State University, along with Dr.
Satish Joshi of the Department of Agricultural, Food, and Resource Economics at
Michigan State University, for being members of my academic guidance committee.
Furthermore, I would like to thank Dr. Selke and Dr. Auras for their support in doing the
study. I would also like to acknowledge James Fitzsimmons for his contribution to
generating LCA data for the study.

iv

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... viii
LIST OF FIGURES......................................................................................................... xii
1. INTRODUCTION ......................................................................................................... 1
2. PACKAGING GROWTH CONSIDERATIONS ............................................................ 3
3. SUSTAINABILITY and LIFE CYCLE ASSESSMENT ................................................ 8
4. PUBLICATIONS ON LIFE CYCLE ASSESSMENT.................................................. 13
4.1 Introductions To Life Cycle Assessment ......................................................... 13
4.2 Life Cycle Assessment Procedural References .............................................. 14
4.3 Impact Methodologies ....................................................................................... 17
4.4 Journals and Other Publications ...................................................................... 21
5. REVIEW OF BASIC LIFE CYCLE ASSESSMENT PRINCIPLES ............................ 23
5.1 Goal Definition and Scope ................................................................................ 24
5.2 Life Cycle Inventory (LCI) .................................................................................. 27
5.3 Life Cycle Impact Assessment (LCIA) .............................................................. 29
5.4 Life Cycle Interpretation .................................................................................... 31
6. LIFE CYCLE ASSESSMENT SOFTWARE .............................................................. 34
6.1 SimaPro .............................................................................................................. 34
6.2 GaBi..................................................................................................................... 35
6.3 openLCA ............................................................................................................. 36
6.4 COMPASS ........................................................................................................... 36
6.5 Package Modeling .............................................................................................. 37
7. RESEARCH SUBJECT ............................................................................................ 38
8. RESEARCH METHODS ........................................................................................... 41
9. PACKAGE SYSTEM COMPARISONS ..................................................................... 50
9.1 Beverage Containers ......................................................................................... 50
9.2 Flexible Versus Rigid Containers ..................................................................... 57
v

9.3 Single Use Box Versus Reusable Crate ........................................................... 62
9.4 Comparison Summary for Package Systems .................................................. 73
10. BASIC MATERIAL COMPARISONS ...................................................................... 75
10.1 COMPASS, SimaPro, and GaBi Comparisons ............................................... 76
10.2 SimaPro and openLCA Comparisons ............................................................ 83
10.3 Other Impact Categories ................................................................................. 87
10.4 Comparison Summary For Basic Materials ................................................... 93
11. UNDERLYING CAUSES OF DISSIMILAR RESULTS ........................................... 94
11.1 IMPACT 2002+ .................................................................................................. 97
11.1.1 Aquatic Ecotoxicity ................................................................................... 98
11.1.2 Global Warming ....................................................................................... 105
11.1.3 Mineral Extraction.................................................................................... 110
11.1.4 Ozone Layer Depletion ............................................................................ 115
11.1.5 Respiratory Organics .............................................................................. 118
11.2 ReCiPe ............................................................................................................ 122
11.2.1 Human Toxicity ........................................................................................ 122
11.2.2 Ionizing Radiation .................................................................................... 127
11.2.3 Marine Ecotoxicity ................................................................................... 129
11.2.4 Marine Eutrophication ............................................................................. 134
11.2.5 Water Depletion ....................................................................................... 138
11.3 TRACI 2 ........................................................................................................... 140
11.3.1 Ecotoxicity ............................................................................................... 140
11.3.2 Global Warming ....................................................................................... 147
11.3.3 Non Carcinogens ..................................................................................... 150
11.3.4 Respiratory Effects .................................................................................. 155
11.4 Summary Of Underlying Causes Of Dissimilar Results ............................. 158
12. RESEARCH SUMMARY & CONCLUSIONS ........................................................ 159
APPENDICES ............................................................................................................. 162
APPENDIX A: Population, MSW, and Oil Consumption Data ............................. 163
APPENDIX B: Container Modeling Assumptions ................................................ 165
APPENDIX C: Container Compositions and Material Weights ........................... 169
APPENDIX D: Product LCA Flow Diagrams......................................................... 179
APPENDIX E: Test Parameter Combinations For Containers ............................ 188
APPENDIX F: Database Files Used In Software Modeling.................................. 197
APPENDIX G: Packaging Systems Comparison Data ........................................ 209
vi

APPENDIX H: Basic Materials Comparison Data ................................................ 214
BIBLIOGRAPHY ......................................................................................................... 218

vii

LIST OF TABLES

Table 1: Emerging markets ............................................................................................. 7
Table 2: Packaging systems compared ........................................................................ 42
Table 3: Parameters for basic material test scenarios .................................................. 47
Table 4: Parameters for beverage container comparison ............................................. 51
Table 5: Parameters for truck and rail transport comparison ........................................ 55
Table 6: Parameters for flexible versus rigid container comparison .............................. 58
Table 7: Parameters for reuse comparison ................................................................... 63
Table 8: Parameters for corrugated box recycled content comparison ......................... 67
Table 9: Parameters for air and ship transport comparisons ........................................ 71
Table 10: Parameters for comparison of four basic materials ....................................... 77
Table 11: Parameters for recycled content comparison of basic corrugated board. ..... 80
Table 12: Aluminum impact assessment data, 50% recycled content .......................... 88
Table 13: Corrugated board impact assessment data, 50% recycled content .............. 89
Table 14: Glass impact assessment data, 55.5% recycled content .............................. 90
Table 15: PET impact assessment data, 0% recycled content ..................................... 91
Table 16A: SimaPro / Impact 2002+ aquatic ecotoxicity for aluminum ....................... 101
Table 16B: GaBi / Impact 2002+ aquatic ecotoxicity for aluminum ............................. 102
Table 16C: SimaPro / Old Impact 2002+ variant aquatic ecotoxicity for aluminum ..... 103
Table 17A: SimaPro / Impact 2002+ global warming for corrugated board, part 1 ..... 106
Table 17B: GaBi / Impact 2002+ global warming for corrugated board, part 1 ........... 107
Table 18A: SimaPro / Impact 2002+ global warming for corrugated board, part 2 ..... 108

viii

Table 18B: GaBi / Impact 2002+ global warming for corrugated board, part 2 ........... 109
Table 19A: SimaPro / Impact 2002+ mineral extraction for glass ............................... 112
Table 19B: GaBi / Impact 2002+ mineral extraction for glass ..................................... 114
Table 20: SimaPro naming differences for mineral extraction ..................................... 114
Table 21A: SimaPro / Impact 2002+ ozone layer depletion for PET ........................... 116
Table 21B: GaBi / Impact 2002+ ozone layer depletion for PET................................. 117
Table 22: SimaPro naming differences for ozone depletion........................................ 117
Table 23A: SimaPro / Impact 2002+ respiratory organics for PET ............................. 119
Table 23B: GaBi / Impact 2002+ photochemical oxidation for PET ............................ 120
Table 24: SimaPro naming differences for respiratory organics ................................. 121
Table 25A: SimaPro / ReCiPe human toxicity for PET ............................................... 124
Table 25B: GaBi / ReCiPe human toxicity for PET ..................................................... 126
Table 26A: SimaPro / ReCiPe ionizing radiation for glass .......................................... 128
Table 26B: GaBi / ReCiPe ionizing radiation for glass ................................................ 128
Table 27A: SimaPro / ReCiPe marine ecotoxicity for corrugated board ..................... 131
Table 27B: GaBi / ReCiPe marine ecotoxicity for corrugated board ........................... 133
Table 28A: SimaPro / ReCiPe marine eutrophication for aluminum ........................... 136
Table 28B: GaBi / ReCiPe marine eutrophication for aluminum ................................. 137
Table 29A: SimaPro / ReCiPe water depletion for aluminum ..................................... 139
Table 29B: GaBi / ReCiPe water depletion for aluminum ........................................... 139
Table 30A: SimaPro / TRACI 2 ecotoxicity for glass ................................................... 143
Table 30B: GaBi / TRACI 2 ecotoxicity for glass ........................................................ 145
Table 31A: SimaPro / TRACI 2 global warming for corrugated board ........................ 148

ix

Table 31B: GaBi / TRACI 2 global warming for corrugated board .............................. 149
Table 32A: SimaPro / TRACI 2 non carcinogens for PET........................................... 152
Table 32B: GaBi / TRACI 2 non carcinogens for PET ................................................ 154
Table 33A: SimaPro / TRACI 2 respiratory effects for aluminum ................................ 157
Table 33B: GaBi / TRACI 2 respiratory effects for aluminum ...................................... 157
Table 34: Summary of differences in GaBi and SimaPro characterization factors ...... 158
Table A1: World population estimates ........................................................................ 163
Table A2: U.S. Population and waste generation estimates ....................................... 163
Table A3: World, U.S., and China oil consumption in thousands of barrels per day ... 164
Table C1: Weight of aseptic carton components ........................................................ 170
Table C2: Packaging material used for aseptic carton ................................................ 170
Table C3: Weight of flexible pouches.......................................................................... 172
Table C4: Packaging material used for flexible pouch ................................................ 172
Table C5: Weight of bottle components ...................................................................... 173
Table C6: Packaging material used for beer bottle ..................................................... 173
Table C7: Label areas for printing estimate ................................................................ 174
Table C8: Packaging material used for PET bottle ..................................................... 175
Table C9: Weight of PLA bottle components .............................................................. 176
Table C10: Packaging material used for PLA bottle ................................................... 176
Table C11: Weight of steel food can components ...................................................... 178
Table C12: Packaging material used for steel can ...................................................... 178
Table E1: Aluminum beverage can test parameter combinations ............................... 188
Table E2: Aseptic carton test parameter combinations ............................................... 189

x

Table E3: Corrugated box test parameter combinations ............................................. 190
Table E4: Flexible pouch test parameter combinations .............................................. 191
Table E5: Glass bottle test parameter combinations .................................................. 192
Table E6: PET bottle test parameter combinations ..................................................... 193
Table E7: PLA bottle test parameter combinations ..................................................... 194
Table E8: Plastic (PP) crate test parameter combinations .......................................... 195
Table E9: Steel food can test parameter combinations............................................... 196
Table F1: GaBi library/database files used ................................................................. 197
Table F2: openLCA library/database files used .......................................................... 201
Table F3: SimaPro library/database files used ........................................................... 202
Table G1: Beverage container comparison data ......................................................... 209
Table G2: Truck and rail transport comparison data for glass bottle ........................... 210
Table G3: Steel can and flexible pouch comparison data ........................................... 210
Table G4: Reuse comparison data on a per use basis ............................................... 211
Table G5: Recycled content comparison data for corrugated box .............................. 212
Table G6: Air and ship transport comparison data for corrugated box and PP crate .. 213
Table H1: Basic material comparison data ................................................................. 214
Table H2: Basic corrugated board recycled content comparison data ........................ 215
Table H3: openLCA basic material comparison data .................................................. 216
Table H4: openLCA corrugated board recycled content comparison data .................. 217

xi

LIST OF FIGURES

Figure 1: Estimate of world population from 1804 to 2028. ............................................ 3
Figure 2: Population and waste material generation in U.S. from 1960 to 2010. ........... 4
Figure 3: Oil consumption for China and the U.S. from 1980 to 2010. ........................... 6
Figure 4: Journal articles using LCA specific software for LCIA. .................................. 39
Figure 5: Greenhouse gas emissions for selected beverage containers. ..................... 52
Figure 6: Fossil fuel/non-renewable energy for selected beverage containers. ........... 52
Figure 7: Eutrophication for selected beverage containers. ......................................... 53
Figure 8: Water depletion/consumption for selected beverage containers. .................. 54
Figure 9: Greenhouse gas emissions for glass bottle using truck transport. ................ 55
Figure 10: Greenhouse gas emissions for glass bottle using rail transport. ................ 57
Figure 11: Greenhouse gas emissions for steel cans and flexible pouch. .................. 59
Figure 12: Fossil fuel/non-renewable energy for steel cans and flexible pouch. ......... 60
Figure 13: Eutrophication for steel cans and flexible pouch. ....................................... 61
Figure 14: Water depletion/consumption for steel cans and flexible pouch. ............... 62
Figure 15: Greenhouse gas emissions for reuse comparison. .................................... 64
Figure 16: Fossil fuel/non-renewable energy for reuse comparison. .......................... 64
Figure 17: Eutrophication for reuse comparison. ........................................................ 65
Figure 18: Water depletion/consumption for reuse comparison. ................................. 66
Figure 19: Greenhouse gas emissions for corrugated box comparison. ..................... 68
Figure 20: Fossil fuel/non-renewable energy for corrugated box comparison. ........... 69
Figure 21: Eutrophication for corrugated box comparison. ......................................... 69

xii

Figure 22: Water depletion/consumption for corrugated box comparison. .................. 70
Figure 23: Greenhouse gas emissions versus air transport distance. ........................ 72
Figure 24: Greenhouse gas emissions for ship transport of plastic (PP) crate. .......... 72
Figure 25: Greenhouse gas emissions for basic material comparison........................ 77
Figure 26: Fossil fuel/non-renewable energy for basic material comparison. ............. 78
Figure 27: Eutrophication for basic material comparison. ........................................... 79
Figure 28: Water depletion/consumption for basic material comparison..................... 79
Figure 29: Greenhouse gas emissions for basic corrugated board............................. 80
Figure 30: Fossil fuel/non-renewable energy for basic corrugated board. .................. 81
Figure 31: Eutrophication for basic corrugated board. ................................................ 82
Figure 32: Water depletion/consumption for basic corrugated board.......................... 82
Figure 33: openLCA greenhouse gas emissions for basic materials. ......................... 83
Figure 34: openLCA non-renewable energy for basic materials. ................................ 84
Figure 35: openLCA eutrophication for basic materials. ............................................. 84
Figure 36: openLCA water depletion for basic materials. ........................................... 85
Figure 37: openLCA greenhouse gas emissions for corrugated board. ...................... 86
Figure 38: openLCA non-renewable energy for corrugated board. ............................. 86
Figure 39: openLCA eutrophication for corrugated board. .......................................... 86
Figure 40: openLCA water depletion for corrugated board. ........................................ 87
Figure D1: Flow diagram for aluminum can used for carbonated beverages. ........... 179
Figure D2: Flow diagram for aseptic carton. ............................................................... 180
Figure D3: Flow diagram for corrugated box. ............................................................ 181
Figure D4: Flow diagram for flexible pouch used for packaging tuna. ....................... 182

xiii

Figure D5: Flow diagram for glass (amber) beer bottle. ............................................ 183
Figure D6: Flow diagram for PET carbonated beverage bottle. ................................ 184
Figure D7: Flow diagram for PLA water bottle. .......................................................... 185
Figure D8: Flow diagram for plastic (PP) crate. ......................................................... 186
Figure D9: Flow diagram for steel food can used for packaging tuna. ...................... 187

xiv

1. INTRODUCTION
The world’s natural resources are under intense pressure to meet the growing
needs and desires of its inhabitants. World population continues to increase, while at
the same time demand for goods of all types increases as undeveloped regions become
more modernized. Over the years conservation efforts have pushed their way to the
forefront of product and packaging development in an attempt to reach a balance
between resources and demands. For the packaging industry this has become an effort
to provide packaging that is environmentally sustainable while still meeting all of the
performance requirements needed to package a given product.
A key requirement in creating sustainable packaging is being able to compare
the environmental impacts of different packaging options. Without a common basis of
comparison there is no credible means of determining that any option is more
environmentally sustainable than the others available. An approach to estimating
environmental impact that is seeing increasing use in packaging is Life Cycle
Assessment (LCA). In conducting an LCA the contribution of each step in a
product/package life cycle, from getting the raw materials all the way through to disposal
at end of useful life, can be looked at and combined to give a picture of the overall
environmental impact.
While LCA can provide a clearer picture of the environmental impact of a
product/package, the task of conducting an LCA can be difficult and time consuming. It
is often impractical for individuals and companies to obtain first hand all the pertinent
data needed for an LCA on a package design, and the effort required increases
proportionally when looking at multiple packaging options utilizing different materials

1

and processes. To overcome this difficulty several software systems have become
available with the intent of reducing the amount of effort needed to do an LCA, or some
subset of a full LCA that the software creators deem appropriate for packaging
applications.
The advent of multiple LCA software systems, while offering the promise of
expedited LCA, also creates questions about their use. Foremost is the question of
consistency in results between software systems; are the results from different software
similar, or does the choice of software affect the results in some manner? As LCA is
becoming a key tool in package design, it is important to know if the selection of
software affects the analysis in ways that can impact the decision making process. This
research is intended to aid in determining if the choice of LCA software may affect
results, and thereby influence decisions based on these results.

2

2. PACKAGING GROWTH CONSIDERATIONS
Packaging is tied to environmental concerns by the immense amount of
resources required to provide packaging for the distribution of products throughout the
world, and by the makeup of the packaging materials. There are other ties between
packaging and the environment, a key function of a package can be to protect the
environment from the package contents, but it is generally the volume and content of
material going into packaging that creates the larger environmental concerns.
One of the driving factors for the volume of packaging material being used is the
size of the population being served. Figure 1 shows the estimated population of the
world since 1804 based on United States (U..S.) Census Bureau data (U.S. Census
2002). There were approximately one billion people in the world in 1804; 118 years later
the number had doubled to 2 billion in 1922; 52 years after that the population had
doubled again to 4 billion in 1974; and it is estimated the population will double again to
8 billion people by 2028. With the increase in population comes an increase in demand
for products to support the population. Unfortunately, the earth’s resources have not

Billions of people

increased, so as population grows there is increased strain on the environment.
9
8
7
6
5
4
3
2
1
0
1800 1825 1850 1875 1900 1925 1950 1975 2000 2025 2050
Year
Population (billions of people)
Figure 1: Estimate of world population from 1804 to 2028. Data
Source: U.S. Census Bureau (U.S. Census 2002).
3

Trends in U.S. population and waste generation over several decades provide an
example of how increasing population can impact the amount of packaging material
required. Figure 2 shows how U.S. population, annual municipal solid waste (MSW)
generation, and annual packaging waste generation have changed since 1960, based
on U.S. Census Bureau and U.S. Environmental Protection Agency (EPA) data (U.S.
Census 2012, 2013; EPA 2011). U.S. population increased 71% over the 50 year time
period shown, while annual MSW generation increased 184% and packaging waste
generation increased 176%. Clearly MSW waste generation and packaging waste
generation have increased with population in the U.S., and at rates significantly greater
than the rate of population increase. Please note that the waste material being
discussed here is the total material entering the waste stream; this includes

250

350

200

300

150

250

100

200

50

150

0
1960

1970

Millions of people

Millions of tons

compostable, recyclable, and non recyclable material.

100
2010

1980

1990
2000
Year
Total Generated Municipal Solid Waste (millions of tons)
Total Containers & Packaging In Waste Stream (millions of tons)
Population (millions of people)

Figure 2: Population and waste material generation in U.S. from 1960 to 2010.
Data Sources: U.S. Census Bureau and U.S. EPA (U.S. Census 2012,
2013; EPA 2011). For interpretation of the references to color in this and
all other figures, the reader is referred to the electronic version of this
dissertation.
4

Another factor that can influence the volume of packaging material being used is
demand for products from countries with emerging industrial economies. As incomes
increase, more consumers can afford items that were previously too expensive for them
to purchase. In these situations it is not just the increase in population driving increased
use of natural resources, there is also the pent-up demand of the existing population
that now expects to be satisfied.
New car sales in China is an excellent example of how pent-up demand that is
released by economic growth can affect product demand, and thereby increase stress
on available resources. China’s economic growth has resulted in the Chinese market for
new cars surpassing the U.S. market in sales in 2009 (Lin 2012), a position it is
expected to retain. With this growth in car sales go all the items needed to produce,
maintain, and operate the vehicles. These items in turn need some form of packaging to
enable them to reach the end user in acceptable condition.
An example of how economic growth impacts resource utilization can be seen in
the changes that have occurred in China’s consumption of oil. Figure 3 shows oil
consumption in China and the U.S. over 3 decades as reported by the U.S. Energy
Information Administration (EIA 2013). While China’s total daily oil usage in 2010 was
approximately half that of the U.S., China had a 429% increase in oil consumption over
the 30 year period from 1980 to 2010, compared to a 12.5% increase for the U.S.
Furthermore, while oil consumption in the U.S. may be stabilizing, the rate of increase
for China appears to be accelerating. This implies that increasing demand for oil and
other resources to fuel Chinese economic growth will cause more strain on the
environment in years to come.
5

Thousands of barrels per day

22000
20000
18000
16000
14000
12000
10000
8000
6000
4000
2000
0
1980

1985

1990

1995
Year

China

2000

2005

2010

U.S.

Figure 3: Oil consumption for China and the U.S. from 1980 to 2010.
Data Source: U.S. EIA (EIA 2013).
China is not the only economy expected to see significant sustained growth.
Table 1 provides a list of the top 20 emerging markets (Bloomberg Markets 2013)
combined with the 2010 estimates of their populations (UN ESA 2013). All of these
countries are expected to have gross domestic product (GDP) growth over the next 4
years of 15.6% to 45.9%. While China is clearly the biggest emerging market in terms of
both expected GDP growth and population, the other 19 countries identified in the table
represent another 1.25 billion people in markets expected to see significant GDP
growth. As these markets mature they will likely drive the demand for products, and
therefore packaging, even higher.

6

Table 1: Emerging markets
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
1
2

Country
China
South Korea
Thailand
Peru
Czech Republic
Malaysia
Turkey
Chile
Russia
Indonesia
Colombia
Poland
Namibia
Zambia
South Africa
Mexico
Brazil
Hungary
Morocco
Philippines

Percent GDP
Growth, 2013
1
to 2017
45.9
22.9
25.9
27.4
21.1
21.8
21.2
24.2
26.6
31.3
21.9
21.2
22.3
31.3
19.9
17.5
22.3
15.6
27.7
20.4

Population
2
Estimate, 2010
1,341,335,000
48,184,000
69,122,000
29,077,000
10,493,000
28,401,000
72,752,000
17,114,000
142,958,000
239,871,000
46,295,000
38,277,000
2,283,000
13,089,000
50,133,000
113,423,000
194,946,000
9,984,000
31,951,000
93,261,000

Data Source: Bloomberg (Bloomberg Markets 2013)
Data Source: United Nations, Department of Economic and Social Affairs
(UN ESA 2013)

7

3. SUSTAINABILITY and LIFE CYCLE ASSESSMENT
Given that increasing demand for products will drive the need for more product
packaging, and given that the earth’s resources are finite, it is clear that packaging must
evolve to a state where it is sustainable by the earth’s resources. What is meant by
saying something is “sustainable”? The EPA (2013) states that “Sustainability creates
and maintains the conditions under which humans and nature can exist in productive
harmony, that permit fulfilling the social, economic and other requirements of present
and future generations.” An alternate definition comes from the World Commission on
Environment

and

Development

(WCED

1987):

“Sustainable

development

is

development that meets the needs of the present without compromising the ability of
future generations to meet their own needs.” To put it simply, packaging must be able to
meet the needs of the present without compromising the ability to meet the needs of the
future.
There are several approaches to improving package sustainability: material
reduction, reuse, recycling, and composting, to name a few. Each approach has merit
and in some cases the approaches can be combined, but how can their effectiveness
be determined in a given situation? The environmental impact of different packaging
options must be quantified to provide a true understanding of the benefits and/or costs
associated with each option. Without such an understanding it is possible for design
changes to do more harm than good.
One of the simplest concepts in packaging sustainability is to use less material,
the rationale being that less packaging material means less environmental impact. Even
this simple concept has caveats; the overall environmental impact cannot always be

8

judged by the weight or volume of the packaging material alone. Different types of
material can have different environmental impacts, and the effects of reuse and
recycling need to be taken into account. For instance, a glass container intended to be
refilled may need to be stronger, and therefore heavier, than a single-use version, but
being used multiple times can result in a more environmentally friendly option. As an
example, while single-use beer bottles are common in the U.S., refillable glass bottles
are common in Canada. With an average return rate of 97%, Canadian brewers claim
the refillable bottle system is responsible for an annual reduction of 187,000 metric tons
in greenhouse gas emissions (CNW 2005).
Wever (2009) proposed that environmental sustainability of packaging is tied to
the transportation efficiency of the product. In essence, improving the cube utilization for
transporting a given product results in less packaging material and energy being
required per item shipped. This is another simple concept that has merit, but also has
the same caveats as using less material. Cube utilization as an indication of
sustainability does not by itself take into account that different materials can have
different environmental impacts, nor does it account for the effects of reuse and
recycling.
If the type of packaging material is not being changed, and reuse or recycling is
not affected, then the amount of material being used or the cube utilization obtained can
give an indication of a relative change of environmental impact, but not an absolute
magnitude of the impact. Furthermore, neither approach takes into account any change
in the amount of product damage that can occur due to the reduction of packaging. A
small increase in product damage can cause more environmental impact than what is

9

avoided by using less packaging. Material reduction and cube utilization improvement
are viable relative improvement indicators when used in proper context, but something
more is needed to address comparisons of different types of packaging materials and
processing methods.
Sand (2010) recommends incorporating sustainability as part of the packaging
value chain. In this context sustainability is looked at in terms of added value provided
to customers at each stage in the packaging cycle. While this may be a useful business
model for evaluating different packaging alternatives, it does not provide the means of
obtaining the necessary information on sustainability. Sand does note that decisions
regarding sustainability require environmental impact information from raw materials
through to disposal, a reflection of the cradle-to-grave philosophy that has become a
cornerstone of sustainability.
The idea of estimating the environmental impact of a product or package by
looking at its entire life cycle, from obtaining raw materials (the “cradle”), through
manufacturing, distribution, end use, and disposal (the “grave”) has been around at
least since the 1960s (ALCAS 2013). Harry E. Teasley, Jr. is credited with conceiving
the concept of doing a study covering the entire life cycle of a package when he worked
for The Coca-Cola Company managing their packaging function in 1969 (Hunt and
Franklin 1996). At the time the company was looking at issues related to packaging
beverages that included whether to self-manufacture cans, use plastic bottles (new at
the time), or use refillable bottles. Teasley envisioned a study that would quantify the
energy, material, and environmental impacts of the very different container options
being considered. The study was carried out, becoming the first life cycle assessment of

10

packaging on record, though at the time it was referred to as a Resource and
Environmental Profile Analysis (REPA). A similar resource and environmental release
quantification method practiced in Europe was called Ecobalance (EPA 2006).
According to Hunt and Franklin (1996), it was not until 1990 that the term Life Cycle
Assessment (LCA) was first used in the U.S.
The Coca-Cola study was not published in full due to the proprietary nature of its
contents, but was used by the company to aid packaging and business decisions in the
early 1970s. One aspect to come out of the study was that the poor environmental
image plastics had at the time was not supported by the findings; thus this first study
showed the power of LCA to dispute an accepted notion. It is interesting that even today
plastics often have a bad environmental reputation that may not be justified in a given
situation. A recent occurrence of this can be found in the carry bag debate where local
governments are banning the use of plastic bags (Lin 2010; Karp 2011). One of the
arguments often used to support such a ban is that plastic is less environmentally
friendly than paper, yet LCA shows the opposite (Lewis et al 2010).
Another study setting ground work for what would later become LCA was done
in the United Kingdom in 1972 by Ian Boustead (EEA 1996). Boustead’s work involved
estimating the total energy used in production of glass, plastic, steel, and aluminum
beverage containers. In this early work the emphasis was on the single issue of energy;
looking at wastes and other emissions was not a priority. Over the years other
environmental concerns besides energy have come to the spotlight. As each new
concern came to prominence, the concept of estimating a product’s contribution to

11

environmental impacts by looking at the entire life cycle of the product found growing
acceptance.
One reason LCA has grown in use is that it offers more than just knowing the
total contribution, or impact, a product has regarding an environmental issue. The ability
of LCA to show where in a product life cycle environmental contributions occur is of
significant value, too. Armed with this information, efforts to reduce environmental
impacts can be concentrated in the areas that will yield the greatest benefit. LCA can
also be used to identify burden shifting, which occurs when a design change reduces
impact in one stage of the product’s life cycle at the expense of increasing it as much, or
more, in another stage. Thus LCA can provide both a guide for improving the
environmental impact of a product and a check on the results. The practice of reviewing
a product’s life cycle for ways to optimize overall environmental performance has come
to be referred to as Life Cycle Thinking, or LCT (EC-JRC 2013).
The systematic approach provided by LCA for estimating environmental impacts
has been found useful in supporting informed decision making across a broad range of
products and issues. Whether it is looking at the production of biodiesel (GonzalezGarcia et al. 2013), determining the carbon footprint for honey production (Kendall et al
2013), or debating the environmental benefit of buying locally produced food (Saunders
et al. 2006), LCA has become a method of choice for environmental sustainability
analysis. The use of LCA is expanding in the packaging industry as well. With the focus
on reducing the environmental impacts of packaging in today’s world, LCA is being
turned to increasingly as a support tool to aid in evaluation of packaging designs
(Sonneveld 2000).

12

4. PUBLICATIONS ON LIFE CYCLE ASSESSMENT
In 1997 the International Organization for Standards (ISO) released the first
version of ISO 14040 in an attempt to bring consistency to the methodology of
conducting an LCA. An updated version of ISO 14040, in combination with the more
recent ISO 14044, provides the general framework for conducting an LCA today (ISO
2006). While these standards provide the basis for doing an LCA, there are many
publications available that seek to explain, expand on, or aid in the use of these
standards. Such publications can be useful in understanding the fundamentals of LCA;
however, they can quickly seem redundant since they are often covering very similar
material. This sense of redundancy can be viewed as an indication of the success ISO
14040 and 14044 have had in standardizing the approach to LCA.

4.1 Introductions To Life Cycle Assessment
A useful introduction to LCA can be found in Environmental Life Cycle Analysis
(Ciambrone, 1997). The book is easy to read and relatively short, allowing it to hold a
reader’s attention from start to end. Readers gain insight into the basic principles of
inventory analysis (gathering data), impact analysis (translating data into environmental
impact categories), and improvement analysis (how LCA results can be used to improve
the product). Setting boundary conditions to define what will be included and excluded
from analysis, developing flow diagrams and process diagrams to identify the data that
will be needed, and how the quality of the data can affect the LCA results are all
covered. While a useful starting point in obtaining a general understanding of LCA, this
book is limited on application examples and may not be suited to someone looking for

13

detailed information on a specific topic. As a side note, the author did try to coin the
term “MANPRINT” to refer to the impact of a product on the environment, something
which thankfully did not catch on.
Another publication that is a good starting point for learning about LCA is Life
Cycle Assessment: Principles and Practice (SAIC 2006). Created by an EPA funded
research project, it is now available as an EPA document. Again this is a short
document, 80 pages total, that gives a straightforward presentation of the basic tenets
of LCA. This publication is newer than the first book mentioned, allowing it to reference
more recent information covering LCA development. It does share a similar limitation of
application examples, so it too may not be suitable for anyone looking for something
beyond an introduction to LCA. One advantage of this publication is that it is available
for free on line from the EPA website.

4.2 Life Cycle Assessment Procedural References
For a general reference on LCA one choice is the Handbook on Life Cycle
Assessment: An Operational Guide to the ISO Standards, commissioned by the Dutch
Government Ministry of Housing (Guinée 2002). The book provides a short discussion
of the background and purpose of LCA, a guide to creating an LCA, an operational
annex that provides support information, and a section on the scientific background
supporting LCA. It is in the last two sections that this book extends beyond being just an
introduction to LCA. Supplying practical support information regarding issues that come
up in conducting an LCA, along with background information on the science behind
LCA, makes this book a useful application reference. Of particular note is the example

14

provided for creating a PEDIGREE MATRIX to aid in determining the quality of data
being used in an LCA. The other publications mentioned so far talk about the
importance of data quality, but do not discuss in any detail a method for determining it.
An alternative reference for LCA practitioners is International Reference Life
Cycle Data System (ILCD) Handbook – General Guide for Life Cycle Assessment –
Detailed Guidance (EC JRC-IES 2010). This publication provides step by step guidance
in understanding how to perform an LCA. Its purpose is to bring consistency to the
practical implementation of LCA methodology based on accepted best practices.
Recommendations are provided for conducting an LCA from initial planning all the way
through to reporting results. The annex (appendix) on data quality is one of many useful
features. Another good feature is the discussion of frequent errors that have occurred in
doing the various steps of an LCA, an attempt to help people learn from the mistakes
made by others in the past. The guidance provided, while detailed, is not focused on
any specific product or service; it is a general reference with concepts that are easily
adaptable. One thing this publication is not intended to be is an introduction to LCA,
people should first read one of the introductory publications already mentioned before
attempting to use this one.
People who are interested in the efficient performance of LCA calculations may
want to read The Computational Structure of Life Cycle Assessment (Heijungs and Suh
2002). The authors go into great detail about performing LCA calculations using matrix
arithmetic. Solutions to problems such as how to handle cutoff of flows or closed loop
recycling with matrix arithmetic are covered. While not for everyone, this book is worth
considering if a spreadsheet, or similar program, is being used to do LCA calculations.

15

Spreadsheets generally have standard matrix arithmetic functions built in, and the
savings in computation time over a more linear calculation approach are reported to be
significant. When using software programs specifically developed to assist in LCA, such
as the ones looked at in this paper, the LCA calculations are performed by the software.
For an overall picture of how LCA and sustainability can affect packaging
decisions, a useful reference is Packaging for Sustainability (Verghese et al. 2012).
With only a single chapter on LCA, one that is geared toward an executive overview,
this book is not about how to do an LCA. Instead, a much broader look is taken at how
sustainability impacts business decisions, and how LCA is a tool that provides
information that can influence these decisions. Emphasis is put on companies needing
to include sustainability as part of their corporate strategy. Marketing, regulatory
compliance, consumer perceptions, and infrastructure availability are all shown to play a
part in how a company includes sustainability in its thinking. Among the many subjects
discussed are that consumers do not necessarily support sustainability with their
purchasing habits, and that a recyclable package will only get recycled if there is
infrastructure in place to handle the task. A section on packaging materials is included
with information covering representative life cycles, general environmental impact
considerations, and recovery and disposal options. This is a worthwhile book to read for
anyone involved in packaging sustainability, which should be anyone involved in
packaging design and decision making.

16

4.3 Impact Methodologies
One area where there are many differing publications is in impact assessment
methodologies. ISO 14040 and 14044 provide a framework for doing LCA, but
purposely leave out specifics on identifying impact categories. It has been left to the
LCA practitioner to identify what impacts are significant to the study being done, and
how best to map inputs and emissions to these impact categories. In order to help fill
this void, work groups of scientists and LCA practitioners have created various impact
methodologies for many common environmental issues. Individual methods are based
on available scientific information and best practices, and are peer reviewed. However,
this does not mean that two methods will agree. In order for LCA practitioners to
understand how an impact assessment methodology works, papers are published
describing the impact categories provided, the conversion of inputs and emissions to
these categories, and the scientific basis behind it all. While there are many different
assessment methods available, the three that are pertinent to this work are Impact
2002+, ReCiPe, and TRACI 2.
The Impact 2002+ assessment method is described in IMPACT 2002+: A New
Life Cycle Impact Assessment Methodology (Jolliet et al. 2003). Impact 2002+ converts
inputs and emissions into 15 midpoint environmental impact categories, and then maps
combinations of midpoint indicators into 4 damage (endpoint) categories. The 4 damage
categories are Human Health, Ecosystem Quality, Climate Change, and Resources
(mineral and non-renewable energy use). Some midpoint categories map into multiple
damage categories. Each midpoint category is expressed in terms of a reference
substance. For instance, an emission that affects global warming is converted into the

17

amount of CO2 emissions to air that produces an equivalent impact. All the inputs and
emissions that affect a midpoint impact category can be converted to an equivalent
amount of the reference substance for that category, and then added together to
determine the midpoint indicator value. For each damage category, midpoint values that
map into the category are multiplied by damage characterization factors to convert them
into a common unit, allowing the midpoint values to be summed into a single value
representing the impact of the damage category. In an attempt to further aid analysis of
the results, normalization factors are provided that convert damage category values into
ratios referenced to the emissions associated with one person for one year, based on
Western Europe.
An interesting side note about Impact 2002+ is that while the authors recommend
considering the 4 damage categories separately, which is in line with ISO 14044, they
also state the categories can be aggregated into a single result with proper weighting
factors. This is a departure from ISO 14044 which states “It should be recognized that
there is no scientific basis for reducing LCA results to a single overall score or number”
(ISO 2006). While not supported by ISO 14044 due to the lack of scientific basis,
coming up with a single LCA score is in essence what companies are doing when they
make decisions based on more than one environmental impact category. Whether it is
applying a formal set of weighting factors, or informally deciding that one impact
category is more important than another, companies implicitly aggregate category
results when setting the course they will follow for a product or service.
The ReCiPe assessment method is covered in ReCiPe 2008: A life cycle
assessment method which comprises harmonized category indicators at the midpoint

18

and the endpoint level, First edition (revised), Report I: Characterization (Goedkoop et
al. 2013). ReCiPe has 18 environmental midpoint categories and 3 endpoint categories,
most of which are similar to those found in Impact 2002+, but that does not mean they
are identical. While using more than one assessment method to verify trends is
advisable when performing an LCA, variation in the methodologies used to determine
indicators can be a source of differences in the values obtained from multiple methods.
Another reason for using multiple assessment methods is that one method may not
cover all the environmental impact categories of interest for a given project. One
midpoint indicator pertinent to this paper that ReCiPe supplies, and for which there is no
counterpart in Impact 2002+, is water depletion (usage).
One feature of ReCiPe is that three different scenarios are provided for grouping
assumptions that can affect the analysis of data. Each scenario provides a distinct
perspective on time-frame and other issues that contribute to uncertainty in the LCA
results. The individualist (I) perspective is based on short-term interest, where impacts
are not being disputed. The hierarchist (H) perspective assumes that the most common
policy principles are followed, and is viewed as the default model to use when a specific
perspective has not been chosen. The egalitarian (E) perspective looks at the longest
time-frame and considers impacts that may not be fully understood yet, but that do have
some basis for characterization. Clearly the egalitarian perspective is likely to have the
greatest uncertainty in analysis results, while the individualist perspective provides a
limited time over which any analysis can be considered valid. The hierarchist
perspective falls somewhere in between, trying to strike a balance between uncertainty,
available knowledge, and useful time-frame. While these different scenarios add some

19

flexibility to using ReCiPe, one must be careful to ensure comparisons are done using a
common scenario.
One interesting part of the discussion of architecture for ReCiPe deals with the
variations that can occur on a regional basis when trying to quantify potential damage
from emissions. Hygienic conditions, weather conditions, background concentrations,
and population density are all examples of issues with significant regional variability that
can influence the effect of emissions on the environment. ReCiPe often uses
environmental models based on Europe, and although attempts have been made to
generalize these models within ReCiPe, the authors do note that “the ReCiPe method
has limited validity for all regions that cannot be defined as well-developed temperate
regions” (Goedkoop et al. 2013). This is an excellent cautionary note for any
assessment method; the validity of the results are only as good as the assessment
model being used, which in turn is dependent upon the appropriateness of the
information the model is based on.
The TRACI 2 assessment method is an updated version of the original TRACI
method developed by the EPA in 2002, with TRACI 2.0 released in 2011 and TRACI 2.1
released in 2012. The TRACI methodology is discussed in the paper TRACI: The Tool
for the Reduction and Assessment of Chemical and Other Environmental Impacts (Bare
et al. 2002). TRACI differs from Impact 2002+ and ReCiPe in that it was developed
specifically for use in the U.S., coming after the conclusion was reached that no tool in
existence at that time was sufficiently applicable to the U.S. to meet the EPA’s needs.
TRACI also differs in that it deals only with category midpoints, leaving it to the user to
convert the midpoint information into desired endpoints. Information provided in the

20

referenced paper does show examples of how different endpoints can be affected by
the impact category midpoints provided. The updated TRACI 2 methodology is covered
in the paper TRACI 2.0: the tool for the reduction and assessment of chemical and other
environmental impacts 2.0 (Bare 2011). TRACI 2 covers 9 impact categories, with 3 of
these (Particulate, Cancer, and Non Cancer) being sub categories listed under Human
Health. The user’s manual for TRACI 2.1 (Bare 2012) does list Fossil Fuel Use as a
tenth impact category; however, the implementations of TRACI 2 found in the
commercial LCA software used for this paper did not include this category.

4.4 Journals and Other Publications
The publications discussed so far are only a small sample of what is available.
Books, papers, and journal articles on LCA abound. Several articles from the
International Journal of Life Cycle Assessment are referenced in this paper, as are
articles from the journal Packaging Technology and Science. Another source of LCA
information is the Journal of Industrial Ecology, which has recently tightened its policy
on what is expected in an LCA paper submitted for publication (Lifset 2013). This has
been done because the editors believe LCA has evolved to the point where there is now
widespread use and that basic case studies are now common. For a paper to be
accepted by the journal now, it must advance the practice of LCA by providing new and
important data that challenges established views, use a novel methodology, or provide
new information on a topic of significance. This change in policy reflects the success
that has been achieved in the development of LCA under ISO 14040 and 14044.

21

A general discussion of LCA publications would not be complete without
mentioning the ongoing publication of research papers that contribute to debates on
how best to model various substance flows within an LCA. One example is the paper
Critical aspects in the life cycle assessment (LCA) of bio-based materials – Reviewing
methodologies and deriving recommendations (Pawelzik et al. 2013). The authors of
this paper are concerned that current LCA standards do not adequately address
biogenic carbon storage associated with bio-based materials, and that failing to do so
can have a significant effect on estimating greenhouse gas emissions for products using
these materials. Papers like this are intended to widen the scope of thought on an
environmental impact issue, in this case the effect of carbon sequestration when
accounting for greenhouse gas emissions, and over time can result in changes being
made to what is considered accepted best practices for performing an LCA. The
research presented in these papers is what drives the continued evolution of LCA.

22

5. REVIEW OF BASIC LIFE CYCLE ASSESSMENT PRINCIPLES
LCA is a method by which environmental impacts can be identified, quantified,
and evaluated to provide information for guiding decision making about a product,
process, or service. The information from an LCA may be used for various purposes
including comparison of design options, product improvement, monitoring environmental
regulatory compliance, and providing a basis for product environmental claims. An LCA
will not provide information on parameters such as production costs, product
performance ratings, marketing considerations, and consumer acceptance, to name a
few. Thus LCA is one of many sources that feed information into the decision making
process for businesses today, and the information niche it fills is that of aiding in
understanding the environmental consequences associated with a decision.
An LCA is divided into four phases: goal definition and scope, inventory analysis,
impact assessment, and interpretation. While these phases are laid out to be performed
sequentially, LCAs often become iterative in nature, with information from a later phase
getting fed back to reevaluate the work done in an earlier phase. For example, it may be
that data from the inventory phase is not as specific as originally desired and the goal or
scope of the LCA has to be modified accordingly. The interpretation phase may reveal
more questions that need to be answered, requiring additional inventory data and
impact assessment. Newer or more product specific data may become available that
replaces older or estimated data, requiring another pass through the LCA to see if
results are affected. In essence, an LCA does not have a flat one dimensional flow;
instead the flow through an LCA is dynamic in nature, with review and updates
occurring for the different phases as new information becomes available.

23

5.1 Goal Definition and Scope
Adequately defining the goal and scope of an LCA is key to having an efficient
LCA process and a useful result. The goals of the LCA determine what information is
needed to fulfill the purpose of the LCA, the information needed drives the type of
assessments that have to be made, which in turn defines the type and quality of the
data that must be acquired. Having clearly defined goals serves to focus effort in
subsequent LCA phases onto what is needed, minimizing wasted effort.
The first step in performing an LCA is to understand who the LCA is being
performed for and how they are going to use the information that is provided. An LCA
can be initiated by various parties for different reasons; one may want information to
support a public policy decision, another may need guidance on selection of the best
option for a new product, someone else may want to know how an existing product
compares to others for a specific type of environmental impact. Trying to do everything
for everyone can result in an overly complex, time consuming, and expensive LCA.
Recognizing what information is actually needed by the stakeholders who initiated the
LCA, how specific that information has to be to meet their needs, their expectations on
the timeliness of the information, and how best to report the information to them can
significantly streamline the LCA process.
After the goals of the LCA are sufficiently defined to insure the information
expectations of the stakeholders will be met, the functional unit and system boundaries
the LCA will be based on must be set. The functional unit is simply a quantifiable
amount of a product, service, or process that can serve as a basis of reference for

24

comparing the environmental impacts of different options. For example, laundry
detergent can be powder or liquid, so kilograms of powder detergent or liters of liquid
detergent are not good units to use for evaluating these options since they cannot be
directly compared to each other. Even when comparing just powder detergents, weight
would not be an appropriate functional unit if the detergents being considered require
differing amounts to be used. A better choice for a functional unit would be to use a
fixed number of loads of laundry. This can then be related back to the amount of
detergent, powder or liquid, needed to do the loads of laundry. Once an appropriate
functional unit has been selected, then the environmental impacts of the different
options being considered can be scaled to reflect the amount of impacts associated with
producing one functional unit, thus allowing comparison of dissimilar options.
System boundaries define the limits of what will be included in an LCA. An LCA
can be an all inclusive cradle-to-grave study, extending from the acquisition of raw
materials for a product all the way through to disposal at end-of-life, or it can be the
“gate-to-gate” contribution of one or more processes in producing a product. The needs
of the stakeholders are what drive the setting of system boundaries. If someone wants
to know the total environmental impact of a product, say a glass bottle filled with beer,
then a cradle-to-grave LCA is called for. On the other hand, if the intent of the LCA is to
help select a cleaning method for the bottles before filling them with beer, then it is the
contribution of the cleaning process options that need to be looked at, not the entire life
cycle of a bottle of beer.
System boundaries may have to be expanded to account for production of coproducts. When dealing with multiple products produced from a common process, it is

25

necessary to fairly allocate the inputs and emissions associated with the process among
the resulting co-products. Since allocation based on characteristics such as weight,
volume, or monetary value of the co-products do not necessarily reflect the true
distribution of environmental burdens, it is preferable to find a way to separate the coproducts. One way to achieve this separation is to subdivide the common process into
sub-processes associated with the various co-products. Where this is not practical, the
principle of avoided burden can be used to separate the environmental burdens of a coproduct from the product of interest. To do this, the co-product is viewed as reducing the
need for the same item generated by more direct means, so the system boundary is
expanded to include production of an equivalent amount of the undesired co-product,
and the environmental burdens associated with this generation are subtracted from the
process that produces the co-products. Ideally this leaves only the burdens associated
with the product of interest; however, care must be taken when applying this method.
Any type of emission associated with equivalent production of an undesired co-product
that does not have an offsetting emission in the original multi-product process will result
in a negative emission being reported.
Assumptions, rules, and methodologies guiding the implementation of the LCA,
such as the use of system expansion, should be stated as part of the scope of the LCA.
Anything that affects what data is collected, how the data is collected, and how the data
is analyzed should be included and updated as needed. Everyone involved with the
LCA, whether it be with implementation, reviewing, or using the results, must work from
a common reference point.

26

5.2 Life Cycle Inventory (LCI)
Performing an LCI amounts to collecting and quantifying all of the inputs and
emissions associated with the product, process, or service being studied. It is from this
inventory that the environmental burdens associated with the product, process, or
service will be determined. In its simplest form, an LCI produces a list of all the
substances involved with the production, use, and disposal of a product, along with the
amount required of each substance. In practice, a product is made from components,
which may have several levels of subcomponents, which are made from materials,
which are comprised of substances. Added to this is that each step of the way from
obtaining raw materials to disposal at end-of-life involves more energy and material
usage for transportation and processing, resulting in additional environmental burden.
Performing an LCI from cradle-to-grave can be a very daunting task; fortunately there
are ways to minimize the effort.
The starting point for an LCI is to generate a flow diagram of the processes within
the system being studied. You must know a system before you can perform an LCA on
it, and if you cannot diagram the system, then you do not know the system well enough.
The system boundaries are established when defining the goal and scope of the LCA
study. The flow diagram then builds on this by linking the input and emissions of the
system to the individual unit processes within the system, and by showing how the unit
processes interconnect to produce the product.
Once a system flow diagram is in place, the next step is to create a plan for
collecting data. What data will be needed and where the data will come from is driven
by the data requirements set down when defining the goal and scope of the LCA. Some

27

data may be generic in nature and found in commercially available databases such as
Ecoinvent (Ecoinvent Center 2013), or government sponsored databases such as the
U.S. Life Cycle Inventory (NREL 2013). Other data may be in the form of industry
averages for the region of interest and found in industry or government publications.
Product or process specific data may require that time and effort be allocated to field
surveys. Each of these sources has advantages and disadvantages. The generic
database approach is relatively easy and fast, but will lack specific information about the
product being studied, and the generic information may not be sufficiently timely.
Industry averages may be more specific and timely than generic information, but will
require more time to locate and evaluate. Product specific information is the most
relevant, but can require substantial effort to obtain.
When collecting data for an LCA, a practical approach to balancing cost versus
data quality involves using a combination of product specific, industry, and generic data.
Product specific data may be collected for key information that differentiates the
product, industry averages may be used for common processes, and generic data used
to fill in data gaps. One way to minimize the data collection effort in comparing multiple
alternatives for a product is to do an initial LCA based on generic data to look for the
most promising options, then gather industry and product specific data only for those
options. In this way the effort associated with data collection becomes more focused on
the better possibilities.
Another aspect of data collection that must be addressed is when to exclude
processes because they do not make a significant contribution to the environmental
burden. Ideally, an LCA includes all processes no matter how small their contribution,

28

but realistically it is not cost effective to spend valuable time collecting data for
something that has little bearing on the final results. Furthermore, as an LCA becomes
cluttered with insignificant processes, it may detract from the analysis of significant
contributions. If it is clear that a process does not make a significant contribution to the
LCA results, then it should be excluded; however, the reasons for excluding it should be
documented. If it is not clear that a process contribution is insignificant, then the process
should be included at least through a first pass of the LCA with generic data.

5.3 Life Cycle Impact Assessment (LCIA)
The LCIA phase of an LCA can be thought of as the processing phase, where
the raw inventory data collected during the LCI phase is converted into estimates of
environmental impact. There are many ways in which the environment can be affected:
global warming, ozone depletion, acidification, eutrophication, and nonrenewable
energy use, to name a few. Each of these environmental issues represents an impact
category that has been studied, and for which there is reasonable scientific basis to
establish a quantifiable causal relationship between human activities and the severity of
the impact to the environment. An LCIA employs impact assessment methodologies
consisting of agreed upon models of quantifiable relationships to convert the inputs and
emissions involved with producing a product, process, or service into indicators
representing the estimated levels of impact that will occur in the environmental
categories of concern.
Conversion

of

input

and

emission

substances

starts

with

applying

characterization factors, conversion values developed from scientific study, that relate

29

the effect a substance has on an impact category to the amount of a reference
substance required to produce an equivalent effect. Multiplying the input and emission
substances in the inventory by the appropriate characterization factors, and then
summing the results for an impact category, produces the impact indicator for that
category. Midpoint indicators generally represent impact categories directly related to
specific environmental mechanisms, such as global warming or ozone depletion.
Endpoint indicators often represent damage to systems, such as human health or
marine life, and may be based on multiple midpoint indicators, making the models the
endpoint calculations are based on more complex. This increase in complexity, and the
associated increase in assumptions and simplifications that have to be made, can
cause greater uncertainty in the results for endpoints versus midpoints (SAIC 2006).
Impact category indicators initially have units that are based on the reference
substance for each category. It may be desirable for comparison purposes to modify
this to something else. Normalization can be used to change the indicators to be a ratio
relative to a desired reference point. One reference point might be an original base line
LCA for a product, allowing subsequent LCAs for new versions to have their impact
indicators expressed as being more than, or less than, the original version. Another
common reference point is to express an indicator as relative to the amount of emission
one person generates in a specific area over a specific amount of time, such as the
amount associated with one European over one year of time. Using, or not using,
normalization depends on the requirements for analyzing and communicating the LCA
results back to the stakeholders in a manner they find useful, something that should be
determined when the goal and scope of the LCA is being defined.

30

Weighting is another means by which the results of an LCA can be adjusted to
aid interpretation or improve communication. Weighting is simply a formal way to apply
a ranking of importance to different impact categories. Each impact category indicator is
multiplied by a weight factor representing its relative importance, and then the weighted
results are summed together into a single value representing the total environmental
impact of the product, process, or service. Weighting is what a company effectively
does when it decides on one product option over another based on multiple impact
categories. The problem with weighting is that there is generally no scientific basis for
saying that one impact category is more important than another. This makes weighting
a judgment call, and something that is significantly more open to questioning than
category indicators. If weighting is used, it should be applied in a consistent manner and
the basis of the weighting must be documented.

5.4 Life Cycle Interpretation
Understanding the results of an LCA is not as simple as saying one option is
better than another. Even if the relative performance of options show a consistent
pattern favoring one option across all the impact categories being considered, there are
still questions about assumptions, uncertainties, and the significance of results that
must be considered. An LCA generally involves a set of assumptions, as well as limits
on the availability, timeliness, and applicability of data used, all of which must be
reviewed to ascertain what affect they may have on the results.
The starting point for life cycle interpretation is to identify where the greatest
contributions to impact categories are coming from. If a few processes account for the

31

majority of an impact in a category, then these are the processes that should be
focused on in the interpretation phase. For example, if one process contributes 90% of
the impact in a category, then any variations in this process caused by changes in
assumptions or issues with data quality will likely produce more pronounced changes in
the impact category result than variations in all the other processes combined. LCAs
generate a lot of data, so being able to focus effort on major areas of contribution is
important to an efficient and cost effective LCA. Identifying the major contributors to an
impact category also provides guidance on where future effort to reduce the impact
should be concentrated.
The next task is to review the LCI and LCIA results to verify they are consistent
with the scope of the study and that they are sufficient to allow the goals of the study to
be met. The quality of the data collected should be verified against the expectations set
forth at the beginning of the LCA. The extent to which LCI data measurement error
affects the inherent uncertainty associated with LCIA results should be checked, as well
as the sensitivity of LCIA results to small changes in data. Anomalies in the data and
results should be identified and explained. How well the collection and assessment of
data meets the needs of the study should be documented, along with the course of
action taken to deal with any deficiencies that were found. Each deficiency must be
addressed as to how it will affect the LCA, and whether more data collection and
analysis is warranted.
Once it is established that the LCI and LCIA results meet the needs of the LCA,
they should be reported along with any caveats, conclusions, and recommendations
required. The report should include all assumptions, limitations, methods, data, and an

32

adequate discussion of the analysis being done. Reporting the assumptions and
limitations of the LCA are just as important as reporting the results, as these define the
context under which the results can be viewed. The report must take into account the
expectations of the stakeholders in regard to the type of information expected, as
covered when defining the goals and scope of the LCA. While the needs of the
stakeholders may dictate a summary form of report be presented to them, this should
always be backed up by a complete report that is intended to go through a peer review
process.

33

6. LIFE CYCLE ASSESSMENT SOFTWARE
The complex and time consuming nature of LCA has prompted the development
of several software programs aimed at reducing the effort involved in performing an
LCA. While general purpose software such as MS Excel, Matlab, or Mathematica can
be set up to do LCA calculations, software specifically designed for LCA facilitates
assembling the LCI data as well as coming preconfigured to do LCIA calculations.
These programs simplify initial modeling of products by providing access to databases
of generic LCA material and process information. Having the product models then flow
as input to the assessment methodologies included with the software eliminates the
effort normally associated with entering and verifying an assessment method in a
spreadsheet or math program.
LCA software programs can be broad-based fully ISO 14040/14044 compliant
packages, or they can be simplified versions geared toward specific applications. The
broad-based packages allow access to multiple databases and impact assessment
methods, allowing them to be more general purpose. The simplified programs are
streamlined versions of LCA, often having only a single database and assessment
method available. Some LCA software is proprietary, others are commercially available.
The LCA software programs discussed in this paper are all commercially available.

6.1 SimaPro
SimaPro (PRé Consultants 2011) is a general purpose LCA software program
that is fully compliant with ISO 14040/14044, providing complete LCI and LCIA
capabilities. A product life cycle is modeled as a collection of assemblies, processes,

34

and waste or disposal scenarios. An assembly is made up of a collection of substances
(minerals, chemicals), materials, processes, and other assemblies (subassemblies).
Disposal scenarios can include reuse, disassembly, and waste scenarios. Multiple
libraries (databases) containing predefined substances, materials, processes, and
waste treatments for modeling products can be installed. Various impact assessment
methodologies can be installed, modified, or created. A high degree of flexibility and
visibility is provided in doing LCA calculations, with all impact characterization factors
accessible, and contributions to an impact category traceable down to a substance
level. This flexibility does make SimaPro more complex than the streamlined LCA
software offerings. SimaPro versions 7.2.4 and 7.3.3 were used for this paper.

6.2 GaBi
GaBi (PE International 2011) is another general purpose LCA software program
that is fully compliant with ISO 14040/14044, providing complete LCI and LCIA
capabilities. A product life cycle is modeled as a plan consisting of processes,
material/energy flows, and other plans. Having plans within plans provides a nesting
capability similar to the use of assemblies and subassemblies in SimaPro. Gabi can
provide access to multiple commercial databases of predefined LCA data, though they
have to be merged into a single database to be used at the same time. GaBi provides
access to multiple LCA impact assessment methods. Flexibility and visibility in doing
LCA calculations in GaBi is on par with SimaPro. All impact characterization factors are
visible and contributions to an impact category are traceable down to a substance level.

35

Like SimaPro, Gabi is significantly more complex to use than streamlined LCA software
offerings. GaBi version 5 was used for this paper.

6.3 openLCA
openLCA (openLCA 2011) is an attempt at providing a free open source software
package for doing LCA. The intent is a software package with LCA capabilities similar to
SimaPro and GaBi. Unfortunately it was found that openLCA was very much a work in
progress at the time it was reviewed. Documentation was poor and several problems
with using the program were encountered. One significant issue found was that the
sequential calculation method could not be made to work properly. Some analysis was
done using the alternative matrix calculation method. The ability to trace contributions to
an impact category down to a substance level was missing. Open LCA version 1.2 was
used for this paper.

6.4 COMPASS
COMPASS (SPC 2011) is a web-based streamlined LCA software program
designed for packaging applications. COMPASS does not provide true LCI capabilities;
instead a limited dataset of impact indicators previously agreed upon by the members of
the Sustainable Packaging Coalition (SPC) has been generated and made available to
users of COMPASS. No other datasets can be accessed through the user software, and
the methodology for generating the dataset is fixed. Thus multiple assessment methods
are not permitted. COMPASS is easier to use than either SimaPro or GaBi, but is also

36

far less flexible. It is not possible to trace contributions to an impact category down to a
substance level. COMPASS 2.0 was used for this paper.

6.5 Package Modeling
Package Modeling is the underlying software on which the WalMart Scorecard is
based. Like COMPASS, it is intended for packaging applications, does not provide true
LCI capabilities, and allows for only a single proprietary assessment method. It
generates a single weighted score based on 9 categories of information. The categories
include generalized indicators such as cube utilization and energy innovation, but
provide little or no information on most environmental emissions other than greenhouse
gases. There is little flexibility in application, and no traceability of impact contributions
back to substance level emissions. Package Modeling 3.0 was used for this paper.

37

7. RESEARCH SUBJECT
Every LCA is based on a combination of assumptions and data, so there is
always some room for variability in results. Examples available in literature confirm
LCAs using different sets of assumptions and data for a common topic can have
disparate conclusions (Villanueva and Wenzel 2007). What has not been well studied is
whether or not the selection of an LCA software program can contribute significantly to
differences in the conclusions that are reached.
To get an idea of the extent software specific to LCA has been used, a review of
three LCA and packaging industry related journals was conducted to identify the
software used for LCIA. Included in the review were all issues published in 2010
through 2012 of International Journal of Life Cycle Assessment, Journal of Industrial
Ecology, and Packaging Technology and Science.

Figure 4 shows the number of

articles each year that were found to use software specific to LCA for impact
assessment. A total of 73 articles over the 3 year period used SimaPro for LCIA, 21
articles used GaBi, and 5 articles use other LCA software. No articles were found to use
more than one program for LCIA. In addition, no articles were found that used
COMPASS, Package Modeling, or openLCA. When commercial LCA software was
used, SimaPro and GaBi were the preferred choices for LCA practitioners submitting
articles to these journals. It is also clear that it is common for LCA practitioners to use
only one software package when doing an LCA, likely selecting the one they are most
versed in.

38

Number of Articles

SimaPro

25
20
15
10
5
0

2010

2011

Other

GaBi

2012

2010

2011

2012

2010

International Journal of Life Cycle Assessment

2011

2012
Year

Journal of Industrial Ecology
Packaging Technology and Science
Figure 4: Journal articles using LCA specific software for LCIA.

By using only a single LCA software package in a study, an implicit assumption
is being made that the choice of software does not affect the outcome of the study. This
assumption is further compounded if, as is often the case, only one impact assessment
method is being used as well. There has been only limited investigation into whether
LCA software and associated impact methodology implementations significantly
influence results. Aissani et al. (2009) presented a comparison of SimaPro and GaBi
that was basically a discussion of the features available in the two software packages.
Lee and Shan (2000) compared SimaPro and GaBi to EcoPro, using corrugated
paperboard and different impact assessment methods, finding that EcoPro reported 50
times the greenhouse gas emissions and 80 times the eutrophication values of the other
programs.
The studies mentioned so far cover only SimaPro and GaBi; no good source of
comparison has been found for streamlined LCA software. There is no information
available as to how consistent results are between streamlined and fully ISO compliant
LCA software, nor is there information comparing just streamlined LCA software. As the
39

use of LCA software becomes increasingly common place, there is a need for reliable
information as to whether or not LCA software choices affect the results obtained, and
how conclusions drawn from those results may be affected. In addition, it would be
useful to develop guidelines as to when it is reasonable to employ streamlined LCA
software versus using fully ISO compliant software.
The research proposed for this paper is intended to help fill the gap in knowledge
that currently exists concerning comparability of LCA software. The basic presumption
made by companies using LCA software to guide decisions is that environmental impact
estimates will be sufficiently consistent between LCA software as long as assumptions,
system boundaries, and other conditions affecting the study remain consistent. To test
this presumption, a common set of packages and distribution systems was defined and
used as the basis for comparison of multiple LCA software programs.

40

8. RESEARCH METHODS
The main interest when considering the comparability of LCA software is whether
or not the conclusions reached about environmental preference of possible alternatives
in a product design will be consistent across the different software. Every software
program has the potential to influence the way environmental impacts are assessed.
The most obvious factors affecting impact assessments of product systems are
differences in data quality and impact assessment methodologies; however, other less
obvious factors can come into play, too. It may be that limits in available datasets
dictate substitutions, deletions, simplifications, or detailed expansions be made to a
product model in one software that would not be done in another. A more subtle
possibility is having the functionality of available modeling tools influence the
implementation of a model by making some things easier to do in one software than in
another. Another aspect is that streamlined LCA software gets that way by limiting
available choices, meaning that some decisions are pre-made in the attempt to simplify
operation, and details of those decisions may not be readily visible to the user.
To study the different LCA software programs, three sets of common packaging
systems were chosen as the basis for comparison. Each set consisted of two or more
containers that could be used for a similar purpose. Table 2 presents a summary of
these packaging systems. Beverage containers were selected for one set as they have
been widely studied and allow for a variety of materials to be considered (RomeroHernández et al. 2009; Mourad et al. 2008). This set included aluminum cans, glass
bottles, polyethylene terephthalate (PET) bottles, aseptic cartons, and polylactic acid
(PLA) bottles. The functional unit for comparison of this set is one liter of fluid

41

(beverage). The second set of containers explores flexible versus rigid packaging
options by comparing multilayer pouches and steel cans used in packaging tuna. This
sets utilizes one kilogram of tuna as the functional unit. The third set of containers is
aimed at studying the effect of reuse by comparing reusable polypropylene (PP) crates
to single-use corrugated boxes for distribution of cut flowers. Each PP crate and
corrugated box can carry an amount of flowers equivalent to ½ flower box, a standard
unit of measure in the cut flower industry and the functional unit for this comparison.
Table 2: Packaging systems compared
Functional
Unit

Containers Per
Functional Unit

Aluminum Carbonated Beverage Can, 12
fluid ounces (354.9mL)

1.00 lt fluid

2.82

Glass Beer Bottle, 12 fluid ounces
(354.9mL)

1.00 lt fluid

2.82

Polyethylene Terephthalate (PET)
Carbonated Beverage Bottle, 12 fluid
ounces (354.9mL)

1.00 lt fluid

2.82

Aseptic Carton, 200mL

1.00 lt fluid

5.00

Polylactic Acid (PLA) Water Bottle, 500mL

1.00 lt fluid

2.00

Multilayer Pouch, 74g net weight

1.00 kg tuna

13.51

Steel Can, 142g net weight

1.00 kg tuna

7.04

Corrugated Box, 1/2 flower box

1/2 flower
box

1.00

Polypropylene Crate (PP), 1/2 flower box

1/2 flower
box

1.00

Container Type

Flow diagrams representing simplified product life cycles were created for the
nine packaging systems selected as a basis of comparison for this study. Simplified life
cycles were used to facilitate comparison of the software systems; consequently the
information presented in this paper should not be construed as providing valid
42

comparisons of the actual packaging systems themselves. The assumptions made to
simplify these systems are listed in Appendix B. Data on the physical characteristics
needed to model the nine containers in the LCA software programs are given in
Appendix C. Some of this data has been taken from previous studies; the rest was
obtained by measuring samples of the different containers. The simplified flow diagrams
are presented in Appendix D. To keep the number of tests that needed to be conducted
down to a manageable level, it was decided to study the effects of varying transport
distances, recycling rates, recycled content, and number of reuses only in selected
comparisons. Appendix E provides details on how test parameters were varied to obtain
the comparison data of interest. U.S. datasets were used wherever possible in creating
software models of packaging systems.
Using the flow diagrams and associated data as common references, models of
the packaging systems were created in SimaPro 7.2.4, GaBi 5, COMPASS 2.0, and
Packaging Modeling 3.0. Problems encountered with using the version of openLCA that
was available when these tests were conducted led to it being dropped from this portion
of the testing.
It quickly became apparent that some software offered choices that could
significantly reduce the amount of work involved, while other software did not. For
example, SimaPro had available data files from Franklin Associates for aluminum,
glass, and PET containers that rolled the entire set of processes for the container, from
getting raw material in the ground to producing a finished container, into a single cradleto-gate file. Thus an LCA practitioner using SimaPro could do either a detailed model for
producing one of these containers where each required process is represented

43

individually, or they could use a single cradle-to-gate file. GaBi did not have these
cradle-to-gate files available, necessitating more detailed and time consuming models
be implemented. COMPASS was the opposite of GaBi, generally providing only one
appropriate conversion (manufacturing) process for each container, with no means for
the LCA practitioner to do their own detailed model. Package Modeling did not provide a
means of adding conversion processes. Faced with this inconsistency in the way
product system models could be implemented across the different LCA software, it was
decided to allow the models to be implemented in the manner that provided the simplest
approach for each software. Clearly this can cause differences in inventory data that
may affect the results, but it does create valid models for each software. Furthermore, it
is believed that the simplest method of implementing a model reflects the most likely
approach users will take in using a software program.
The lack of comparable impact categories is an obvious difficulty when
evaluating LCA software. The impact assessment methodology used for SimaPro and
GaBi in this part of the study was Impact 2002+ version 2.1, supplemented by ReCiPe
Midpoint (H) version 1.04 in SimaPro and version 1.05 in GaBi for water consumption.
SimaPro and GaBi have 15 midpoint categories using Impact 2002+, and 18 with
ReCiPe, but only global warming, non-renewable energy, aquatic eutrophication, and
water depletion can be matched to output indicators in COMPASS. Once Package
Modeling is added, global warming becomes the only category of impact information
common to these 4 software programs. Things are further complicated when trying to
include openLCA. Its implementation of Impact 2002+ reports only 2 of the standard 15
midpoint categories, neither of which is global warming. The other impacts are

44

translated directly to damage (endpoint) categories and reported in “points”. Among the
other software being studied, only SimaPro and GaBi have the option of viewing impact
categories in units of “points”, a technique for normalizing results in terms of the
average impact to one person in one year for a specified region (Humbert 2005). Thus
openLCA results using Impact 2002+ could not be compared directly to COMPASS or
Package Modeling. Both COMPASS and Package Modeling report impact categories
that have no comparable counterpart in the other software, examples are a “Material
Health” score reported in COMPASS and the “Innovation Different from Energy
Standard” metric used in Package Modeling. With the wide variation in the type and
presentation of impact information available, the potential exists for software to
influence decision making not only by the impact category results presented, but by
which categories are included. Evaluations of this type of influence would be subjective
and are not within the scope of this paper. However, the lack of common impact
categories does create limitations for this study that must be addressed. All
comparisons in this paper will be based on the values of an impact category indicator
common to the software being included, meaning that some evaluations will involve only
a subset of the software programs being studied, the others being eliminated from
specific comparisons due to the absence of the impact category under consideration.
Using packaging system flow diagrams as the starting point, while allowing for a
useful broad comparison, does not provide the means to easily track down the cause of
differences that may occur. Any variation in impact assessment can be the result of an
amalgamation of the different data, assumptions, and modeling choices dictated by, or
at least influenced by, the functionality of the software being compared. For example,

45

SimaPro and GaBi can use commercially available datasets of LCA information while
streamlined LCA software such as COMPASS does not. What COMPASS makes
available to users is a web based set of impact indicators that were previously compiled
using a proprietary assessment method and SimaPro. An attempt to duplicate the
process outlined for COMPASS, using the input datasets identified for COMPASS along
with information about the proprietary assessment method, resulted in significant
differences in impact values compared to those reported by COMPASS. Without access
to the actual inventory database used to produce the impact indicator dataset for
COMPASS, it was not possible to definitively determine the causes of any differences
involving COMPASS. Attempts to trace differences in impact results between SimaPro
and GaBi for the packaging systems also proved difficult.
As a means of providing a base line of comparison, one that would aid the effort
in exploring causes of any impact differences that might be encountered, it was decided
to do another series of comparisons using more minimalistic systems to cut down on the
number of variables involved. Each system would consist of obtaining and disposing of
1 kg of a single basic packaging material. Four materials were selected: aluminum,
corrugated board, glass, and PET. Table 3 shows the parameters used with these basic
material test scenarios. SimaPro 7.2.4, GaBi 5, openLCA 1.2, and COMPASS 2.0 were
used for the basic material comparisons. Packaging Modeling was not included as it is
designed specifically for packages, not for obtaining and disposing of materials.
Values for recycling rate, landfill, and waste-to-energy at end-of-life (EOL) for the
base materials were chosen to match those used in COMPASS, as COMPASS did not
allow modification of these parameters. Input data for openLCA, SimaPro and GaBi was

46

limited to files from the Ecoinvent 2.2 database, with efforts made to identify and use the
same data files called out in reference documentation as being used in creating the
COMPASS dataset (Mistry 2010). Because many U.S. datasets are adapted from
European datasets by adjusting energy inputs to match U.S. energy production
averages, European datasets originating in the EcoInvent 2.2 database were chosen for
the base materials to eliminate potential variations that might have occurred in doing the
energy conversions. The impact assessment methodology used with SimaPro and GaBi
for the basic material tests was Impact 2002+ version 2.1, supplemented by ReCiPe
Midpoint (H) version 1.04 in SimaPro and version 1.05 in GaBi for water consumption.
The methodologies used with openLCA were Impact 2002+ and ReCiPe; however, no
information was found that identified the versions being used.

Table 3: Parameters for basic material test scenarios

Basic Material
Aluminum
Corrugated Board
Corrugated Board
Corrugated Board
Packaging Glass
Polyethylene
Terephthalate
(PET)

Functional
Unit
(kg)

Recycled Recycled
Content At End of
(%)
Life (%)

Landfill
At End of
Life (%)

Waste To
Energy At End
of Life (%)

1
1
1
1
1

50
12
50
87
55.5

67.9
76.4
76.4
76.4
63.8

32.1
10.7
10.7
10.7
36.2

0
12.9
12.9
12.9
0

1

0

26.1

45.1

28.8

The final part of this study involved tracing any discrepancies in results between
software to their root causes. SimaPro and GaBi lend themselves well to examining
differences in impact assessment values coming from the basic material tests. Both
47

software programs provide complete transparency in all calculations, and in the
characterization factors used to implement impact assessment methodologies.
COMPASS and openLCA were less transparent, which can make tracing impact
contributions back to input sources difficult, if not impossible. For SimaPro and GaBi,
built-in functions allow all the substances contributing to an impact assessment category
indicator value to be viewed, as well as the characterization factors used to implement
the assessment method.
To trace how differences in impact assessment values occur, the approach used
was to start with the LCIA output and work backward to the inputs. While there can
potentially be hundreds, or even thousands, of substances contributing to an impact
category, generally only a small number are significant. By sorting on total contribution
to an impact category, it is easy to see which substances contribute the most to the
category. To minimize time, it was originally decided to look at the substances that
combined for the first 90% of an impact indicator value, but it was often found that the
number of significant contributors was small enough to make it practical to look at
substances responsible for 95% or more of the total contributions. Once the significant
contributors were identified, the amount of each input substance could be found, as well
as the characterization factor used to convert the input substance to the common
reference substance of the impact category. With the significant inputs, input amounts,
characterization factors, and impact category contributions known, all the pertinent
information was available for pointing to the cause of any important discrepancy in
impact category indicator values between software programs.

48

Due to an upgrade, SimaPro 7.3.3 was used for this last part of the study, with
the impact assessment methodologies for SimaPro and GaBi being Impact 2002+
version 2.1, ReCiPe Midpoint (H) version 1.06 in SimaPro and version 1.05 in GaBi,
plus TRACI 2. TRACI 2 methodology was added for both SimaPro and GaBi as a third
point of reference when looking for inconsistencies in results.

49

9. PACKAGE SYSTEM COMPARISONS
The stated goal of the packaging system comparisons was to evaluate whether
the choice of LCA software was likely to influence decisions based on the LCA results.
Since some variation in magnitude of impact indicators was likely due to differences in
datasets and modeling, tests were created that looked at the relative order each
software would rank a group of similar containers, as well as provide a comparison of
impact indicator magnitudes across software. While the life cycle scenarios of
containers were varied between tests to study different facets of how impact
assessment was affected, the key point for data presented in this paper is that every
software program was presented the same set of container scenarios for a given
comparison.
A note of caution is in order regarding impact category results presented here; no
representation is being made as to what constitutes an environmentally significant
amount of emissions or energy use for any impact category. Determining if the value of
an impact category indicator represents a significant environmental impact for a given
situation is beyond the scope of this paper. What is being investigated here is the
consistency of results between LCA software programs.
The data for all graphs presented in this section can be found in Appendix G.

9.1 Beverage Containers
The parameters used in the beverage container comparison presented here are
given in Table 4. This represents a small subset of the comparisons conducted for this
study; a more complete listing of test parameter combinations explored is given in

50

Appendix E. All beverage container comparison data is based on a functional unit of 1
liter of fluid (beverage).
Table 4: Parameters for beverage container comparison
Container Type

Recycled
Content
(%)

Aluminum can
PET bottle
Glass bottle
Aseptic carton
PLA Bottle
1

70
10
25
0
0

Recycled
1
At EOL
(%)
50
30
40
0
0

Composted
1
At EOL
(%)
0
0
0
0
0

Truck
(km)

Rail
(km)

Air
(km)

Ship
(km)

100
100
100
100
100

0
0
500
0
0

0
0
0
0
0

0
0
0
0
0

EOL = End of Life.
The results obtained for the global warming impact category are shown in Figure

5. While the relative ranking order (lowest to highest) of the beverage containers was
the same in COMPASS and SimaPro, the rankings of the aluminum can versus PLA
bottle differed in GaBi. Further, the results for the beverage containers did not maintain
the same ratios to each other when moving from COMPASS to SimaPro, indicating that
changes in weights of the containers being compared could affect the relative rankings
differently in each software. Package Modeling differed substantially from the other
software. While COMPASS, SimaPro, and GaBi all ranked the aluminum can, glass
bottle, and PET bottle in that order relative to each other for lowest to highest
emissions, Package Modeling had the aluminum can and glass bottle approximately
equal, with both at twice the emissions of the PET bottle. Package Modeling had the
emissions for the aluminum can being greater than any of the other software, and the
emissions for the PET and glass bottles significantly less than the other software.

51

0.7
kg CO2 equiv

0.6

0.5
0.4
0.3

0.2
0.1
0.0
COMPASS
SimaPro
GaBi
Package Modeling
Al can
PET bottle
Glass bottle
Aseptic carton
PLA bottle
Figure 5: Greenhouse gas emissions for selected beverage containers.

MJ Primary Energy

12
10
8
6
4
2
0

COMPASS
Al can
PET bottle

SimaPro
Glass bottle

GaBi
Aseptic carton
PLA bottle

Figure 6: Fossil fuel/non-renewable energy for selected beverage containers.
Figure 6 shows the comparison results for fossil fuel/non-renewable energy use.
The COMPASS fossil fuel category did not include electricity generated from nuclear
energy, part of the standard U.S energy mix, while the Impact 2002+ non-renewable
energy category used in SimaPro and Gabi did include it. Despite this, COMPASS fossil
fuel results were 6% higher for PET bottles and 18% higher for glass bottles than the
corresponding non-renewable energy values from SimaPro. The relative rankings of the
containers also differed. While glass and PET bottles in this comparison required
roughly the same amount of energy in COMPASS, glass required less energy than PET
52

in SimaPro and more energy than PET in GaBi. PLA bottles required approximately the
same energy in COMPASS and SimaPro, yet GaBi reported less than half as much
energy required.
Distinctly different results arose from the three software programs used to look at
eutrophication in the beverage container comparison, as shown in Figure 7. COMPASS
and GaBi reported significantly higher impact values than SimaPro, with ratios ranging
from 1.39 to 140. There was little agreement between software on how the containers
should be ranked relative to each other for this impact category. COMPASS and
SimaPro ranked the PLA bottle as the highest impact, but disagreed on the order for the
remaining containers. COMPASS and GaBi only agreed that PET had the second
highest impact; there was no agreement between them for the remaining containers.
0.0008
0.0007

kg PO4 equiv

0.0006
0.0005
0.0004
0.0003
0.0002
0.0001
0.0000
COMPASS
Al can

PET bottle

SimaPro
Glass bottle

Aseptic carton

GaBi
PLA bottle

Figure 7: Eutrophication for selected beverage containers.
Results of the beverage container comparison for water depletion, or water
consumption as it is referred to in COMPASS, are shown in Figure 8. Notice that the
vertical scale for this graph is logarithmic; this is necessitated by GaBi reporting water
53

depletion that was several orders of magnitude greater than COMPASS or SimaPro.
The greatest difference occurred for the aluminum can; COMPASS and SimaPro
reported less than 1.0 liter of water depletion while GaBi reported over 4,200 liters.
Even the smallest relative difference was sizable, occurring for the aseptic carton with
COMPASS reporting 3.6 liters, SimaPro 2.3 liters, and GaBi 376 liters. In addition, the
relative ranking of the containers for water depletion varied with software; COMPASS
had the glass bottle requiring the most water, in SimaPro it was the aseptic carton, and
there was a tie in GaBi between the aluminum can and the glass bottle
1.0E+04
1.0E+03

liters

1.0E+02
1.0E+01
1.0E+00
1.0E-01

1.0E-02
1.0E-03
COMPASS
Al can

PET Bottle

SimaPro
Glass bottle

Aseptic Carton

GaBi
PLA Bottle

Figure 8: Water depletion/consumption for selected beverage containers.
The effect of truck and rail transport on impact indicators was looked at using
beverage container scenarios as the basis of comparison. Package Modeling was
removed from consideration in these tests after it was discovered that the available
transport distance selections had no effect on the greenhouse gas emissions reported
by this software. Table 5 lists parameters for the truck and rail comparisons for the glass
bottle.
54

Table 5: Parameters for truck and rail transport comparison
Container Type

Recycled Recycled Composted
Truck
1
1
Content At EOL
At EOL
(km)
(%)
(%)
(%)

Rail
(km)

Air
(km)

Ship
(km)

Truck Tests
Glass bottle
Glass bottle

25
25

40
40

0
0

100
1000

500
500

0
0

0
0

Rail Tests
Glass bottle
Glass bottle
Glass bottle

25
25
25

40
40
40

0
0
0

100
100
100

0
500
4000

0
0
0

0
0
0

1

EOL = End of Life.

0.7

kg CO2 equiv

0.6
0.5
0.4
0.3
0.2
0.1
0.0
0

100 200 300 400 500 600 700 800 900 1000
km
COMPASS

SimaPro

GaBi

Figure 9: Greenhouse gas emissions for glass bottle using truck transport.

Figure 9 shows the plot of greenhouse gas emissions versus truck transport
distance for the glass bottle; all other parameters were held constant. SimaPro and
GaBi produced basically identical slopes, 4.87x10

-5

kg CO2/km and 4.88x10

CO2/km respectively, while the slope from COMPASS was 1.04x10

-4

-5

kg

kg CO2/km, more

than double the others. Given the initial starting points at 100 km, COMPASS results
55

grew increasing larger than those of SimaPro from the start, but it would take
approximately 1,600 km of truck transport distance before COMPASS would report
higher greenhouse gas emissions than GaBi. The change in emissions from 100 km to
1,000 km represented a 15.5% increase in COMPASS, an 8.4% increase in SimaPro,
and a 6.9% increase in GaBi.
Figure 10 shows how greenhouse gas emissions varied with rail transport
distance for the glass bottle. Rail transport distance was the only variable; all other
parameters were held constant. The slope produced by SimaPro, 2.81 x10
CO2/km, was similar to that from GaBi at 2.63 x10

-5

-5

kg CO2/km, while the 1.15 x10

-5

kg
kg

CO2/km slope from COMPASS was less than half the others. Given the initial offset
points, GaBi results grew increasing larger than COMPASS from the start, but it took
approximately 2,250 km of rail transport distance before SimaPro reported higher
greenhouse gas emissions than COMPASS. The change in emissions from 0 km to
4,000 km represented a 9.2% increase in COMPASS, a 24.2% increase in SimaPro,
and an 18.1% increase in GaBi.

56

0.8

kg CO2 equiv

0.7
0.6
0.5
0.4
0.3
0.2

0.1
0.0
0

500

1000 1500 2000 2500 3000 3500 4000
km

COMPASS

SimaPro

GaBi

Figure 10: Greenhouse gas emissions for glass bottle using rail transport.

9.2 Flexible Versus Rigid Containers
The parameters used in the flexible versus rigid container comparison presented
here are given in Table 6. All data are based on a functional unit of 1 kilogram of tuna. A
difficulty arose in trying to account for recycled content of steel when comparing results
from different software. The database used with GaBi fixed the recycled content of steel
at 25%, Package Modeling fixed it at 28%, COMPASS fixed it at 37%, and the database
used with SimaPro allowed 0% to 100%. Since it was not possible to compare results
from all the software with the same amount of recycled content for steel, it was decided
to compare GaBi to SimaPro at 25%, and COMPASS to SimaPro at 37%. With Package
Modeling only able to provide a comparison to the other software for global warming, it
was further decided to group it with the GaBi and SimaPro comparison for this single
impact category.

57

Table 6: Parameters for flexible versus rigid container comparison
Container Type
Flexible Pouch

Recycled Recycled Composted
1
1
Content At EOL
At EOL
(%)
(%)
(%)
0
0
0

Steel Can
Steel Can
1

25
37

70
70

0
0

Truck
(km)

Rail
(km)

Air
(km)

Ship
(km)

100

0

0

0

100
100

0
0

0
0

0
0

EOL = End of Life.

The comparison of greenhouse gas emissions for the global warming impact
category is given in Figure 11. It can be seen in Figure 11A that while the relative
ranking of the 37% recycled content steel can versus the flexible pouch was the same in
SimaPro and COMPASS, SimaPro reported significantly higher greenhouse gas
emissions for the steel can than COMPASS did, 62% higher. Emissions for the flexible
pouch had better agreement, with SimaPro 5% lower than COMPASS. Figure 11B
shows the 25% to 28% recycled content steel can was consistently ranked by all three
software programs as having higher greenhouse gas emissions than the flexible pouch.
The variation of greenhouse gas emissions reported for the steel can was less, as
shown Figure 11B, with SimaPro results 26% higher than Package Modeling. On the
other hand, variation of greenhouse gas emissions for the flexible pouch was higher, as
shown in Figure 11B, with SimaPro results 73% higher than Package Modeling. For
both containers, GaBi results fell between those of SimaPro and Package Modeling.

58

kg CO2 equiv

0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0

kg CO2 equiv

0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
COMPASS

SimaPro

SimaPro

GaBi

Package
Modeling

Steel Can - 37% recycled content

Steel Can - 25% to 28% recycled content

Flexible Pouch

Flexible Pouch

Part A
Part B
Figure 11: Greenhouse gas emissions for steel cans and flexible pouch.

Figure 12 presents results for the fossil fuel/non-renewable energy impact
category. COMPASS and SimaPro reversed rankings for the 37% recycled content
steel can versus the flexible pouch, as can be seen in Figure 12A. COMPASS reported
the steel can 20% lower than the pouch while SimaPro placed the can 8% higher.
SimaPro was 58% higher in reported equivalent energy for the steel can when
compared to COMPASS, and 17% higher for the flexible pouch. Figure 12B shows the
results for the 25% recycled content steel can versus flexible pouch in SimaPro and
GaBi. There was a reversal in ranking between the software; SimaPro reported the steel
can 13% higher than the pouch while GaBi placed the can 5% lower. SimaPro reported
higher equivalent energy requirements compared to GaBi, 43% higher for the steel can
and 20% higher for the flexible pouch.

59

12.0

10.0

10.0
MJ Primary Energy

MJ Primary Energy

12.0

8.0
6.0
4.0
2.0
0.0

8.0
6.0
4.0
2.0
0.0

COMPASS

SimaPro

SimaPro

GaBi

Steel Can - 37% recycled content

Steel Can - 25% recycled content

Flexible Pouch

Flexible Pouch

Part A
Part B
Figure 12: Fossil fuel/non-renewable energy for steel cans and flexible pouch.

Eutrophication results are shown in Figure 13. It can be seen in Figure 13A that
COMPASS and SimaPro agreed on the ranking order of the 37% recycled content steel
can versus the flexible pouch; however, COMPASS results were considerably larger in
magnitude, by a multiplication factor of 4.7 for the steel can and 5.6 for the pouch.
SimaPro and GaBi reversed the ranking order for the 25% recycled content steel can
versus the flexible pouch, as shown in Figure 13B. SimaPro reported emissions for the
steel can were 14% lower than the pouch, while GaBi had the can 22% higher. GaBi
reported eutrophication results that were 2.8 times those of SimaPro for the steel can,
and 2.0 for the flexible pouch.

60

0.0005

0.0004

0.0004

kg PO4 equiv

kg PO4 equiv

0.0005

0.0003
0.0002

0.0003
0.0002

0.0001

0.0001

0.0000

0.0000
COMPASS

SimaPro

SimaPro

GaBi

Steel Can - 37% recycled content

Steel Can - 25% recycled content

Flexible Pouch

Flexible Pouch

Part A
Part B
Figure 13: Eutrophication for steel cans and flexible pouch.

Water depletion/consumption results for the steel cans and flexible pouch are
shown in Figure 14. The results reported by GaBi greatly exceed those from SimaPro
and COMPASS, forcing the use of a logarithmic scale for the comparison graphs, just
as happened in the beverage container comparison for water depletion. GaBi reported
502 liters of water required for the 25% recycled content steel can, compared to 6.4
liters reported by SimaPro. GaBi had the flexible pouch requiring 1,460 liters of water,
SimaPro estimated 0.69 liters, and COMPASS 1.0 liter. GaBi also reversed the ranking
order compared to the other software, with lower water usage reported for the steel can
instead of the flexible pouch.

61

1.0E+04

1.0E+03

1.0E+03

1.0E+02

1.0E+02

1.0E+01

1.0E+01

liters

liters

1.0E+04

1.0E+00

1.0E+00

1.0E-01

1.0E-01
COMPASS

SimaPro

SimaPro

GaBi

Steel Can - 37% recycled content

Steel Can - 25% recycled content

Flexible Pouch

Flexible Pouch

Part A
Part B
Figure 14: Water depletion/consumption for steel cans and flexible pouch.

9.3 Single Use Box Versus Reusable Crate
Test parameters for exploring how the software handled comparison of a single
use corrugated box to a reusable plastic (PP) crate are given in Table 7. All data is
based on a functional unit of ½ flower box, a standard unit of measure for the cut flower
industry. For this comparison all containers were subject to 200 km of truck transport
and 2,500 km of air transport on the out going journey, with the return trip involving
2,100 km by ship (ocean) transport and 1,200 km additional truck transport. This loop is
based on a model for shipment of cut flowers which required air shipment for quick
delivery, but the return of the empty reusable PP crate was done by more cost effective
means. Please note that there is no return of the PP crate from its last use, meaning
that a single use crate did not go through a return cycle.

62

Table 7: Parameters for reuse comparison
Container
Type
PP Crate,
1 use
PP Crate,
10 uses
PP Crate,
100 uses
Corrugated
Box, 1 use
1
2

Recycled Recycled Composted
2
1
1
Truck
Content At EOL
At EOL
(km)
(%)
(%)
(%)

2

2

2

Rail
(km)

Air
(km)

Ship
(km)

0

10

0

200

0

2,500

0

0

10

0

1,400

0

2,500

2,100

0

10

0

1,400

0

2,500

2,100

50

80

0

200

0

2,500

0

EOL = End of Life.
Transport distances are on per use basis, except no return transport on last use.

Results of the reuse comparison for the global warming impact category are
presented in Figure 15 in terms of average emissions per use, where each use
constitutes one trip through the distribution cycle. As previously stated, it was found that
Package Modeling did not vary reported greenhouse gas emissions with changes in
shipping distance; thus it did not provide a means to account for the contribution to
emissions made by each pass through the distribution cycle. For this reason, the results
from Package Modeling are shown only for reference as single use values. Looking at
results from the other software, it can be seen that each software produced a different
result. GaBi had the single use corrugated box having 19% lower greenhouse gas
emissions than the 100-use PP crate, SimaPro reversed the order with the corrugated
box having 73% more emissions, and COMPASS rated them approximately equal with
only a 2% difference.

63

kg CO2 eqiv

7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
COMPASS

SimaPro

GaBi

Package
Modeling
PP crate
PP crate
PP crate
Corrugated Box
1 use
10 uses *
100 uses *
1 use
* Values are average per use.
Figure 15: Greenhouse gas emissions for reuse comparison.
Figure 16 shows the reuse comparison for the fossil fuel/non-renewable energy
impact category, again in terms of average per use values. Here SimaPro and GaBi
agreed on the ranking order of the results, though not on magnitudes. GaBi had the
corrugated box taking 30% more energy than the 100-use PP crate, while SimaPro had
the corrugated box taking 62% more energy. COMPASS ranked the corrugated box and

MJ Primary Energy

100-use PP crate about equal, with only a 6% difference in energy between them.
225
200
175
150
125
100
75
50
25
0
COMPASS
PP crate
1 use

SimaPro

PP crate
10 uses *

GaBi

PP crate
Corrugated Box
100 uses *
1 use
* Values are average per use.
Figure 16: Fossil fuel/non-renewable energy for reuse comparison.

64

The eutrophication results for the reuse comparison are given in Figure 17.
Notice that the vertical scale of the graph is logarithmic; this is due to the results
reported by SimaPro being substantially smaller in magnitude than results from
COMPASS or GaBi. COMPASS values are 7 to 136 times those of SimaPro, and GaBi
values are 5 to 24 times those of SimaPro. The three software programs differed on
where they ranked the corrugated box relative to the PP crate options; SimaPro ranked
the corrugated box as having the highest emissions, GaBi ranked it as second highest,
and COMPASS ranked it as third highest.

kg PO4 equiv

1.0E-02
1.0E-03
1.0E-04
1.0E-05
1.0E-06
COMPASS
PP crate
1 use

SimaPro

PP crate
10 uses *

PP crate
100 uses *

GaBi
Corrugated Box
1 use

* Values are average per use.
Figure 17: Eutrophication for reuse comparison.

Water depletion/consumption results for the reuse comparison are given in
Figure 18. Note the vertical scale of the graph is logarithmic, a consequence of GaBi
producing numbers several orders of magnitude greater than the other software. As an
example, GaBi reported approximately 11,200 liters of water depletion associated with
the single use PP crate, COMPASS put the number at 12.6 liters, and SimaPro at 8.6

65

liters. These water depletion/consumption magnitude differences are consistent with
those obtained for the other packaging systems considered in this paper; GaBi always
reported significantly higher water depletion than the other software. In addition to
differences in magnitudes, GaBi also differed from the other programs by reporting the
single use PP crate required more water than the corrugated box.

1.0E+05
1.0E+04

liters

1.0E+03
1.0E+02
1.0E+01
1.0E+00
1.0E-01
1.0E-02

COMPASS
PP crate
1 use

PP crate
10 uses *

SimaPro
PP crate
100 uses *

GaBi
Corrugated Box
1 use

* Values are average per use.
Figure 18: Water depletion/consumption for reuse comparison.

The corrugated box package system was also used to explore how the software
compared when dealing with varying amounts of recycled content in a package. Table 8
lists the parameters associated with this series of tests. The only parameter being
varied is the recycled content; all other parameters remained constant. As originally
conceived, the tests called for recycled content of 0%, 25%, 50%, and 100%. However,
it was discovered that COMPASS only allowed recycled content of corrugated board to
be 12% to 87%. Rather than alter the basis of comparison in the middle of collecting
data, it was decided to look at this as an LCA practitioner trying to model a package as
66

closely as possible to a set of requirements s/he has been given. Under this premise,
12% and 87% are substituted in the COMPASS model as the closest match possible to
the desired values of 0% and 100%.

Table 8: Parameters for corrugated box recycled content comparison
Container
Type
Corrugated
Box - 1 use
Corrugated
Box - 1 use
Corrugated
Box - 1 use
Corrugated
Box - 1 use
1
2

Recycled
Content
(%)
2

Recycled Composted
1
1
At EOL
At EOL
(%)
(%)

Truck
(km)

Rail
(km)

Air
(km)

Ship
(km)

80

0

200

0

2,500

0

25

80

0

200

0

2,500

0

50

80

0

200

0

2,500

0

80

0

200

0

2,500

0

0 (12)

100 (87)

2

EOL = End of Life.
Values in parentheses are for COMPASS.

Greenhouse gas emissions reported by COMPASS, SimaPro, and GaBi for
corrugated boxes with varying amounts of recycled fiber content are shown in Figure
19. Package Modeling had no means of varying recycled content of corrugated board,
so it could not be included in this comparison. While there are differences in magnitudes
of the emissions reported by the remaining three programs, the most striking result of
the comparison is the complete reversal of ranking order between software. COMPASS
and GaBi reported the lowest greenhouse gas emissions occurred when recycled
content was small, meaning large virgin fiber content, and the highest emissions
occurred when recycled content was large. SimaPro reported the exact opposite order,
with highest emissions coinciding with the smallest recycled content. The implication of

67

this discrepancy is that COMPASS and GaBi would drive a design toward using less
recycled fiber content to reduce greenhouse gas emissions, while SimaPro would direct
designs toward using more recycled fiber content to accomplish the same goal.
1.2
kg CO2 eqiv

1.0
0.8

0.6
0.4
0.2
0.0
COMPASS

0% (12%)*

SimaPro

25%

50%

GaBi

100% (87%)*

Percent Recycled Content
*Values in parentheses are for COMPASS
Figure 19: Greenhouse gas emissions for corrugated box comparison.

Fossil fuel/non-renewable energy results for the corrugated box recycled content
comparison are given in Figure 20. All the software were in agreement regarding
ranking order; there was no reversal for this impact category. There was some variation
in the magnitudes of the reported emissions, with SimaPro ranging from 15% to 27%
higher than COMPASS, and 28% to 35% higher than GaBi.

68

MJ Primary Energy

18
16
14
12
10
8
6
4
2
0
COMPASS

SimaPro

GaBi

0% (12%)*

25%
50%
100% (87%)*
Percent Recycled Content
*Values in parentheses are for COMPASS

Figure 20: Fossil fuel/non-renewable energy for corrugated box comparison.

Figure 21 shows the results of the corrugated box recycled content comparison
for eutrophication. The software agreed on the ranking order, however, there was
noticeable disagreement in the reported magnitudes of emissions. SimaPro emission
numbers ranged from 6% to 21% of those from COMPASS, and from 11% to 23% of

kg PO4 equiv

the values from GaBi.
0.0018
0.0016
0.0014
0.0012
0.0010
0.0008
0.0006
0.0004
0.0002
0.0000
COMPASS

SimaPro

GaBi

0% (12%)*

25%
50%
100% (87%)*
Percent Recycled Content
*Values in parentheses are for COMPASS

Figure 21: Eutrophication for corrugated box comparison.
69

Water

depletion/consumption

for

the

corrugated

box

recycled

content

comparison is given in Figure 22. Once again the vertical scale of the graph is
logarithmic to accommodate GaBi consistently reporting water depletion more than two
orders of magnitude greater than either COMPASS or SimaPro. All the software agreed
on the ranking order for this impact category.
10000

liters

1000
100
10
1
COMPASS
0% (12%)*

SimaPro
25%

50%

GaBi
100% (87%)*

Percent Recycled Content
*Values in parentheses are for COMPASS
Figure 22: Water depletion/consumption for corrugated box comparison.

The effect of air and ship (ocean) transport on impact indicators was looked at
using the corrugated box and PP crate reuse scenarios as the basis of comparison. As
with the truck and rail comparisons earlier, Package Modeling was removed from
consideration in these tests after it was discovered that the available transport distance
selections had no effect on the greenhouse gas emissions reported by this software.
Table 9 lists parameters for the air and ship transport comparisons. The only
parameters varied in these tests were the transport distances, and only one mode of
transport was varied at a time for a given packaging system.

70

Table 9: Parameters for air and ship transport comparisons
Container
Type
Air
Corrugated
Box - 1 use
Corrugated
Box - 1 use
PP Crate
- 10 uses
PP Crate
- 10 uses
Ship
PP Crate
- 10 uses
PP Crate
- 10 uses
1
2

Recycled Recycled Composted
2
1
1
Truck
Content At EOL
At EOL
(km)
(%)
(%)
(%)

2

2

2

Rail
(km)

Air
(km)

Ship
(km)

50

80

0

200

0

500

0

50

80

0

200

0

2500

0

0

10

0

1,400

0

500

2,100

0

10

0

1,400

0

2,500

2,100

0

10

0

1,400

0

2,500

500

0

10

0

1,400

0

2,500

2,100

EOL = End of Life.
Transport distances are on per use basis, except no return transport on last use.

Figure 23 gives the greenhouse gas emissions reported for the air transport
comparisons. It can be seen in Figure 23A that the slopes of the plots for the corrugated
box were all similar, ranging from 4.18x10

-5

kg CO2/km to 4.42x10

-5

kg CO2/km. Figure

23B shows the slopes of the plots for the PP crate had good agreement, ranging from
1.07 x10

-4

kg CO2/km to 1.09 x10

-4

kg CO2/km. The difference in slopes between the

corrugated box versus the PP crate was due primarily to the difference in weight, the
corrugated box was 0.7kg and the PP crate was 1.8kg. Figure 24 gives the greenhouse
gas emissions reported for the ship (ocean) transport comparisons. The slopes of these

71

plots were also very similar, ranging from 2.89x10

-5

kg CO2/km to 2.92x10

-5

kg

1.4

1.2

1.2
kg CO2 equiv

1.4
1.0
0.8
0.6
0.4

1.0
0.8
0.6
0.4

0.2

0.2

0.0

0.0
0

500 1000 1500 2000 2500
km

COMPASS

SimaPro

0

GaBi

500 1000 1500 2000 2500
km

COMPASS

SimaPro

GaBi

Part A – Corrugated Box
Part B – Plastic (PP) Crate
Figure 23: Greenhouse gas emissions versus air transport distance.

1.4
1.2
kg CO2 equiv

kg CO2 equiv

CO2/km.

1.0
0.8
0.6
0.4
0.2
0.0
0

500

1000

1500

2000

2500

km
COMPASS

SimaPro

GaBi

Figure 24: Greenhouse gas emissions for ship transport of plastic (PP) crate.

72

9.4 Comparison Summary for Package Systems
Significant discrepancies were found in the results reported by the various LCA
software programs for the packaging system comparisons. Notable among them were
the large magnitudes of water depletion/consumption reported by GaBi relative to
COMPASS and SimaPro for every package system comparison, and the complete
reversal of ranking order that COMPASS and GaBi had relative to SimaPro when
looking at how greenhouse gas emissions varied with recycled content of a corrugated
box. All four impact categories studied had variations in magnitudes and ranking order
of results at some point in the comparisons, and all packaging systems studied had
multiple inconsistencies in results between software programs.
In work contemporary to what was done here for package system comparisons,
researchers from the Laboratory for Manufacture and Sustainability at the University of
California, Berkeley, performed a study comparing various LCA software (Simon et al.
2012). Their approach involved doing detailed LCAs on 19 products in each of 5 LCA
software programs: GaBi, SimaPro, COMPASS, Wal-Mart Sustainable Packaging
Scorecard (Package Modeling), and Sustainable Minds.

They came across many

discrepancies in results, concluding “that each tool has advantages and opportunities
for improvement” and recommending “users select the tool that best suits their needs.”
While providing guidance in what could be expected in using the various LCA software,
their published research ends at this point. In contrast, the package system
comparisons in this paper represent the start of the research. With the presence of
significant discrepancies in using the various software established, work then turned to

73

finding an effective and efficient way to isolate and identify the causes of the
discrepancies.

74

10. BASIC MATERIAL COMPARISONS
The package system comparisons, while providing a high level look at the
consistency of results across several LCA software programs, do not directly address
the question of where discrepancies in the results come from. Differences could be due
to several things, including choices made when implementing system models. To
reduce the human factor in the comparisons, and to provide a more consistent basis for
further exploration of inconsistencies, another series of tests was performed using
simple systems, ones that minimized the number of variables even more than the
package system comparisons. To accomplish this, four basic packaging materials were
selected: aluminum, corrugated board, glass, and PET. Each basic system studied
consisted of obtaining and disposing of a single material. No additional materials,
converting, distribution, or other processes were included. The functional unit for each
system was 1 kilogram of the selected material.
It was decided to limit SimaPro and GaBi to using only files from the Ecoinvent
2.2 database, with every effort made to use the same database components for both
software programs. The intent was to create a set of results that could be used to
identify likely areas for further research in tracing any discrepancies down to the level of
how specific input files from the database were handled. COMPASS was included in the
basic material comparisons to see how a streamlined LCA program would stack up
against the fully ISO 14040/14044 compliant programs, though the proprietary nature of
the COMPASS dataset precluded it from being part of the effort to determine the causes
of any differences. Package Modeling is designed for complete packages, not materials,
so it was not included in these tests.

75

Consideration of using openLCA in the basic material comparisons faced a set of
pros and cons. Pros included that openLCA could use the Ecoinvent database, as well
as the Impact 2002+ and ReCiPe impact assessment methods. Cons included that
openLCA was found to be a work in progress, documentation was poor, several
operational problems were encountered, and the implementation of Impact 2002+ for
openLCA would only produce endpoint values for many impact categories, not the
desired midpoint values. Another issue was that no reference could be found to identify
the versions of Impact 2002+ and ReCiPe that were available to use with openLCA. In
the end, it was decided that some openLCA results would be compared with SimaPro
separately for reference purposes, but openLCA was not included in the analysis of
causes of differences in results.
The data for all graphs presented in this section can be found in Appendix H.

10.1 COMPASS, SimaPro, and GaBi Comparisons
The parameters used for the comparisons of the four basic materials are listed in
Table 10. The greenhouse gas emissions reported for 1 kg of each of these materials
are shown in Figure 25. All three software programs reported aluminum as having the
highest emissions, and PET as the second highest. There was some variation in the
magnitude of emissions reported for aluminum, with GaBi numbers being 19% greater
than SimaPro and 21% greater than COMPASS. The magnitudes of emissions for PET
varied less than 4% across the three software programs. The ranking order of
corrugated board and glass did vary between the software. COMPASS had emissions
for corrugated board 13% lower than the emissions for glass, while SimaPro reported

76

corrugated board 9% higher than glass, and GaBi reported corrugated board 28%
higher.

Table 10: Parameters for comparison of four basic materials
Functional
Unit
(kg)

Basic Material

Recycled Recycled
Content At End of
(%)
Life (%)

Landfill
At End of
Life (%)

Waste To
Energy At End
of Life (%)

1
1
1

50
50
55.5

67.9
76.4
63.8

32.1
12.9
36.2

0
10.7
0

Polyethylene
Terephthalate (PET)

1

0

26.1

45.1

28.8

kg CO2 equiv

Aluminum
Corrugated Board
Packaging Glass

8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
COMPASS

SimaPro

GaBi

Aluminum
Corrugated Board
Glass
PET
Figure 25: Greenhouse gas emissions for basic material comparison.

The comparison of fossil fuel/non-renewable energy requirements for the four
basic materials is given in Figure 26. All software reported the same ranking order, with
aluminum having the greatest energy requirements and corrugated board the lowest.
There was some variation in reported magnitudes across the software, most notably
that COMPASS reported lower magnitudes than either SimaPro or GaBi. COMPASS
77

only reported fossil fuel use; it did not include nuclear energy, which may be a factor

MJ Energy

here.
100
90
80
70
60
50
40
30
20
10
0

COMPASS

SimaPro

GaBi

Aluminum
Corrugated Board
Glass
PET
Figure 26: Fossil fuel/non-renewable energy for basic material comparison.

Eutrophication emissions for the basic materials are presented in Figure 27.
SimaPro and GaBi had the same ranking order for the materials, placing aluminum as
having the highest emissions and PET as second highest. COMPASS reversed the
order of the first two, placing PET higher in emissions than aluminum. There were
differences in magnitudes of emissions, with GaBi reporting emissions for aluminum
that were a multiplication factor of 3.6 of those reported by COMPASS, and 6.1 of those
reported by SimaPro. At the same time, GaBi reported emissions for the other materials
that were only 52% to 67% of what COMPASS reported. Emissions reported by
SimaPro were consistently lower than the other software, 10% to 27% of what
COMPASS reported and 19% to 40% of what GaBi reported.
Water depletion/consumption results for the basic material comparisons are
given in Figure 28. Note the use of a logarithmic vertical scale; this is due to GaBi
reporting water use that was orders of magnitude greater than what was reported by
78

either COMPASS or SimaPro. As an example, GaBi reported 165,000 liters of water
associated with producing and disposing of 1kg of aluminum, compared to 32 liters
reported by both COMPASS and SimaPro. This is similar to results for water
depletion/consumption found in the comparisons of packaging systems; estimates from

kg PO4 equiv

GaBi were consistently orders of magnitude higher than estimates from other software.

0.010
0.009
0.008
0.007
0.006
0.005
0.004
0.003
0.002
0.001
0.000
COMPASS

SimaPro

GaBi

Aluminum
Corrugated Board
Glass
PET
Figure 27: Eutrophication for basic material comparison.

1,000,000
100,000
liters

10,000
1,000
100
10
1
COMPASS

SimaPro

GaBi

Aluminum
Corrugated Board
Glass
PET
Figure 28: Water depletion/consumption for basic material comparison.

The effect recycled content can have on impact category indicators was explored
for the basic materials using corrugated board. The parameters used for the recycled
79

content comparison are listed in Table 11. The estimates of greenhouse gas emissions
for varying recycled fiber content of corrugated board are given in Figure 29. COMPASS
and GaBi had emissions increasing as recycled content increased, similar to what was
found for the corrugated box in the packaging systems comparisons. Results from
SimaPro had a nearly flat response to varying recycled content, clearly disagreeing with
COMPASS and GaBi, but somewhat different than what was found for the corrugated
box in the packaging systems comparison. The flat response effect was traced to
energy usage data coming from European datasets for the basic materials, while energy
data for comparisons of packaging systems came from U.S. datasets.
.
Table 11: Parameters for recycled content comparison of basic corrugated board.

Basic Material

Functional
Unit
(kg)

Corrugated Board
Corrugated Board
Corrugated Board

1
1
1

Recycled Recycled
Content At End of
(%)
Life (%)
12
50
87

76.4
76.4
76.4

Landfill
At End of
Life (%)

Waste To
Energy At End
of Life (%)

12.9
12.9
12.9

10.7
10.7
10.7

1.4
kg CO2 equiv

1.2
1.0
0.8
0.6
0.4
0.2
0.0
COMPASS

SimaPro

GaBi

12%
50%
87%
Figure 29: Greenhouse gas emissions for basic corrugated board.
80

Results for fossil fuel/non-renewable energy requirements when varying recycled
content of corrugated board are given in Figure 30. COMPASS and GaBi reversed
ranking order, with COMPASS having energy requirements increasing as recycled
content increased, and GaBi having energy requirements decreasing. SimaPro had a
flat response, keeping energy use nearly constant as recycled fiber content varied.
SimaPro reported magnitudes of energy usage that were 6% to 14% higher than GaBi,

MJ Energy

and 15% to 36% higher than COMPASS.
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
COMPASS

SimaPro

GaBi

12%
50%
87%
Figure 30: Fossil fuel/non-renewable energy for basic corrugated board.

Figure 31 covers the effect of varying recycled fiber content on eutrophication
emissions for corrugated board. While all three software programs had the same
ranking order, with emissions decreasing as recycled content increased, there are
noticeable differences in the magnitudes of emissions reported by SimaPro versus the
other software. The values from SimaPro are 20% to 32% of those from COMPASS,
and 24% to 32% of the values obtained from GaBi.

81

kg PO4 equiv

0.0020
0.0018
0.0016
0.0014
0.0012
0.0010
0.0008
0.0006
0.0004
0.0002
0.0000
COMPASS

SimaPro

GaBi

12%
50%
87%
Figure 31: Eutrophication for basic corrugated board.

Water depletion/consumption results versus recycled fiber content for corrugated
board are shown in Figure 32. Note that the vertical scale of the graph is logarithmic, as
once again GaBi reported significantly higher water depletion than COMPASS and
SimaPro. GaBi had water consumption ranging from 1,480 liters to 4,080 liters, while
COMPASS ranged from 7.7 liters to 40.2 liters and SimaPro ranged from 12.5 liters to
37.0 liters. All three software programs reported the same ranking order, with water
depletion decreasing as recycled fiber content increased.
10,000

liters

1,000
100

10
1

COMPASS

SimaPro

GaBi

12%
50%
87%
Figure 32: Water depletion/consumption for basic corrugated board.

82

10.2 SimaPro and openLCA Comparisons
The SimaPro and openLCA basic material comparisons used the same test
parameters listed in Table 10 of section 10.1 for the COMPASS, SimaPro, and GaBi
comparisons. The difference is that some of the impact categories are expressed in
points rather than equivalent amounts of reference substances. Figure 33 shows the
comparisons for greenhouse gas emissions. The ranking orders from the programs
were the same, but SimaPro did have corrugated board and glass more evenly matched
than openLCA. SimaPro results were consistently higher in points than openLCA.
Similar results for non-renewable energy can be seen in Figure 34.
0.0007
0.0006
points

0.0005
0.0004
0.0003
0.0002
0.0001
0.0000
SimaPro

openLCA

Aluminum
Corrugated Board
Glass
PET
Figure 33: openLCA greenhouse gas emissions for basic materials.

Eutrophication results are shown in Figure 35. SimaPro again had higher
magnitudes, with openLCA values only 17% to 39% those of SimaPro. There was a
slight change in ranking order of aluminum and corrugated board between the two
software programs, with SimaPro placing emissions for aluminum 4% higher than
emissions for corrugated board, while openLCA reversed the order with corrugated
board 15% higher than aluminum.
83

0.0007
0.0006
points

0.0005
0.0004
0.0003

0.0002
0.0001
0.0000
SimaPro

openLCA

kg PO4 equiv

Aluminum
Corrugated Board
Glass
PET
Figure 34: openLCA non-renewable energy for basic materials.

0.0016
0.0014
0.0012
0.0010
0.0008
0.0006
0.0004
0.0002
0.0000
SimaPro

openLCA

Aluminum
Corrugated Board
Glass
PET
Figure 35: openLCA eutrophication for basic materials.

Water depletion results are given in Figure 36. The vertical scale is logarithmic,
this time due to the significantly smaller amount of water depletion reported by openLCA
for glass. SimaPro had 9.0 liters of water being depleted for 1 kg of glass, while
openLCA placed the amount at 0.08 liters. There was a reversal of ranking order for
aluminum and corrugated board; SimaPro reported aluminum needed 34% more water
than corrugated board, while openLCA indicated aluminum used 25% less water.

84

1.0E+02

liters

1.0E+01
1.0E+00
1.0E-01
1.0E-02
SimaPro

openLCA

Aluminum
Corrugated Board
Glass
PET
Figure 36: openLCA water depletion for basic materials.

Tests looking at the effects varying recycled content had on impact assessment
indicators were done with openLCA, similar to the comparisons for COMPASS,
SimaPro, and GaBi reported in section 10.1. The same set of parameters listed in Table
11 for the earlier tests were also used with openLCA. Greenhouse gas emissions for
these openLCA tests are shown in Figure 37, along with results from SimaPro for
comparison. The emissions estimates from openLCA increased slightly as the recycled
fiber content increased, while SimaPro estimates remain flat. The magnitudes of
openLCA estimates were 50% to 60% of the estimates from SimaPro. Similar results
were obtained for non-renewable energy, shown in Figure 38.
Eutrophication estimates, shown in Figure 39, have values reported by openLCA
that are 11% to 26% of those reported by SimaPro. The two programs did provide the
same ranking order, with emissions going down as recycled content increased. Ranking
order for water depletion was also the same between the two programs, as can be seen
in Figure 40.

85

0.00012

points

0.00010
0.00008
0.00006
0.00004
0.00002

0.00000
SimaPro
openLCA
12%
50%
87%
Figure 37: openLCA greenhouse gas emissions for corrugated board.
0.00012

points

0.00010
0.00008
0.00006
0.00004
0.00002
0.00000
SimaPro
openLCA
12%
50%
87%
Figure 38: openLCA non-renewable energy for corrugated board.

kg PO4 equiv

0.0015
0.0010
0.0005
0.0000
SimaPro
openLCA
12%
50%
87%
Figure 39: openLCA eutrophication for corrugated board.

86

liters

45
40
35
30
25
20
15
10
5
0
SimaPro

openLCA

12%
50%
87%
Figure 40: openLCA water depletion for corrugated board.

10.3 Other Impact Categories
Up to this point the comparisons between software were limited to four impact
categories due to the inclusion of COMPASS. However, additional comparisons can be
made between SimaPro and GaBi. Impact 2002+ and ReCiPe impact assessment
methods have been used with SimaPro and GaBi throughout this paper. Impact 2002+
has 15 different midpoint impact categories, and ReCiPe has 18. Therefore, the
question of how many of these impact categories might show significant differences for
the basic materials being considered was examined.
Table 12 shows all Impact 2002+ and ReCiPe impact category results for
aluminum as used in the earlier basic material comparisons. Rows that are highlighted
in bold text indicate impact categories where the result from one software program is at
least twice the result from the other. This factor of 2 equates to a difference value in
Table 12 of either less than or equal to -50%, or greater than or equal to 100%, where
% Difference = 100% x (GaBi - SimaPro) / SimaPro. It can be seen that 7 of the Impact
2002+ categories fit this criteria for aluminum, and 5 categories from ReCiPe also fit.
87

Table 13 shows that 9 Impact 2002+ categories fit the factor of 2 criteria for corrugated
board, while 4 categories fit for ReCiPe, with marine eutrophication falling out of the
group. The same 9 categories for Impact 2002+ fit the factor of 2 criteria for glass as
shown in Table 14; however, marine eutrophication replaced terrestrial ecotoxicity for
the ReCiPe group. For PET, as shown in Table 15, the same 9 Impact 2002+ categories
fit the criteria, and the number of categories that fit the criteria for ReCiPe stays at 4,
though terrestrial ecotoxicity replaced marine eutrophication.

Table 12: Aluminum impact assessment data, 50% recycled content
Impact Category
Impact 2002+
Aquatic acidification
Aquatic ecotoxicity
Aquatic eutrophication
Carcinogens
Global warming
Ionizing radiation
Land occupation
Mineral extraction
Non-carcinogens
Non-renewable energy
Ozone layer depletion
Photochemical oxidation
- Respiratory organics
Respiratory effects Respiratory inorganics
Terrestrial acid/nutrification
Terrestrial ecotoxicity
ReCiPe
Agricultural land
occupation
Climate change
Fossil depletion
Freshwater ecotoxicity
Freshwater eutrophication
Human toxicity

Unit

SimaPro

GaBi

% Difference

kg SO2 eq
kg TEG water
kg PO4 P-lim
kg C2H3Cl eq
kg CO2 eq
Bq C-14 eq
m2org.arable
MJ surplus
kg C2H3Cl eq
MJ primary
kg CFC-11 eq

3.013E-02
1.217E+03
1.523E-03
3.040E-01
5.845E+00
1.790E+02
2.701E-02
1.416E+00
3.645E-02
9.169E+01
3.181E-08

2.225E-02
1.230E+06
9.270E-03
9.620E-02
6.959E+00
1.708E+02
3.216E-06
1.779E+00
1.290E+00
8.839E+01
4.430E-07

-26
100,960
509
-68
19
-5
-100
26
3,439
-4
1,293

kg C2H4 eq

1.477E-04

1.126E-03

662

kg PM2.5 eq

5.735E-03

7.821E-03

36

kg SO2 eq
kg TEG soil

8.596E-02 8.598E-02
7.669E+01 1.451E+02

0
89

m a

8.306E-02

0

kg CO2 eq
kg oil eq
kg 1,4-DB eq
kg P eq
kg 1,4-DB eq

6.812E+00 6.812E+00
1.752E+00 1.752E+00
7.355E-02 6.239E-02
3.059E-03 3.059E-03
3.218E+00 1.111E+00

2

88

8.306E-02

0
0
-15
0
-65

1

Table 12 (cont’d)
Ionizing radiation
Marine ecotoxicity
Marine eutrophication
Metal depletion
Natural land transformation
Ozone depletion
Particulate matter
formation
Photochemical oxidant
formation
Terrestrial acidification
Terrestrial ecotoxicity
Urban land occupation

kg U235 eq
kg 1,4-DB eq
kg N eq
kg Fe eq

Water depletion

m

1
2

m
kg CFC-11 eq

1.705E+00
7.843E-02
1.056E-03
3.964E-01
1.243E-03
5.988E-07

5.637E-01
6.096E-02
4.979E-03
3.925E-01
1.243E-03
5.988E-07

-67
-22
371
-1
0
0

kg PM10 eq

1.319E-02

1.319E-02

0

kg NMVOC

1.703E-02

1.703E-02

0

kg SO2 eq
kg 1,4-DB eq

2.787E-02
7.209E-04
3.806E-02

2.787E-02
1.917E-04
3.806E-02

0
-73
0

3.174E-02

1.651E+02

520,108

2

2

m a
3

% Difference = 100% x (GaBi - SimaPro) / SimaPro
Rows in bold text have at least a factor of 2 difference between SimaPro and GaBi.

Table 13: Corrugated board impact assessment data, 50% recycled content
Impact Category
Impact 2002+
Aquatic acidification
Aquatic ecotoxicity
Aquatic eutrophication
Carcinogens
Global warming
Ionizing radiation
Land occupation
Mineral extraction
Non-carcinogens
Non-renewable energy
Ozone layer depletion
Photochemical oxidation
- Respiratory organics
Respiratory effects Respiratory inorganics
Terrestrial acid/nutrification
Terrestrial ecotoxicity

Unit

SimaPro

GaBi

% Difference

kg SO2 eq
kg TEG water
kg PO4 P-lim
kg C2H3Cl eq
kg CO2 eq
Bq C-14 eq
m2org.arable
MJ surplus
kg C2H3Cl eq
MJ primary
kg CFC-11 eq

4.244E-03 2.341E-03
1.701E+02 3.938E+03
3.955E-04 1.363E-03
2.288E-02 5.583E-03
9.455E-01 5.892E-01
1.927E+01 1.838E+01
4.154E-01 3.336E-05
1.546E-02 1.341E-01
6.085E-03 6.505E-02
1.520E+01 1.380E+01
5.042E-09 8.844E-08

kg C2H4 eq

5.769E-05

3.726E-04

546

kg PM2.5 eq

8.385E-04

9.012E-04

7

kg SO2 eq
kg TEG soil

1.881E-02 1.886E-02
4.826E+00 2.970E+01

89

-45
2,215
245
-76
-38
-5
-100
768
969
-9
1,654

0
515

1

Table 13 (cont’d)
ReCiPe
Agricultural land
occupation
Climate change
Fossil depletion
Freshwater ecotoxicity
Freshwater eutrophication
Human toxicity
Ionizing radiation
Marine ecotoxicity
Marine eutrophication
Metal depletion
Natural land transformation
Ozone depletion
Particulate matter
formation
Photochemical oxidant
formation
Terrestrial acidification
Terrestrial ecotoxicity
Urban land occupation
Water depletion
1
2

2

m a

3.133E+00 3.133E+00

0

kg CO2 eq
kg oil eq
kg 1,4-DB eq
kg P eq
kg 1,4-DB eq
kg U235 eq
kg 1,4-DB eq
kg N eq
kg Fe eq
m
kg CFC-11 eq

1.142E+00 1.116E+00
3.091E-01 3.090E-01
1.021E-02 8.118E-03
4.500E-04 4.491E-04
4.561E-01 8.425E-02
1.837E-01 6.179E-02
8.375E-03 5.806E-03
1.300E-03 2.153E-03
3.946E-02 3.907E-02
5.513E-04 5.523E-04
8.860E-08 8.857E-08

-2
0
-21
0
-82
-66
-31
66
-1
0
0

kg PM10 eq

1.458E-03

1.460E-03

0

kg NMVOC

3.660E-03

3.656E-03

0

kg SO2 eq
kg 1,4-DB eq

3.758E-03
2.899E-04
5.432E-02

3.761E-03
1.041E-04
5.424E-02

0
-64
0

2.462E-02

2.763E+00

11,123

2

2

m a
m

3

% Difference = 100% x (GaBi - SimaPro) / SimaPro
Rows in bold text have at least a factor of 2 difference between SimaPro and GaBi.

Table 14: Glass impact assessment data, 55.5% recycled content
Impact Category
Impact 2002+
Aquatic acidification
Aquatic ecotoxicity
Aquatic eutrophication
Carcinogens
Global warming
Ionizing radiation
Land occupation
Mineral extraction
Non-carcinogens
Non-renewable energy

Unit

SimaPro

kg SO2 eq
kg TEG water
kg PO4 P-lim
kg C2H3Cl eq
kg CO2 eq
Bq C-14 eq
m2org.arable
MJ surplus
kg C2H3Cl eq
MJ primary
90

GaBi

7.771E-03 5.470E-03
1.309E+02 1.963E+04
1.494E-04 6.620E-04
1.514E-02 5.632E-03
8.655E-01 8.020E-01
1.554E+01 1.482E+01
3.126E-02 2.640E-06
6.366E-03 5.229E-02
9.931E-03 8.118E-02
1.558E+01 1.434E+01

% Difference
-30
14,896
343
-63
-7
-5
-100
721
717
-8

1

Table 14 (cont’d)
Ozone layer depletion
Photochemical oxidation
- Respiratory organics
Respiratory effects Respiratory inorganics
Terrestrial acid/nutrification
Terrestrial ecotoxicity
ReCiPe
Agricultural land
occupation
Climate change
Fossil depletion
Freshwater ecotoxicity
Freshwater eutrophication
Human toxicity
Ionizing radiation
Marine ecotoxicity
Marine eutrophication
Metal depletion
Natural land transformation
Ozone depletion
Particulate matter
formation
Photochemical oxidant
formation
Terrestrial acidification
Terrestrial ecotoxicity
Urban land occupation

kg CFC-11 eq

6.444E-09

1.005E-07

1,460

kg C2H4 eq

2.435E-05

2.379E-04

877

kg PM2.5 eq

1.181E-03

1.227E-03

4

kg SO2 eq
kg TEG soil

2.433E-02 2.434E-02
3.898E+00 8.731E+00

Water depletion

m

1
2

2

0
124

m a

2.516E-01

2.516E-01

0

kg CO2 eq
kg oil eq
kg 1,4-DB eq
kg P eq
kg 1,4-DB eq
kg U235 eq
kg 1,4-DB eq
kg N eq
kg Fe eq
m
kg CFC-11 eq

8.985E-01
3.267E-01
5.612E-03
2.236E-04
3.637E-01
1.480E-01
5.724E-03
2.267E-04
2.602E-02
2.833E-04
1.006E-07

8.985E-01
3.267E-01
4.462E-03
2.183E-04
1.438E-01
5.050E-02
4.362E-03
1.372E-03
2.567E-02
2.833E-04
1.006E-07

0
0
-21
-2
-60
-66
-24
505
-1
0
0

kg PM10 eq

2.191E-03

2.191E-03

0

kg NMVOC

4.121E-03

4.120E-03

0

kg SO2 eq
kg 1,4-DB eq

7.004E-03
2.627E-04
9.111E-03

7.004E-03
2.374E-04
9.111E-03

0
-10
0

8.641E-03

1.652E+00

19,021

2

2

m a
3

% Difference = 100% x (GaBi - SimaPro) / SimaPro
Rows in bold text have at least a factor of 2 difference between SimaPro and GaBi.

Table 15: PET impact assessment data, 0% recycled content
Impact Category
Impact 2002+
Aquatic acidification
Aquatic ecotoxicity
Aquatic eutrophication

Unit

SimaPro

kg SO2 eq
kg TEG water
kg PO4 P-lim
91

GaBi

1.054E-02 7.063E-03
5.689E+02 1.003E+04
5.818E-04 3.040E-03

% Difference
-33
1,663
423

1

Table 15 (cont’d)
Carcinogens
Global warming
Ionizing radiation
Land occupation
Mineral extraction
Non-carcinogens
Non-renewable energy
Ozone layer depletion
Photochemical oxidation
- Respiratory organics
Respiratory effects Respiratory inorganics
Terrestrial acid/nutrification
Terrestrial ecotoxicity
ReCiPe
Agricultural land
occupation
Climate change
Fossil depletion
Freshwater ecotoxicity
Freshwater eutrophication
Human toxicity
Ionizing radiation
Marine ecotoxicity
Marine eutrophication
Metal depletion
Natural land transformation
Ozone depletion
Particulate matter
formation
Photochemical oxidant
formation
Terrestrial acidification
Terrestrial ecotoxicity
Urban land occupation

kg C2H3Cl eq
kg CO2 eq
Bq C-14 eq
m2org.arable
MJ surplus
kg C2H3Cl eq
MJ primary
kg CFC-11 eq

1.307E+00 1.616E-02
3.266E+00 3.270E+00
6.499E+01 6.195E+01
1.232E-02 1.297E-06
6.255E-02 5.740E-01
2.665E-02 2.321E-01
8.238E+01 7.618E+01
1.134E-08 1.448E-07

-99
0
-5
-100
818
771
-8
1,177

kg C2H4 eq

1.527E-04

1.922E-03

1,158

kg PM2.5 eq

1.702E-03

1.922E-03

13

kg SO2 eq
kg TEG soil

3.472E-02 3.473E-02
8.800E+00 1.944E+01

0
121

Water depletion

m

1
2

2

m a

5.022E-02

kg CO2 eq
kg oil eq
kg 1,4-DB eq
kg P eq
kg 1,4-DB eq
kg U235 eq
kg 1,4-DB eq
kg N eq
kg Fe eq
m2
kg CFC-11 eq

3.514E+00 3.514E+00
1.745E+00 1.745E+00
4.449E-02 4.047E-02
1.003E-03 1.003E-03
1.133E+00 3.869E-01
6.185E-01 2.115E-01
4.416E-02 3.803E-02
3.317E-03 3.133E-03
2.025E-01 2.007E-01
3.649E-04 3.649E-04
1.450E-07 1.450E-07

kg PM10 eq

3.409E-03

3.409E-03

0

kg NMVOC

8.445E-03

8.299E-03

-2

kg SO2 eq
kg 1,4-DB eq

9.719E-03
2.725E-04
1.347E-02

9.719E-03
1.308E-04
1.347E-02

0
-52
0

1.476E-02

6.990E+00

47,257

2

m a
3

5.022E-02

0
0
0
-9
0
-66
-66
-14
-6
-1
0
0

% Difference = 100% x (GaBi - SimaPro) / SimaPro
Rows in bold text have at least a factor of 2 difference between SimaPro and GaBi.

92

10.4 Comparison Summary For Basic Materials
Even with the test scenarios reduced to nothing more than the creation and
disposal of a basic packaging material, significant differences were found in the results
provided by the LCA programs being studied. Notable among these differences were
the large magnitudes of water depletion/consumption reported by GaBi, and the ranking
orders reported by the various software for greenhouse gas emissions when varying
recycled fiber content of corrugated board. These results were similar to what was
found in the higher level packaging systems comparisons, indicating that at least some
of the issues causing differences in the higher level comparisons are likely rooted in
how the basic materials are handled. Combined with the sheer number of discrepancies
found in the limited testing done so far, there was reason for concern that the choice of
LCA software may affect the results obtained. This raised the question of why are these
discrepancies happening? Could it be firmly established that the issues encountered
were truly with the software, and not how the systems were implemented in the
software? This was the next challenge to be addressed.

93

11. UNDERLYING CAUSES OF DISSIMILAR RESULTS
Having established that using the same impact assessment methods in SimaPro
and GaBi can produce dissimilar results in several impact assessment categories,
attention now turned to establishing why this was happening. The information presented
so far was not sufficient to rule out any cause. Human error in modeling, software errors
in calculations, implementation differences for the assessment methods, or some issue
not yet thought of were all possibilities. The task at hand was to isolate the causes in
order to ascertain why the differences were occurring.
SimaPro and GaBi both allow users to view life cycle inventories (LCI) down to a
basic substance level, covering all the inputs and emissions that are associated with a
system. Both software programs also provide a means to see the contribution each
substance makes to an impact category indicator, and the characterization factors used
to convert inputs and emissions to impact indicators. These capabilities, when used in
conjunction with the Ecoinvent database and basic material models of the earlier tests,
allow common inputs to be traced through to impact indicators, providing a basis for
comparing what is happening in the software. In practice, things are made difficult by
SimaPro and GaBi categorizing inputs in different manners. SimaPro appears to map
Ecoinvent data on a one-to-one basis, while GaBi may combine items. For example, in
one test it was found that SimaPro had 4 varieties of copper ore as input, while GaBi
combined the mass of these 4 ores into a single entry for copper. Other differences in
the way inputs are handled between the two software programs also exist.
SimaPro maps information from the Ecoinvent database into compartments such
as air, raw, soil, and water. Air, soil, and water generally contain emissions, while the

94

“raw” compartment covers raw materials used in production, including things like ore
mined from the ground. Each of these compartments can be divided into subcompartments. For example; groundwater, groundwater long-term, lake, ocean, and
river are all sub-compartments of water. A given emission to water can be assigned to a
sub-compartment, or it can be allocated to the general compartment, in which case it is
referred to as unassigned. A substance or emission can have a unique characterization
factor assigned to it for each compartment and sub-compartment. If a substance does
not have a characterization factor assigned to a sub-compartment, then the software
defaults to using the characterization factor of the substance assigned for the general
compartment. Zero is considered a valid characterization factor for a substance in a
SimaPro sub-compartment, and in this case there is no defaulting to the compartment
value. If a substance has no characterization factors assigned at all, either in the
general compartment or the sub-compartments, then zero is used.
GaBi often combines one or more of the inputs for a substance into a single entry
in the LCI, whereas SimaPro typically places them in separate sub-compartments.
There does not appear to be a great deal of consistency in how this is done; in some
cases all the mass of a substance is combined together, and in other cases it may be
split between two or more categories. For example; an amount of a substance SimaPro
places in the groundwater long-term sub-compartment may end up in GaBi under the
category “ecoinvent long-term to fresh water”, or it may be combined with other
amounts of the same substance and place in a category such as “Heavy metals to fresh
water” or “Inorganic emissions to fresh water”. The only way to accurately compare
inputs and emissions between SimaPro and GaBi is to trace out and match the total

95

amount of a substance for each type of input and emission, and then examine what
characterization factors are used in converting the mass to an impact assessment
indicator. In this way it can be established whether both programs are dealing with the
same amount of input for a given substance, what characterization factors are used with
the substance, and what contribution the substance makes to the impact indicator in
question. This can be time consuming to do, but is a viable method of ensuring a
consistent comparison of the software programs.
A total of 14 impact assessment categories were examined covering 3 different
impact assessment methodologies: Impact 2002+, ReCiPe, and TRACI 2. All
comparisons were done using SimaPro 7.3.3 and GaBi 5. Comparisons between
software were made using pairs of tables, each pair consisting of a table for SimaPro
(Table xxA) and a table for GaBi (Table xxB). Even with the simple basic material
systems being used as the basis for comparison, there can be hundreds, possibly
thousands, of substances included in a life cycle inventory. However, generally only a
small subset are significant to the final impact indicator value, and it is these subsets
that are included in the tables. Item numbers in tables have been arranged so there is a
one-to-one correspondence between tables within a pair; meaning that item 5 in the
SimaPro table corresponds to item 5 in the GaBi table. Any differences in
characterization factors between SimaPro and GaBi are highlighted in bold text in the
tables.

96

11.1 IMPACT 2002+
Up to this point, the implementation of Impact 2002+ version 2.1 being used for
SimaPro was one downloaded via an internet link contained in the methods library
manual for SimaPro (Goedkoop et al. 2008). When SimaPro was upgraded to version
7.3.3, another copy of Impact 2002+ version 2.1 was included. While both files indicated
they were the same version, a check of the comments included in the files found the
one from the software upgrade was last modified in 2010, while the one from the
download link had no changes listed since 2005. After downloading the methods
manual again, a follow up check found that the original file is still the one accessed via
the internet link. Faced with two potentially different variants of Impact 2002+ version
2.1 being available, it was decided to do the tests in this section using both for
completeness. Unless otherwise noted, all SimaPro Impact 2002+ data presented in
comparison tables in this section came from the software upgrade variant, or “new”
Impact 2002+ version 2.1. Any differences found with the original variant, or “old” Impact
2002+ version 2.1, are noted in the discussion of the results accompanying each pair of
tables. All GaBi Impact 2002+ data comes from the Impact 2002+ version 2.1 method
included with GaBi 5.
One issue that does exist for the old variant of Impact 2002+, but not the new
one, is an error in the downloaded installation file. The file contained an inadvertent
blank line in the middle of the data for the terrestrial ecotoxicity impact category. A blank
line represents the end of a impact category in the installation file, so SimaPro stopped
loading impact characterization factors for terrestrial ecotoxicity when it encountered the
blank line. No warning was issued because SimaPro saw this as a proper indication of

97

the end of the category. It was not until investigating differences in results between
software packages that it was discovered that 671 (48.3%) of the 1389 impact
characterization factors for terrestrial ecotoxicity never got installed. The blank line was
removed from the installation file and the old variant of Impact 2002+ was reinstalled to
fix the problem. This problem was still present in the file posted on the download
website at the time this paper was being written.

11.1.1 Aquatic Ecotoxicity
Impact 2002+ data for the aquatic ecotoxicity impact category when obtaining
and disposing of 1 kg of aluminum with 50% recycled content is given in Table 16A for
SimaPro, and Table 16B for GaBi. The first thing to note at the bottom of the tables is
that SimaPro reported 2,385 kg TEG (Tri Ethylene Glycol) to water versus 1,229,700 kg
TEG to water reported by GaBi, meaning the two software programs differed by a
multiplication factor of 516. The next thing to note at the bottom of the tables is that the
19 items in each table account for 2,357 kg TEG to water for SimaPro and 1,229,643 kg
TEG to water for GaBi, or in simpler terms 98.8% of the total impact reported by
SimaPro and effectively 100% for GaBi. This establishes that SimaPro and GaBi
differed significantly from each other for this impact category, and that the 19 items
contained in each table were the significant contributors to the impact category results.
Looking at the far right column in the two tables, the total contribution each item
made to the impact indicator value can be seen. Item 1, aluminum (U.S. spelling) or
“aluminium” (English spelling), clearly had a significant difference in impact contribution
between SimaPro and GaBi. This was caused by the Impact 2002+ implementation in

98

SimaPro setting the characterization factor for aluminum in the “groundwater, long term”
sub-compartment (item 1A) to zero, effectively removing it from making any
contribution. The Impact 2002+ implementation in SimaPro did have a characterization
factor for aluminum in the “river” sub-compartment (item 1B), the same factor that GaBi
used for “aluminum (+III)”. The total mass of aluminum for items 1A and 1B in SimaPro
(Table 16A) matched the mass GaBi had for item 1 (Table 16B), showing that total
aluminum emissions to water were identical for the two programs. It should also be
noted that GaBi assigned all the aluminum mass to the category “Inorganic emissions to
fresh water”, even though GaBi did use the category “ecoinvent long-term to fresh
water” for other substances.
While item 1 was clearly the dominant issue in this comparison of SimaPro and
GaBi, it was not the only problem. Items 2 through 11A had characterization factors of
zero in SimaPro and non zero in GaBi. Examination of the method file in SimaPro found
that all the characterization factors for substances assigned to the “groundwater” and
“groundwater,

long-term”

sub-compartments

were

set

to

zero.

When

the

characterization factor for an impact category is set to zero, a substance makes no
contribution to the reported impact indicator value. While these items were not
significant in this comparison due to their limited mass, they could easily be significant
in other comparisons where one or more of them has greater emissions to water.
The roles were reversed for items 12, 13, 16, 17, and 18. These items had nonzero characterization factors in SimaPro, but did not contribute to the impact reported by
GaBi, even though the mass was accounted for. A check of the Impact 2002+
implementation in GaBi showed no characterization factors for these substances in the

99

stated categories, which for GaBi means an effective characterization factor of zero.
Again this is a case where any of these items could be significant in other comparisons
where they have greater emissions to water.
The Impact 2002+ method file in SimaPro did include a comment that while
characterization factors for “groundwater” and “groundwater, long-term” were set to zero
for aquatic ecotoxicity and some other impact categories, it did not mean there were no
impacts, just that such characterization factors were not currently available. The GaBi
version of Impact 2002+ referred those seeking information on Impact 2002+ to an
internet link that returned an error when accessed.
There were two significant differences between the new Impact 2002+ variant for
SimaPro, used in Table 16A, and the old Impact 2002+ variant. In the old variant, shown
in Table 16C, the name “aluminium” (English spelling) was used as the substance
name, while “aluminum” (U.S. spelling) was used for the characterization factor.
Because the names differed by a single character, SimaPro could not match the
characterization factor to the input substance, resulting in an effective value of zero for
the characterization factor of items 1B, 13, 16, and 17. In addition, items 2, 3, 6, 7, 9,
10, and 11A had characterization factors in the old SimaPro Impact 2002+ variant that
matched those of GaBi, allowing them to contribute to the reported impact indicator
value, while the new SimaPro Impact 2002+ variant had characterization factors of zero.

100

Table 16A: SimaPro / Impact 2002+ aquatic ecotoxicity for aluminum
50% Recycled Content
Characterization
Substance
Amount
Item
Factor
(kg)
Name
Compartment
Sub-compartment
(kg TEG water)
1A
Aluminium
Water
groundwater, long-term
0.34044
0
1B
Aluminium
Water
river
0.00052
3.60E+06
Total Aluminum In Water: 0.34096 kg
2
Antimony
Water
groundwater, long-term
4.68E-06
0
3
Barium
Water
groundwater, long-term
8.74E-05
0
4
Cobalt
Water
groundwater, long-term
6.44E-05
0
5
Copper, ion
Water
groundwater, long-term
8.73E-05
0
6
Lead
Water
groundwater, long-term
1.82E-05
0
7
Mercury
Water
groundwater, long-term
6.05E-07
0
8
Nickel
Water
groundwater, long-term
2.10E-04
0
9
Selenium
Water
groundwater, long-term
1.37E-05
0
10
Zinc, ion
Water
groundwater, long-term
5.76E-04
0
11A
Arsenic, ion
Water
groundwater; groundwater, long-term 1.98E-05
0
11B
Arsenic, ion
Water
river
2.13E-05
3.88E+05
Total Arsenic, Ion In Water: 4.10E-5 kg
12
Chromium VI
Water
river
3.31E-05
4.53E+05
13
Aluminium
air
(combined)
4.36E-04
4.93E+05
14
Copper
air
(combined)
9.78E-06
2.94E+06
15
Zinc
air
(combined)
7.92E-05
2.04E+05
16
Aluminium
soil
agricultural
6.21E-07
3.50E+06
17
soil
industrial
1.89E-05
3.50E+06
Aluminium
18
Chromium VI
soil
(combined)
9.60E-06
4.49E+05
19
Copper
soil
(combined)
6.00E-06
2.04E+07
Total kg TEG Water For Items In Table
Reported kg TEG Water
1

Items in bold text differ in characterization factors between SimaPro and GaBi.
101

Impact
(kg TEG
water)
0
1879
0
0
0
0
0
0
0
0
0
0
8.25
15.00
215
28.76
16.15
2.17
66.20
4.31
123
2357
2385

Table 16B: GaBi / Impact 2002+ aquatic ecotoxicity for aluminum
50% Recycled Content
Characterization
Substance
Amount
Item
Factor (kg TEG to
(kg)
water per kg)
Category
Name
1
Inorganic emissions to fresh water
Aluminum (+III)
0.34096
3.60E+06
2
ecoinvent long-term to fresh water
Antimony
4.68E-06
2.10E+06
3
ecoinvent long-term to fresh water
Barium
8.74E-05
8.05E+04
4
ecoinvent long-term to fresh water
Cobalt
6.44E-05
3.86E+06
5
ecoinvent long-term to fresh water
Copper (+II)
8.73E-05
2.06E+07
6
ecoinvent long-term to fresh water
Lead (+II)
1.82E-05
2.64E+05
7
Heavy metals to fresh water
Mercury (+II)
6.22E-07
1.58E+07
8
Heavy metals to fresh water
Nickel
2.11E-04
1.27E+06
9
ecoinvent long-term to fresh water
Selenium
1.37E-05
3.40E+06
10
ecoinvent long-term to fresh water
Zinc (+II)
5.76E-04
1.40E+06
11
Heavy metals to fresh water
Arsenic (+V)
4.10E-05
3.88E+05
12
Heavy metals to fresh water
Chromium (+VI) 3.37E-05
0
13
Particles to air
Aluminum
4.36E-04
0
14
Heavy metals to air
Copper (+II)
9.78E-06
2.94E+06
15
Heavy metals to air
Zinc (+II)
7.92E-05
2.04E+05
16 Inorganic emissions to agricultural soil
Aluminum
6.21E-07
0
17
Inorganic emissions to industrial soil
Aluminum
1.89E-05
0
18
Heavy metals to industrial soil
Chromium (VI)
9.60E-06
0
19
Heavy metals to industrial soil
Copper (+II)
6.00E-06
2.04E+07
Total kg TEG To Water For Items In Table
Reported kg TEG To Water
1

Items in bold text differ in characterization factors between SimaPro and GaBi.

102

Impact
(kg TEG
to water)
1,226,263
9.84
7.04
249
1,794
4.79
9.82
268
46.52
808
15.93
0
0
28.76
16.15
0
0
0
123
1,229,643
1,229,700

Item
1A
1B
2
3
4
5
6
7
8
9
10
11A
11B
12
13
14
15
16

Table 16C: SimaPro / Old Impact 2002+ variant aquatic ecotoxicity for aluminum
50% Recycled Content
Characterization
Substance
Amount
Factor
(kg)
Name
Compartment
Sub-compartment
(kg TEG water)
Aluminium
Water
groundwater, long-term
0.34044
0
Aluminium
Water
river
0.00052
3.60E+06
Total Aluminium In Water: 0.34096 kg
Antimony
Water
groundwater, long-term
4.68E-06
2.10E+06
Barium
Water
groundwater, long-term
8.74E-05
8.05E+04
Cobalt
Water
groundwater, long-term
6.44E-05
0
Copper, ion
Water
groundwater, long-term
8.73E-05
0
Lead
Water
groundwater, long-term
1.82E-05
2.64E+05
Mercury
Water
groundwater, long-term
6.05E-07
1.58E+07
Nickel
Water
groundwater, long-term
2.10E-04
0
Selenium
Water
groundwater, long-term
1.37E-05
3.40E+06
Zinc, ion
Water
groundwater, long-term
5.76E-04
1.40E+06
Arsenic, ion
Water
groundwater; groundwater, long-term 1.98E-05
3.88E+05
Arsenic, ion
Water
river
2.13E-05
3.88E+05
Total Arsenic, Ion In Water: 4.10E-5 kg
Chromium VI
Water
river
3.31E-05
4.53E+05
Aluminium
air
(combined)
4.36E-04
4.93E+05
Copper
air
(combined)
9.78E-06
2.94E+06
Zinc
air
(combined)
7.92E-05
2.04E+05
soil
agricultural
6.21E-07
3.50E+06
Aluminium

17
soil
Aluminium
18
Chromium VI
soil
19
Copper
soil
Total kg TEG Water For Items In Table
Reported kg TEG Water

industrial
(combined)
(combined)

103

1.89E-05
9.60E-06
6.00E-06

3.50E+06
4.49E+05
2.04E+07

Impact
(kg TEG
water)
0
2
0
9.84
7.04
0
0
4.79
9.56
0
46.52
808
7.67
8.25
15.00
2
0
28.76
16.15
0

2
2

0
4.31
123
1089
1217

Table 16C (cont’d)
1
2

Items in bold text differ in characterization factors between old and new variants of SimaPro.
Difference between substance and characterization factor names caused a zero.

104

11.1.2 Global Warming
Impact 2002+ data for the global warming impact category when obtaining and
disposing of 1 kg of corrugated board with 12% recycled content is shown in Table 17A
for SimaPro, and Table 17B for GaBi. The tables show that SimaPro reported 1.008 kg
CO2 equivalent while GaBi reported only 0.140 kg CO2 equivalent, meaning the two
software programs differ by a factor of 7.20. The 9 items in the table account for 1.006
kg CO2 equivalent of the impact reported by SimaPro, or 99.8%, and 0.137 kg CO2
equivalent of what GaBi reported, or 97.7%, again giving good coverage of the
significant contributors to the impact category.
The most significant difference occurred with item 3 in the tables. GaBi effectively
gave a credit for the sequestration of carbon in the biomass used to produce the
corrugated board, and SimaPro did not. The second difference between the programs is
item 1, where GaBi included biotic carbon dioxide while SimaPro did not. GaBi also
included biotic carbon monoxide in item 5 while SimaPro did not. GaBi and SimaPro
both included variants of methane in items 8 and 9, but with different characterization
factors.
The old variant of SimaPro Impact 2002+ had different characterization factors
for items 4, 8, and 9. “Carbon dioxide, land transformation” (item 4) and “methane,
biogenic” (item 8) both had zero as characterization factors in the old variant. “Methane,
fossil” (item 9) had a characterization factor in the old variant of 7.00, the same as used
by GaBi. Checking comments in the new variant SimaPro Impact 2002+ method file, a
statement was found that characterization factors for “methane, biogenic” and
“methane, fossil” were changed to their current values in 2010. There was also a

105

comment that “carbon dioxide, land transformation” was included in 2008. There were
no comments available with the GaBi implementation of Impact 2002+.

Table 17A: SimaPro / Impact 2002+ global warming for corrugated board, part 1
12% Recycled Content
Compartment

Amount
(kg)

Characterization
Factor
(kg CO2 per kg)

Impact
(kg CO2
Equiv.)

air

1.48E+00

0

0

air

9.15E-01

1.00

0.91490

raw

2.35E+00

0

0

air

4.13E-04

1.00

0.00041

air

8.06E-05

0

0

air

4.41E-03

1.57

0.00693

air

5.70E-05

156

0.00890

Substance
Item
1
2
3
4
5
6
7

Name
Carbon dioxide,
biogenic
Carbon dioxide,
fossil
Carbon dioxide, in
air
Carbon dioxide,
land transformation
Carbon monoxide,
biogenic
Carbon monoxide,
fossil
Dinitrogen
monoxide

8

Methane, biogenic

air

7.19E-03

7.60

0.05465

9

Methane, fossil

air

1.92E-03

10.35

0.01984

Total kg CO2 For Items In Table

1.006

Reported kg CO2

1.008

1

Items in bold text differ in characterization factors between SimaPro and GaBi.

106

Table 17B: GaBi / Impact 2002+ global warming for corrugated board, part 1
12% Recycled Content
Substance
Item
1
2
3

Characterization
Factor
(kg CO2 per kg)

Impact
(kg CO2
Equiv.)

Category

Name

Inorganic
emissions to air
Inorganic
emissions to air
Renewable
resources

Carbon dioxide
(biotic)

1.50E+00

1.00

1.50000

Carbon dioxide

9.15E-01

1.00

0.91500

Carbon dioxide

2.35E+00

-1.00

-2.35000

4.13E-04

1.00

0.00041

8.49E-05

1.57

0.00013

4.41E-03

1.57

0.00692

5.72E-05

156

0.00892

6.00E-03

7.00

0.04200

1.92E-03

7.00

0.01344

4

Inorganic
emissions to air

5

Inorganic
emissions to air

Carbon dioxide,
land
transformation
Carbon
monoxide
(biotic)
Carbon
monoxide
Nitrous oxides
(laughing gas)

Inorganic
emissions to air
Inorganic
7
emissions to air
Organic
8
emissions to air Methane (biotic)
(group VOC)
Organic
9
emissions to air
Methane
(group VOC)
Total kg CO2 For Items In Table
6

0.137
0.140

Reported kg CO2
1

Amount
(kg)

Items in bold text differ in characterization factors between SimaPro and GaBi.

In order to look at how the differences between SimaPro and GaBi influence
reported impacts when recycled content is varied, the Impact 2002+ global warming
data for creating and disposing of 1 kg of corrugated board with 87% recycled content is
given in Table 18A for SimaPro, and Table 18B for GaBi. SimaPro and GaBi came
close to reporting the same impact value, 1.007 kg CO2 equivalent for SimaPro and

107

1.026 kg CO2 equivalent for GaBi. This was due primarily to items 1 and 3 in GaBi,
where the amount of biotic carbon dioxide and the amount of sequestered carbon from
carbon dioxide in the air both decreased with the increase in recycled content. At 12%
recycled content these two items combined for 1.50 - 2.35 = -0.85 kg CO2 equivalent,
but at 87% recycled content the combined value decreased to 0.468 – 0.429 = -0.039
kg CO2 equivalent, so the net carbon credit given by GaBi for using virgin fiber was
reduced, bring the results reported by the two programs into better agreement.

Table 18A: SimaPro / Impact 2002+ global warming for corrugated board, part 2
87% Recycled Content
Substance
Item
1
2
3
4
5
6
7
8
9

Name

Compartment
air

4.41E-01

0

0

air

9.17E-01

1.00

0.91664

raw

4.29E-01

0

0

air

4.27E-04

1.00

0.00043

air

4.14E-05

0

0

air

2.47E-03

1.57

0.00388

air

6.46E-05

156

0.01007

air
air

7.07E-03
1.98E-03

7.60
10.35

0.05373
0.02050
1.005

Carbon dioxide,
biogenic
Carbon dioxide,
fossil
Carbon dixoide, in
air
Carbon dioxide,
land transformation
Carbon monoxide,
biogenic
Carbon monoxide,
fossil
Dinitrogen
monoxide
Methane, biogenic
Methane, fossil

Total kg CO2 For Items In Table

1.007

Reported kg CO2
1

Characterization Impact
Factor
(kg CO2
(kg CO2 per kg) Equiv.)

Amount
(kg)

Items in bold text differ in characterization factors between SimaPro and GaBi.
108

Table 18B: GaBi / Impact 2002+ global warming for corrugated board, part 2
87% Recycled Content
Substance
Item
1
2
3

Characterization
Factor
(kg CO2 per kg)

Impact
(kg CO2
Equiv.)

Category

Name

Inorganic
emissions to air
Inorganic
emissions to air
Renewable
resources

Carbon dioxide
(biotic)

4.68E-01

1.00

0.46800

Carbon dioxide

9.17E-01

1.00

0.91700

Carbon dioxide

4.29E-01

-1.00

-0.42900

4.27E-04

1.00

0.00043

4.57E-05

1.57

0.00007

2.47E-03

1.57

0.00388

6.47E-05

156

0.01009

5.88E-03

7.00

0.04116

1.98E-03

7.00

0.01386

4

Inorganic
emissions to air

5

Inorganic
emissions to air

Carbon dioxide,
land
transformation
Carbon
monoxide
(biotic)
Carbon
monoxide
Nitrous oxides
(laughing gas)

Inorganic
emissions to air
Inorganic
7
emissions to air
Organic
8
emissions to air Methane (biotic)
(group VOC)
Organic
9
emissions to air
Methane
(group VOC)
Total kg CO2 For Items In Table
6

1.025
1.026

Reported kg CO2
1

Amount
(kg)

Items in bold text differ in characterization factors between SimaPro and GaBi.

109

11.1.3 Mineral Extraction
Impact 2002+ data for the mineral extraction impact category when obtaining and
disposing of 1 kg of glass with 55.5% recycled content is given in Table 19A for
SimaPro, and Table 19B for GaBi. SimaPro reported 0.00645 MJ equivalent while GaBi
reported 0.0523 MJ equivalent, differing by a factor of 8.11. The 11 items in each table
combine for effectively the entire impact. There are several issues causing the
difference between reported values from the software; each has the potential of being
significant in comparisons that include larger amounts of the substances involved.
The most significant contribution to the difference in this comparison came from
item 9. The characterization factor for nickel ore in SimaPro was 16.32 MJ per kg, but
GaBi used 163 MJ per kg, a factor of 10 greater. According to literature, mineral
extraction characterization factors for Impact 2002+ were taken from the Eco-indicator
99 assessment method (Jolliet et al. 2003). A check of Eco-indicator 99 in GaBi
revealed a characterization factor for nickel ore of 16.32 MJ per kg, matching the factor
SimaPro had for Impact 2002+, making it appear the factor used for Impact 2002+ in
GaBi may have been entered incorrectly. Interestingly, a check of Eco-indicator 99 in
SimaPro found the characterization factor for nickel ore to be 23.75 MJ per kg, which
matched the factor found in Eco-indicator 99 literature (Goedkoop and Spriensma
2001), but not anything else.
Another problem dealing with nickel ore in SimaPro occurred in item 9A of Table
19A. Here the substance name differed somewhat from the closest name assigned a
non zero characterization factor, resulting in a factor of zero being used. See Table 20
for the difference in names.

110

Item 4 in the Table 19 A and B was the next problem. SimaPro distributed
copper ore among 4 types, but had a characterization factor assigned only to the type in
item 4A; the remaining 3 types of copper ore did not contribute to the reported impact.
GaBi combined all the copper under a single entry and applied a characterization factor
that matched what SimaPro had for the more general “copper, in ground” substance
name.
A similar problem occurred with molybdenum ore in item 8. Here SimaPro
distributed the ore among 5 types, but had no effective characterization factor. The
closest SimaPro came was for the ore type in item 8E, but again a mismatch of
characterization and substance names caused a factor of zero to be used. Table 20
shows the difference in names. GaBi combined all the molybdenum into one entry and
applied a characterization factor that matches what SimaPro had for the more general
“molybdenum, in ground” substance name.
Item 11 of Tables 19 A and B was another case of SimaPro not matching
substance names to characterization factor names, this time for zinc ore, with the result
that a factor of zero was used (see Table 20). The characterization factor that GaBi
applied did not match any for zinc in the new variant of SimaPro Impact 2002+, but did
match what the old variant had for “zinc (in ore)”.
Item 6 in Tables 19 A and B was a case of SimaPro not having a characterization
factor for any specific type of lead ore that came close to matching the substance name.
The closest SimaPro came was the more general “lead, in ground” which had a factor of
7.35, matching what GaBi used.

111

Item 3 in Tables 19A and B shows SimaPro had a characterization factor for
cinnabar while GaBi did not. A check of a spreadsheet version of Impact 2002+
obtained from the same internet link that the old variant came from did not have an
entry for cinnabar, yet the old variant of Impact 2002+ had the same entry as the new
variant.
One additional naming difference was found between the old and new variants of
SimaPro Impact 2002+ version 2.1. When dealing with chromium ore, item 2 in Table
19A, the old variant left out a percent sign in the characterization factor name, causing
another mismatch that resulted in a factor of zero being used. See Table 20 for the
difference in names.
Table 19A: SimaPro / Impact 2002+ mineral extraction for glass
55.5% Recycled Content
Item

1
2
3
4A

4B

4C

4D

Substance
2
Name
Aluminium, 24% in bauxite, 11% in
crude ore, in ground
Chromium, 25.5% in chromite, 11.6%
in crude ore, in ground
Cinnabar, in ground
Copper, 0.99% in sulfide, Cu 0.36%
and Mo 8.2E-3% in crude ore, in
ground
Copper, 1.18% in sulfide, Cu 0.39%
and Mo 8.2E-3% in crude ore, in
ground
Copper, 1.42% in sulfide, Cu 0.81%
and Mo 8.2E-3% in crude ore, in
ground
Copper, 2.19% in sulfide, Cu 1.83%
and Mo 8.2E-3% in crude ore, in
ground
Total Copper Ore: 2.46E-04 kg
112

Amount
(kg)

Characteri
zation
Factor
(MJ per kg)

Impact
(MJ Equiv.)

4.63E-04

2.38

0.001102

9.39E-05

0.9165

0.000086

1.00E-08

165.5

0.000002

1.61E-05

36.80

0.000593

8.88E-05

0

0

2.36E-05

0

0

1.17E-04

0

0

Table 19A (cont’d)
5
6
7
8A

8B

8C

8D

8E

9A
9B

Iron, 46% in ore, 25% in crude ore, in
ground
Lead, 5.0% in sulfide, Pb 3.0%, Zn, Ag,
Cd, In, in ground
Manganese, 35.7% in sedimentary
deposit, 14.2% in crude ore, in ground
Molybdenum, 0.010% in sulfide, Mo
8.2E-3% and Cu 1.83% in crude ore, in
ground
Molybdenum, 0.014% in sulfide, Mo
8.2E-3% and Cu 0.81% in crude ore, in
ground
Molybdenum, 0.022% in sulfide, Mo
8.2E-3% and Cu 0.36% in crude ore, in
ground
Molybdenum, 0.025% in sulfide, Mo
8.2E-3% and Cu 0.39% in crude ore, in
ground
Molybdenum, 0.11% in sulfide, Mo
4.1E-2% and Cu 0.36% in crude ore, in
ground
Total Molybdenum Ore: 4.26E-06
Nickel, 1.13% in sulfide, Ni 0.76% and
Cu 0.76% in crude ore, in ground
Nickel, 1.98% in silicates, 1.04% in
crude ore, in ground
Total Nickel Ore: 2.52E-04 kg

Tin, 79% in cassiterite, 0.1% in crude
ore, in ground
Zinc, 9.0% in sulfide, Zn 5.3%, Pb, Ag,
11
Cd, In, in ground
Total MJ Equivalent For Items In Table
Reported MJ Equivalent
10

1
2
3

5.76E-03

0.051

0.000294

6.41E-06

0

0

1.91E-05

0.313

0.000006

2.18E-06

0

0

3.09E-07

0

0

2.11E-07

0

0

1.13E-06

0

0

4.22E-07

37.14

0

1.28E-06

16.32

0

2.50E-04

16.32

0.004086

4.73E-07

600.00

0.000284

1.39E-04

3.8367

0

3

3

3

0.00645
0.00645

Items in bold text differ in characterization factors between SimaPro and GaBi.
Substances are from the SimaPro compartment "Raw".
Difference between substance and characterization factor names caused a zero.

113

Table 19B: GaBi / Impact 2002+ mineral extraction for glass
55.5% Recycled Content
Item

Substance
2
Name

Amount
(kg)

Characterization Factor
(MJ per kg)

Impact
MJ Equiv.)

2.38
0.917
0
36.7
0.051
7.35
0.313
41.0
163
600
1.89

0.001102
0.000086
0
0.009023
0.000294
0.000047
0.000006
0.000175
0.041015
0.000284
0.000263
0.0523
0.0523

1
Aluminum
4.63E-04
2
Chromium
9.39E-05
3
Cinnabar
1.00E-08
4
Copper
2.46E-04
5
Iron
5.76E-03
6
Lead
6.41E-06
7
Manganese
1.91E-05
8
Molybdenum
4.26E-06
9
Nickel
2.52E-04
10
Tin
4.73E-07
11
Zinc
1.39E-04
Total MJ Equivalent For Items In Table
Reported MJ Equivalent
1
2

Items in bold text differ in characterization factors between SimaPro and GaBi.
Substances are from the GaBi category "Non renewable elements".

Table 20: SimaPro naming differences for mineral extraction
Substance
Name

Characterization
Factor Name

Impact 2002+
Variant Affected

Chromium, 25.5% in chromite,
11.6% in crude ore, in ground

Chromium, 25.5 in chromite,
11.6% in crude ore, in ground

Old

Molybdenum, 0.11% in
sulfide, Mo 4.1E-2% and Cu
0.36% in crude ore, in ground
Nickel, 1.13% in sulfide, Ni
0.76% and Cu 0.76% in crude
ore, in ground
Zinc, 9.0% in sulfide, Zn 5.3%,
Pb, Ag, Cd, In, in ground

Molybdenum, 0.11% in
sulfide, Mo 0.41% and Cu
0.36% in crude ore, in ground

Old and New

Nickel, 1.13% in sulfides,
0.76% in crude ore, in ground

Old and New

Zinc 9%, Lead 5%, in sulfide
in ground

Old and New

1

Differences between names are highlighted in bold text.

114

11.1.4 Ozone Layer Depletion
In earlier work for this paper using the old SimaPro Impact 2002+ variant, Table
15 (section 10.3) showed GaBi reported an impact for ozone layer depletion that was
different from what SimaPro had by more than a factor of 10 when obtaining and
disposing of 1kg of PET with 0% recycled content. Data in Tables 21A and 21B using
the new SimaPro Impact 2002+ variant show both software reported 1.448E-07 kg
CFC_11 equivalent. This difference between variants was investigated. It was found
that for items 4, 5, and 6 in the tables, the old variant of SimaPro Impact 2002+ had a
single character difference in the names of characterization factors versus names of
substances. In each case a hyphen (“-“) was substituted for a space in the
characterization factor name between “Halon” and the number at the end of the name.
This mismatch led to factors of zero being used, which had the effect of eliminating
most of the impact from being reported. Table 22 shows the difference in names.

115

Table 21A: SimaPro / Impact 2002+ ozone layer depletion for PET
0% Recycled Content
Item

Substance
1
Name

Amount
(kg)

Characterization
Factor
(kg CFC-11 per kg)

Impact
(kg CFC-11
Equiv.)

0.120

2.773E-13

0.730

7.364E-10

0.0200

1.315E-12

6.00

7.409E-08

12.00

5.935E-08

0.380

9.371E-18

1.000

1.012E-12

1.000

9.432E-11

0.940

7.677E-09

1.000

2.898E-10

0.0400

2.495E-14

0.0500

2.537E-09

Ethane, 1,1,1-trichloro-,
2.31E-12
HCFC-140
Methane, tetrachloro-, CFC2
1.01E-09
10
3
Methane, monochloro-, R-40
6.58E-11
Methane,
4
bromochlorodifluoro-, Halon
1.23E-08
1211
Methane, bromotrifluoro-,
5
4.95E-09
Halon 1301
6
Methane, bromo-, Halon 1001 2.47E-17
Methane, trichlorofluoro-,
7
1.01E-12
CFC-11
Ethane, 1,1,2-trichloro-1,2,28
9.43E-11
trifluoro-, CFC-113
Ethane, 1,2-dichloro-1,1,2,29
8.17E-09
tetrafluoro-, CFC-114
Methane, dichlorodifluoro-,
10
2.90E-10
CFC-12
Methane, dichlorofluoro-,
11
6.24E-13
HCFC-21
Methane, chlorodifluoro-,
12
5.07E-08
HCFC-22
Total kg CFC-11 Equivalent For Items In Table
Reported kg CFC-11 Equivalent
1

1

Substances are from the SimaPro compartment "Air".

116

1.448E-07
1.448E-07

Table 21B: GaBi / Impact 2002+ ozone layer depletion for PET
0% Recycled Content
Item

Substance
1
Name

Amount
(kg)

Characterization
Factor
(kg CFC-11 per kg)

Impact
(kg CFC11 Equiv.)

0.120

2.773E-13

0.730

7.364E-10

0.0200

1.315E-12

6.00
12.00
0.380
1.000
1.000

7.408E-08
5.935E-08
9.371E-18
1.012E-12
9.432E-11

0.940

7.677E-09

1.000
0.0400
0.0500

2.898E-10
2.495E-14
2.537E-09
1.448E-07
1.448E-07

1

1,1,1-Trichloroethane
2.31E-12
Carbon tetrachloride
2
1.01E-09
(tetrachloromethane)
Chloromethane (methyl
3
6.58E-11
chloride)
4
Halon (1211)
1.23E-08
5
Halon (1301)
4.95E-09
6
Methyl bromide
2.47E-17
7
R 11 (trichlorofluoromethane)
1.01E-12
8
R 113 (trichlorofluoroethane)
9.43E-11
R 114
9
8.17E-09
(dichlorotetrafluoroethane)
10
R 12 (dichlorodifluoromethane) 2.90E-10
11
R 21 (Dichlorofluoromethane)
6.24E-13
12
R 22 (chlorodifluoromethane)
5.07E-08
Total kg CFC-11 Equivalent For Items In Table
Reported kg CFC-11 Equivalent
1

Substances are from GaBi category "Organic emissions to air (group VOC)".

Table 22: SimaPro naming differences for ozone depletion
Substance
Name

Characterization
Factor Name

Impact 2002+
Variant Affected

Methane, bromochlorodifluoro-,
Halon 1211

Methane, bromochlorodifluoro-,
Halon-1211

Old

Methane, bromotrifluoro-,
Halon 1301

Methane, bromotrifluoro-,
Halon-1301

Old

Methane, bromo-, Halon 1001

Methane, bromo-, Halon-1001

Old

1

Differences (spaces versus hyphens) between names are highlighted in bold text.

117

11.1.5 Respiratory Organics
Simapro referred to this impact category as “Respiratory Organics”; however,
GaBi referred to it as “Photochemical Oxidation”. In earlier work for this paper using the
old SimaPro Impact 2002+ variant, Table 15 (section 10.3) showed GaBi reported an
impact for photochemical oxidation/respiratory organics that was different from what
SimaPro had by more than a factor of 10 when obtaining and disposing of 1 kg of PET
with 0% recycled content. The data in Tables 23A and 23B using the new SimaPro
Impact 2002+ variant show a much smaller difference, 0.00185 kg C2H4 equivalent for
SimaPro and 0.00192 kg C2H4 equivalent for GaBi. Investigating the difference
between the two SimaPro Impact 2002+ variants uncovered two more cases of
mismatched names for characterization factors and substances, item 3 for “aldehydes,
unspecified” and item 16 for “NMVOC, non-methane volatile organic compounds,
unspecified origin”. One additional mismatch of names was discovered that is common
between the variants, item 13 for “methane, fossil”. In all 3 cases the mismatch caused
a characterization factor of zero to be used, resulting in no contribution to reported
impact. Item 16 was the dominant impact contributor, causing most of the difference
between Impact 2002+ variants in SimaPro. Item 13 accounted for most of the
difference between the new SimaPro Impact 2002+ variant (Table 23A) and GaBi
(Table 23B). See Table 24 for the difference in names.
Tables 23A and 23B show that for items 3, 5, and 6 GaBi had zero for the
characterization factors while SimaPro had non-zero values. The lack of any
contribution from these 3 items to the impact reported by GaBi accounts for the
remainder of the difference between SimaPro and GaBi. Item 5 is interesting in that
118

while GaBi assigned the mass to “Butane” which had no characterization factor in GaBi
for this impact category, GaBi did have a characterization factor assigned to “Butane (nbutane) [Group NMVOC to air]” which matched the factor SimaPro used.

Table 23A: SimaPro / Impact 2002+ respiratory organics for PET
0% Recycled Content
Item

Substance
2
Name

Amount
(kg)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Acetaldehyde
1.12E-06
Acetic acid
2.22E-04
Aldehydes, unspecified
1.40E-08
Benzene
9.77E-06
Butane
1.64E-05
Ethane
5.01E-05
Ethene
6.41E-05
Ethyl acetate
3.56E-06
Formaldehyde
4.07E-06
Heptane
1.83E-06
Hexane
6.02E-06
Hydrocarbons, aromatic
3.59E-04
Methane, fossil
1.36E-02
Methanol
1.24E-04
Methyl ethyl ketone
3.56E-06
NMVOC, non-methane volatile
16 organic compounds,
2.24E-03
unspecified origin
17 Pentane
2.14E-05
18 Propane
2.42E-05
19 Propene
2.67E-06
20 Toluene
4.81E-06
21 Xylene
6.47E-06
Total kg C2H4 Equivalent For Items In Table
Reported C2H4 Equivalent
1
2
3

Characterization
Factor
(kg C2H4 per
kg)
0.638
0.100
0.657
0.220
0.355
0.124
1.000
0.216
0.521
0.521
0.479
0.9859
0.00601
0.132
0.380

Impact
(kg C2H4
Equiv.)
0.0000007
0.0000222
9.20E-09
0.0000021
0.0000058
0.0000062
0.0000641
0.0000008
0.0000021
0.0000010
0.0000029
0.0003535
3

0
0.0000163
0.0000014

0.601

0.0013450

0.400
0.180
1.117
0.638
1.038

0.0000085
0.0000043
0.0000030
0.0000031
0.0000067
0.001850
0.001855

Items in bold text differ in characterization factors between SimaPro and GaBi.
All substances are from the SimaPro compartment "Air".
Difference in substance and characterization factor names cause a zero.
119

Table 23B: GaBi / Impact 2002+ photochemical oxidation for PET
0% Recycled Content
Characterization
Substance
Amount
Factor
Item
2
(kg)
Name
(kg C2H4 per kg)
1
Acetaldehyde (Ethanal)
1.12E-06
0.638
2
Acetic acid
2.22E-04
0.100
3
Aldehyde (unspecified)
1.40E-08
0
4
Benzene
9.77E-06
0.220
5
Butane
1.64E-05
0
6
Ethane
5.01E-05
0
7
Ethene (ethylene)
6.41E-05
1.000
Ethylene acetate (ethyl
8
3.56E-06
0.216
acetate)
Formaldehyde
9
4.07E-06
0.521
(methanal)
10 Heptane (isomers)
1.83E-06
0.521
11 Hexane (isomers)
6.02E-06
0.479
Hydrocarbons,
12
3.59E-04
0.986
aromatic
13 Methane
1.36E-02
0.00601
14 Methanol
1.24E-04
0.132
Butanone (methyl ethyl
15
3.56E-06
0.380
ketone)
16 NMVOC (unspecified)
2.24E-03
0.601
17 Pentane (n-pentane)
2.14E-05
0.400
18 Propane
2.42E-05
0.180
19 Propene (propylene)
2.67E-06
1.117
Toluene (methyl
20
4.81E-06
0.638
benzene)
Xylene (dimethyl
21
6.47E-06
1.038
benzene)
Total kg C2H4 Equivalent For Items In Table

2

0.0000008
0.0000021
0.0000010
0.0000029
0.0003535
0.0000816
0.0000163
0.0000014
0.0013450
0.0000085
0.0000043
0.0000030
0.0000031
0.0000067
0.001919
0.001922

Reported C2H4 Equivalent
1

Impact
(kg C2H4
Equiv.)
0.0000007
0.0000222
0
0.0000021
0
0
0.0000641

Items in bold text differ in characterization factors between SimaPro and GaBi.
All substances are from the GaBi category "Organic emissions to air (group
VOC)".

120

Table 24: SimaPro naming differences for respiratory organics
Substance
Name

Characterization
Factor Name

Aldehydes, unspecified
Aldehydes
Methane, fossil
Methane
NMVOC, non-methane volatile organic
NMVOC
compounds, unspecified origin
1

Differences between names are highlighted in bold text.

121

Impact 2002+
Variant Affected
Old
Old and New
Old

11.2 ReCiPe
All comparisons for this section were done using ReCiPe Midpoint (H) version
1.06 with SimaPro 7.3.3, and ReCiPe Midpoint (H) version 1.05 with GaBi 5. These are
the versions of ReCiPe that were supplied with the released versions of SimaPro and
GaBi software available at the time of this study.

11.2.1 Human Toxicity
ReCiPe data for the human toxicity category when obtaining and disposing of 1
kg of PET with 0% recycled content is given in Table 25A for SimaPro, and Table 25B
for GaBi. Amounts listed at the bottom of the tables show that SimaPro reported 1.133
kg 1,4-DB equivalent while GaBi reported 0.3869 kg 1,4-DB equivalent, differing by a
factor of 2.93. Totals for the 20 items in each table show they account for 97.8% of the
impact reported by SimaPro, and 95.7% of the impact reported by GaBi.
The majority of the difference between SimaPro and GaBi for human toxicity had
to do with manganese emissions to water, items 12 and 19. SimaPro had non zero
characterization factors for these items, allowing them to contribute to the reported
impact. GaBi had no impact factors assigned, causing zeros to be used and resulting in
no contribution to reported impact. The characterization factors used by SimaPro are
the same as found in spreadsheet versions of ReCiPe 1.05, used by GaBi, and the
more recent ReCiPe 1.08 files that were downloaded from the ReCiPe website (ReCiPe
2013).
Items 1 through 7 in SimaPro had mass split between two different
characterization factors, while GaBi had only a single entry for each item that accounted

122

for the same amount of mass. Furthermore, GaBi had a characterization factor for only
one of these items, phosphorus (item 6). The remaining 6 emissions to air did not
contribute to the impact reported by GaBi, yet they did contribute in SimaPro. All the
characterization factors used by SimaPro were found to match the spreadsheets for
ReCiPe 1.05 and ReCiPe 1.08.
The last difference comes from item 8, phosphorus emissions to agricultural soil.
GaBi did not have a characterization factor for this item, however, this had to do with the
version of ReCiPe being used. There is no characterization factor for phosphorus
emissions to agricultural soil in the spreadsheet for ReCiPe 1.05, but it is present in the
spreadsheet version of ReCiPe 1.08.

123

Table 25A: SimaPro / ReCiPe human toxicity for PET
0% Recycled Content
Substance
Name

Item
1A
1B
2A
2B
3A
3B
4A
4B
5A
5B
6A
6B
7A

Name

Compartment

Arsenic

air

Sub-compartment(s)

(unspecified), high. pop.
low. pop.; low. pop., longArsenic
air
term
Total arsenic in air: 6.14E-07
Cadmium
air
(unspecified); high. pop.
low. pop.; low. pop., longCadmium
air
term
Total cadmium in air: 2.39E-07 kg
Lead
air
(unspecified); high. pop.
low. pop.; low. pop., longLead
air
term
Total lead in air: 1.91E-06 kg
Mercury
air
(unspecified); high. pop.
low. pop.; low. pop., longMercury
air
term
Total mercury in air: 5.92E-08
Nickel
air
(unspecified); high. pop.
low. pop.; low. pop., longNickel
air
term
Total nickel in air: 9.26E-06 kg
Phosphorus
air
(unspecified); high. pop.
low. pop.; low. pop., longPhosphorus
air
term
Total phosphorus in air: 1.43E-07 kg
Vanadium
air
(unspecified); high. pop.
124

6.23E-08

Characterization
Factor
(kg 1,4-DB eq
per kg)
51,300

3.20E-03

5.51E-07

72,000

3.97E-02

8.27E-08

36,000

2.98E-03

1.57E-07

45,200

7.08E-03

3.36E-07

15800

5.30E-03

1.57E-06

16200

2.55E-02

3.02E-08

518,000

1.56E-02

2.90E-08

56,600

1.64E-03

8.24E-06

439

3.62E-03

1.05E-06

66.9

7.02E-05

1.26E-07

18,800

2.36E-03

1.73E-08

14,800

2.56E-04

6.82E-06

3,490

2.38E-02

Amount
(kg)

Impact
(kg 1,4DB eq)

Table 25A (cont’d)
low. pop.; low. pop., longterm
Total vanadium in air: 6.91E-06 kg
8
Phosphorus
soil
agricultural
9
Phosphorus
soil
industrial
10
Antimony
water
groundwater, long-term
11
Barium
water
groundwater, long-term
12
Manganese
water
groundwater, long-term
13
Molybdenum
water
groundwater, long-term
14
Selenium
water
groundwater, long-term
15
Vanadium
water
groundwater, long-term
16
Zinc, ion
water
groundwater, long-term
(unspecified);
17
Antimony
water
groundwater; river
(unspecified);
groundwater;
18
Arsenic, ion
water
groundwater, long-term,
river
(unspecified);
19
Manganese
water
groundwater; river
20
Selenium
water
groundwater; river
Total kg 1,4-DB eq For Items In Table
Reported kg 1,4-DB eq
7B

1

Vanadium

air

9.10E-08

1,070

9.74E-05

2.43E-07
2.55E-07
4.64E-05
6.57E-05
8.62E-04
5.42E-06
4.85E-06
2.41E-04
2.41E-04

9,440
9,440
574
412
700
1,300
10,600
372
36.2

2.30E-03
2.40E-03
2.66E-02
2.71E-02
6.03E-01
7.05E-03
5.14E-02
8.96E-02
8.71E-03

2.05E-05

574

1.18E-02

9.31E-06

14,900

1.39E-01

5.25E-06

700

3.67E-03

3.76E-07

10,600

3.98E-03
1.108
1.133

Items in bold text differ in characterization factors between SimaPro and GaBi.

125

Table 25B: GaBi / ReCiPe human toxicity for PET
0% Recycled Content
Substance
Item
Category
1
Heavy metals to air
2
Heavy metals to air
3
Heavy metals to air
4
Heavy metals to air
5
Heavy metals to air
6
Inorganic emissions to air
7
Heavy metals to air
8
Inorganic emissions to agricultural soil
9
Inorganic emissions to industrial soil
10
ecoinvent long-term to fresh water
11
ecoinvent long-term to fresh water
12
ecoinvent long-term to fresh water
13
ecoinvent long-term to fresh water
14
ecoinvent long-term to fresh water
15
ecoinvent long-term to fresh water
16
ecoinvent long-term to fresh water
17
Heavy metals to fresh water
18
Heavy metals to fresh water
19
Heavy metals to fresh water
20
Heavy metals to fresh water
Total kg 1,4-DB eq For Items In Table
Reported kg 1,4-DB eq
1

Name
Arsenic (+V)
Cadmium (+II)
Lead (+II)
Mercury (+II)
Nickel (+II)
Phosphorus
Vanadium (+III)
Phosphorus
Phosphorus
Antimony
Barium
Manganese (+II)
Molybdenum
Selenium
Vanadium (+III)
Zinc (+II)
Antimony
Arsenic (+V)
Manganese (+II)
Selenium

Amount
(kg)
6.14E-07
2.39E-07
1.91E-06
5.92E-08
9.29E-06
1.43E-07
6.91E-06
2.43E-07
2.55E-07
4.64E-05
6.57E-05
8.62E-04
5.42E-06
4.85E-06
2.41E-04
2.41E-04
2.05E-05
9.31E-06
5.25E-06
3.76E-07

Characterization
Factor
(kg 1,4-DB eq
per kg)
0
0
0
0
0
18,800
0
0
9,440
574
412
0
1,300
10,600
372
36.2
574
14,900
0
10,600

Items in bold text differ in characterization factors between SimaPro and GaBi.

126

Impact
(kg 1,4-DB eq)
0
0
0
0
0
2.69E-03
0
0
2.41E-03
2.66E-02
2.71E-02
0
7.05E-03
5.14E-02
8.96E-02
8.72E-03
1.18E-02
1.39E-01
0
3.98E-03
0.3701
0.3869

11.2.2 Ionizing Radiation
ReCiPe data for the ionizing radiation category when obtaining and disposing of
1 kg of glass with 55.5% recycled content is given in Table 26A for SimaPro, and Table
26B for GaBi. SimaPro reported 0.1480 kg U235 equivalent while GaBi reported 0.0505
kg U235 equivalent, differing by a factor of 2.93. Totals for the 3 items in each table
show they account for 99.6% of the impact reported by SimaPro, and 99.4% of the
impact reported by GaBi.
The only difference between software for this impact category was item 3, radon222 long-term emissions to air. SimaPro had a characterization factor that matched the
one in the spreadsheets for ReCiPe 1.05 and ReCipe 1.08, GaBi had no
characterization factor assigned. It should be noted that GaBi did assign the same
factor as SimaPro for item 2, radon-222 emissions to air (short-term).

127

Table 26A: SimaPro / ReCiPe ionizing radiation for glass
55.5% Recycled Content
Substance
Item
Name

Compartment

1
Carbon-14
air
2
Radon-222
air
3
Radon-222
air
Total kg U235 Equiv. For Items In Table
Reported kg U235 Equiv.
1

Sub-compartment(s)
low. pop.
low. pop.
low. pop., long-term

Characterization
Amount
Factor
(kBq)
(kg U235 Equiv.
per kBq)
4.79E-03
1.00E+01
2.04E+00
1.14E-03
8.53E+01
1.14E-03

Impact
(kg U235
Equiv.)
4.79E-02
2.33E-03
9.72E-02
0.1474
0.1480

Items in bold text differ in characterization factors between SimaPro and GaBi.

Table 26B: GaBi / ReCiPe ionizing radiation for glass
55.5% Recycled Content
Substance

Amount
(kBq)

Item
Category

Name

1
Radioactive emissions to air
Carbon (C14)
2
Radioactive emissions to air
Radon (Rn222)
3
ecoinvent long-term to air
Radon (Rn222)
Total kg U235 Equiv. For Items In Table
Reported kg U235 Equiv.
1

4.79E-03
2.04E+00
8.53E+01

Characterization
Factor
(kg U235 Equiv. per
kBq)
1.00E+01
1.14E-03
0

Items in bold text differ in characterization factors between SimaPro and GaBi.

128

Impact
(kg U235
Equiv.)
4.79E-02
2.33E-03
0
0.0502
0.0505

11.2.3 Marine Ecotoxicity
ReCiPe data for the marine ecotoxicity category when obtaining and disposing of
1 kg of corrugated board with 50% recycled content is given in Table 27A for SimaPro,
and Table 27B for GaBi. SimaPro reported 0.008375 kg 1,4-DB equivalent while GaBi
reported 0.005806 kg 1,4-DB equivalent, differing by a factor of 1.44. The 18 items in
each table account for 97.4% of the impact reported by SimaPro, and 97.9% of the
impact reported by GaBi.
The majority of the difference between software came from item 9, manganese
emissions to long-term groundwater. SimaPro had a non zero characterization factor for
this item, allowing it to contribute to the reported impact. GaBi had no characterization
factor assigned, causing a zero to be used and resulting in no contribution to reported
impact. The 4.14 kg 1,4-DB equivalent per kg characterization factor used by SimaPro
matched the factor found in the spreadsheet versions of ReCiPe 1.05 and ReCiPe 1.08.
Another source of difference between software was that GaBi did not have
characterization factors for any of the heavy metal emissions to air, items 1 through 4,
so these did not contribute to the impact value reported by GaBi. The impact factors
used by SimaPro matched those in the spreadsheet versions of ReCiPe 1.05 and
ReCiPe 1.08. It should also be noted that SimaPro split the mass of each of these 4
items between two characterization factors; GaBi accounted for the mass of each item
in a single entry per item.
GaBi did not have a characterization factor for item 5, Chlorothalonil emissions to
agricultural soil, resulting in no contribution to the impact reported by GaBi. The impact
factor used by SimaPro matched the one in the spreadsheet versions of ReCiPe 1.05

129

and ReCiPe 1.08. GaBi did have a characterization factor for Chlorothalonil emissions
to industrial soil.
GaBi did not have a characterization factor for item 16, tributyltinoxide emissions
to sea water, resulting in no contribution to the impact reported by GaBi. The impact
factor used by SimaPro matched the one in the spreadsheet version of ReCiPe 1.05 for
tributyltin compounds, and the one in the spreadsheet version of ReCiPe 1.08 for
tributyltin-oxide to ocean.

130

Table 27A: SimaPro / ReCiPe marine ecotoxicity for corrugated board
50% Recycled Content
Substance
Name

Item
1A
1B
2A
2B
3A
3B
4A
4B
5
6
7
8
9
10
11
12

Name

Compartment

Copper

air

Sub-compartment(s)

(unspecified); high. pop.
low. pop.; low. pop.,
Copper
air
long-term
Total copper in air: 5.85E-7 kg
Nickel
air
(unspecified); high. pop.
low. pop.; low. pop.,
Nickel
air
long-term
Total nickel in air: 9.63E-07 kg
Vanadium
air
(unspecified); high. pop.
low. pop.; low. pop.,
Vanadium
air
long-term
Total vanadium in air: 3.02E-06 kg
Zinc
air
(unspecified); high. pop.
low. pop.; low. pop.,
Zinc'
air
long-term
Total zinc in air: 1.16E-06 kg
Chlorothalonil
soil
agricultural
Barium
water
groundwater, long-term
Beryllium
water
groundwater, long-term
Cobalt
water
groundwater, long-term
Manganese
water
groundwater, long-term
Selenium
water
groundwater, long-term
Vanadium, ion
water
groundwater, long-term
Zinc, ion
water
groundwater, long-term
131

3.43E-07

Characterization
Factor
(kg 1,4-DB
Equiv. per kg)
270

2.41E-07

388

9.36E-05

8.00E-07

87.8

7.02E-05

1.63E-07

125

2.04E-05

2.99E-06

71.8

2.15E-04

2.34E-08

102

2.38E-06

5.96E-07

22.4

1.33E-05

5.60E-07

32.3

1.81E-05

1.82E-05
2.69E-05
7.49E-07
9.06E-06
4.47E-04
1.55E-06
3.49E-06
6.63E-05

4.20
2.77
453
33.1
4.14
89.3
94.7
6.01

7.64E-05
7.45E-05
3.39E-04
3.00E-04
1.85E-03
1.38E-04
3.31E-04
3.99E-04

Amount
(kg)

Impact
(kg 1,4-DB
Equiv.)
9.27E-05

Table 27A (cont’d)
13

Arsenic, ion

water

14

Bromine

water

15

Nickel, ion
water
Tributyltin
16
water
compounds
17
Zinc, ion
water
18
Phosphorus
water
Total kg 1,4-DB Equiv. For Items In Table
Reported kg 1,4-DB Equiv.
1

combined
(unspecified);
groundwater; river
combined

2.23E-06

15.1

3.36E-05

3.66E-06

4.65

1.70E-05

3.94E-05

95.9

3.78E-03

ocean

8.42E-09

2060

1.74E-05

ocean
river

1.99E-06
2.11E-05

69.6
6.71

1.38E-04
1.42E-04
0.008158
0.008375

Items in bold text differ in characterization factors between SimaPro and GaBi.

132

Table 27B: GaBi / ReCiPe marine ecotoxicity for corrugated board
50% Recycled Content
Substance

Amount
(kg)

Item
Category
1
Heavy metals to air
2
Heavy metals to air
3
Heavy metals to air
4
Heavy metals to air
5
Pesticides to agricultural soil
6
ecoinvent long-term to fresh water
7
ecoinvent long-term to fresh water
8
ecoinvent long-term to fresh water
9
ecoinvent long-term to fresh water
10
ecoinvent long-term to fresh water
11
ecoinvent long-term to fresh water
12
ecoinvent long-term to fresh water
13
Heavy metals to fresh water
14
Inorganic emissions to fresh water
15
Heavy metals to fresh water
16
Pesticides to sea water
17
Heavy metals to sea water
18
Inorganic emissions to fresh water
Total kg 1,4-DB Equiv. For Items In Table
Reported kg 1,4-DB Equiv.
1

Name
Copper (+II)
Nickel (+II)
Vanadium (+III)
Zinc (+II)
Chlorothalonil
Barium
Beryllium
Cobalt
Manganese (+II)
Selenium
Vanadium (+III)
Zinc (+II)
Arsenic (+V)
Bromine
Nickel (+II)
Tributyltinoxide
Zinc (+II)
Phosphorus

5.85E-07
9.63E-07
3.02E-06
1.16E-06
1.82E-05
2.68E-05
7.45E-07
9.06E-06
4.47E-04
1.58E-06
3.50E-06
6.52E-05
2.22E-06
3.66E-06
3.94E-05
8.42E-09
1.99E-06
2.11E-05

Items in bold text differ in characterization factors between SimaPro and GaBi.

133

Characterization
Factor
(kg 1,4-DB
Equiv. per kg)
0
0
0
0
0
2.77
453
33.1
0
89.3
94.7
6.01
15.1
4.65
95.9
0
69.6
6.71

Impact
(kg 1,4-DB
Equiv.)
0
0
0
0
0
7.44E-05
3.37E-04
3.00E-04
0
1.41E-04
3.31E-04
3.92E-04
3.35E-05
1.70E-05
3.78E-03
0
1.38E-04
1.42E-04
0.005684
0.005806

11.2.4 Marine Eutrophication
ReCiPe data for the marine eutrophication category when obtaining and
disposing of 1 kg of aluminum with 50% recycled content is given in Table 28A for
SimaPro, and Table 28B for GaBi. SimaPro reported 0.001056 kg N equivalent while
GaBi reported 0.004979 kg N equivalent, differing by a factor of 4.71. The 12 items in
each table account for 99.9% of the impact reported by SimaPro, and effectively 100%
of the impact reported by GaBi.
The majority of the difference between software came from the characterization
factors for item 2, nitrogen oxide emissions to air. GaBi used a factor of 0.389, 10 times
the value of 0.039 that SimaPro used. The characterization factor used by SimaPro
matched the one in the spreadsheet versions of ReCiPe 1.05 and ReCiPe 1.08.
The next issue between the software was the characterization factors for items 5
and 11, organic bound nitrogen emissions to long-term groundwater and to ocean
water. SimaPro used characterization factors of 1.00, which matched the factors in the
spreadsheets for ReCiPe 1.05 and ReCiPe 1.08. GaBi had no characterization factor
assigned, so these items did not contribute to the impact reported by GaBi. GaBi did
use a characterization factor of 1.00 for organic bound nitrogen emission to fresh water
(short-term), item 12.
SimaPro used a characterization factor of 0.23 for item 9, nitrate emissions to
ocean (sea) water, which matched the factor in the spreadsheet version of ReCiPe
1.08. The spreadsheet version of ReCiPe 1.05 had a factor of 0.226, making it appear
that at some point between the 1.05 and 1.08 versions the factor was rounded off or
adjusted upward. GaBi did not have a characterization factor assigned for nitrate

134

emissions to sea water; however, it did use the 0.226 factor for nitrate emissions to
fresh water, item 8.
GaBi did not have a characterization factor for item 7, cyanide emissions to fresh
water, which is in agreement with the spreadsheet version of ReCiPe 1.05. SimaPro
used a factor of 0.54, which agrees with the spreadsheet for ReCiPe 1.08.

135

Table 28A: SimaPro / ReCiPe marine eutrophication for aluminum
50% Recycled Content
Substance
Name

Item

2.08E-04

0.092

1.92E-05

1.13E-02

0.039

4.39E-04

1.61E-06
2.21E-03

0.780
0.230

1.26E-06
5.08E-04

2.63E-06

1.00

2.63E-06

9.59E-06

0.780

7.48E-06

2.18E-06
1.10E-04
8.64E-06
2.62E-05

0.540
0.230
0.230
1.00

1.18E-06
2.53E-05
1.99E-06
2.62E-05

ocean

2.15E-06

1.00

2.15E-06

river

2.11E-05

1.00

2.11E-05

Compartment

1

Ammonia

air

2

Nitrogen oxides

air

3
4

Ammonium ion

water
water

(unspecified); high.
pop.; low. pop.
(unspecified); high.
pop.; low. pop.
groundwater, long-term
groundwater, long-term

water

groundwater, long-term

6
7
8
9
10

2

Nitrate
Nitrogen,
organic bound
Ammonium ion,
to water, river
Cyanide
2

water
water
water
water
water

Nitrate
Nitrate
Nitrogen
Nitrogen,
11
water
organic bound
Nitrogen,
12
water
organic bound
Total kg N Equiv. For Items In Table
Reported kg N Equiv.
2

Impact
(kg N
Equiv.)

Name

5

1

Amount
(kg)

Characterization
Factor
(kg N Equiv.
per kg)

Sub-compartment(s)

(unspecified);
groundwater; river
river
groundwater, river
ocean
river

0.001055
0.001056

Items in bold text differ in characterization factors between SimaPro and GaBi.
Characterization factor appears rounded to two digit accuracy versus 3 digit in GaBi.

136

Table 28B: GaBi / ReCiPe marine eutrophication for aluminum
50% Recycled Content
Substance
Item

Amount
(kg)

Characterization
Factor
Impact
(kg N Equiv. per (kg N Equiv.)
kg)
0.0920
1.92E-05

Category

Name

1

Inorganic emissions to air

Ammonia

2.08E-04

2

Inorganic emissions to air

Nitrogen oxides

1.13E-02

0.389

4.38E-03

3

ecoinvent long-term to fresh water

1.61E-06

0.778

1.25E-06

4

ecoinvent long-term to fresh water

2.21E-03

0.226

4.99E-04

5

ecoinvent long-term to fresh water

2.63E-06

0

0

6

Inorganic emissions to fresh water

9.59E-06

0.778

7.46E-06

7
8
9
10

Inorganic emissions to fresh water
Inorganic emissions to fresh water
Inorganic emissions to sea water
Inorganic emissions to fresh water

2.21E-06
1.10E-04
8.64E-06
2.62E-05

0
0.226
0
1.00

0
2.49E-05
0
2.62E-05

11

Inorganic emissions to sea water

2.15E-06

0

0

12

Inorganic emissions to fresh water

2.11E-05

1.00

2.11E-05

Ammonium /
ammonia
Nitrate
Nitrogen organic
bounded
Ammonium /
ammonia
Cyanide
Nitrate
Nitrate
Nitrogen
Nitrogen organic
bounded
Nitrogen organic
bounded

Total kg N Equiv. For Items In Table
Reported kg N Equiv.
1

0.004979
0.004979

Items in bold text differ in characterization factors between SimaPro and GaBi.

137

11.2.5 Water Depletion
ReCiPe data for the water depletion category when obtaining and disposing of 1
kg of aluminum with 50% recycled content is given in Table 29A for SimaPro, and Table
29B for GaBi. SimaPro reported 0.03174 m3 of water while GaBi reported 165.1 m3,
differing by a factor of 5,200. The 7 items in each table account for 100% of the impact
reported by both programs.
The greatest difference between results occurred due to item 6, “water, turbine
use, unspecified natural origin/m3”. Tracing this resource backward through SimaPro, it
appears to be water flowing through electrical power generation turbines. GaBi had a
characterization factor of 1.00 for this item, and as such it is included in the impact
reported by GaBi. SimaPro had a characterization factor of zero, so it did not contribute
to the impact reported by SimaPro. Water going through a turbine had a zero
characterization factor in the spreadsheet for ReCiPe 1.05, the version used with GaBi,
but was changed to a factor of 1.00 in the spreadsheet for ReCiPe 1.08.
SimaPro split the volume of item 1 into two parts, one part with a characterization
factor of 1.00 and the other with a zero factor, while GaBi combined the volume into a
single entry with a characterization facto of 1.00. The spreadsheet for ReCiPe 1.05
agreed with SimaPro; however, the spreadsheet for ReCiPe 1.08 agreed with GaBi.
This seems strange since the GaBi version is stated as being ReCiPe 1.05, while the
SimaPro version is stated as being ReCiPe 1.06.
Items 4 and 5 in the tables had characterization factors of 1.00 for GaBi, and
factors of zero for SimaPro. In both cases the spreadsheets for ReCiPe 1.05 and
ReCiPe 1.08 matched SimaPro.

138

Table 29A: SimaPro / ReCiPe water depletion for aluminum
50% Recycled Content
Substance
2
Name

Item

Water, cooling, unspecified natural
origin/m3
3
Water, unspecified natural origin/m , raw

1A
1B

Total Water, unspecified: 1.31E-01 m
2
Water, lake
3
Water, river
4
Water, salt, ocean
5
Water, salt, sole
Water, turbine use, unspecified natural
6
origin/m3
7
Water, well, in ground, raw
Total m3 For Items In Table
Reported m3
1
2

Amount
3
(m )

Characteriza
tion Factor
3
3
(m per m )

Impact
3
(m )

1.22E-01

0

0

9.47E-03

1.00

0.009472

8.38E-05
1.80E-02
2.37E-03
4.78E-04

1.00
1.00
0
0

0.000084
0.018023
0
0

1.65E+02

0

0

4.16E-03

1.00

0.004157
0.03174
0.03174

3

Items in bold text differ in characterization factors between SimaPro and GaBi.
All substances are from the SimaPro compartment "Raw".
Table 29B: GaBi / ReCiPe water depletion for aluminum
50% Recycled Content

6
7

1.31E-01
8.38E-05
1.80E-02
2.37E-03
4.78E-04

0.131352
0.000084
0.018023
0.002371
0.000478

1.65E+02

1.00

164.9

4.16E-03

1.00

0.004157
165.1

Amount
3
(m )

Water
Water (lake water)
Water (lake river)
Water (sea water)
Water, salt, sole
Water, turbine use, unspecified
natural origin
Water (ground water)

Item
1
2
3
4
5

Characterization
Factor
3
3
(m per m )
1.00
1.00
1.00
1.00
1.00

Substance
2
Name

3

Total m For Items In Table
Reported m
1
2

3

Impact
3
(m )

165.1

Items in bold text differ in characterization factors between SimaPro and GaBi.
All substances are from the GaBi category "Renewable resources".
139

11.3 TRACI 2
All comparisons for TRACI 2 were done using versions of the methodology
supplied with the released versions of SimaPro and GaBi software available at the time.
Comments in the SimaPro 7.3.3 method file for TRACI 2 state a version number of 4.00,
with the last change occurring in 2012. There were no accessible comments or notes for
the GaBi 5 implementation of TRACI 2. A spreadsheet containing the characterization
factors for TRACI 2.1 was obtained for reference purposes from the EPA, the U.S.
governmental agency responsible for the development of the TRACI methodology.

11.3.1 Ecotoxicity
TRACI 2 data for the ecotoxicity category when obtaining and disposing of 1 kg
of glass with 55.5% recycled content is given in Table 30A for SimaPro, and Table 30B
for GaBi. SimaPro reported 0.850 CTUe (Comparative Toxic Units, ecotoxicity potential)
while GaBi reported 4.811 CTUe, differing by a factor of 5.66. The 43 items in each
table account for 98.5% of the impact reported by SimaPro, and 99.8% of the impact
reported by GaBi.
A total of 24 of the 43 items in each table differ significantly in impact contribution
between SimaPro and GaBi. The largest contributors to the difference in reported
results came from items 31, 25, 30, and 39, in order of largest difference first. These
were all metal ion emissions to water: zinc, copper, vanadium, and nickel. In all 4 cases
GaBi had a non zero characterization factor assigned, while SimaPro had no factors
assigned. The result was that these items contributed to the impact reported by GaBi,
but not to the impact reported by SimaPro. For zinc, copper and nickel (items 31, 25,

140

and 39) GaBi matched the characterization factors assigned to the substances in the
spreadsheet version of TRACI 2.1. For vanadium, item 30, GaBi used the
characterization factor for vanadium (V) from the spreadsheet version of TRACI 2.1.
These same relationships between GaBi and the spreadsheet hold true for items 43, 37,
and 42, which were smaller emissions of zinc, copper, and vanadium to other
classifications of water.
While the aforementioned substances account for the large majority of difference
in results between software for this comparison, there are 7 more instances where GaBi
had non zero characterization factors while SimaPro used zeros. These could become
significant in other comparisons as the mix of substances involved shift. For cobalt
emissions to water in item 24, GaBi used the TRACI 2.1 characterization factor for
cobalt (II). For silver ion, or silver, emissions to water in items 27 and 40, GaBi used the
TRACI 2.1 characterization factor of silver (I). For cadmium ion, or cadmium (+II),
emissions to water in item 35, GaBi matched the characterization factor for this
substance in TRACI 2.1. For tin ion, or tin (+IV), emissions to water in items 29 and 41,
GaBi used the TRACI 2.1 characterization factor of tin (II). For tin emissions to
agricultural soil in item 15, GaBi again used the characterization factor of tin (II).
There are 5 instances shown in the tables where SimaPro had non zero
characterization factors while GaBi used zeros. Two of these, items 3 and 12, were for
chromium emissions to air and agricultural soil. In both cases it appears SimaPro used
the average of the TRACI 2.1 spreadsheet characterization factors for chromium (III)
and chromium (VI). The remaining 3 cases where GaBi had a zero characterization

141

factor, items 10, 17, and 21, all involved barium. SimaPro used the TRACI 2.1
spreadsheet factor for barium (II).
There are 4 instances where SimaPro and GaBi both have non zero
characterization factors that do not match; these all involved antimony. For items 1, 9,
20, and 32 SimaPro used the TRACI 2.1 spreadsheet characterization factors for
antimony (V), while GaBi used the factors for antimony (III). The characterization factors
for antimony (V) are approximately 155 times those of antimony (III).

142

Table 30A: SimaPro / TRACI 2 ecotoxicity for glass
55.5% Recycled Content
Substance
Item

Name

Compartment

1

Antimony

2

Arsenic

3

Chromium

4

Copper

5

Nickel

2

2
2

7

Vanadium

8

Zinc

air
air

2

Selenium

9
10
11
12
13
14
15
16
17
18
19
20

air

2

6

air

air
air
2

2

Antimony
Barium
Chlorothalonil
Chromium
Copper
Nickel
Tin
Zinc
Barium
Copper
Zinc
Antimony

air
air
soil
soil
soil
soil
soil
soil
soil
soil
soil
soil
soil
water

Sub-compartment(s)
high. pop.; low. pop.; low. pop.,
long-term
high. pop.; low. pop.; low. pop.,
long-term
(unspecified); high. pop.; low. pop.
(unspecified); high. pop.; low.
pop.; low. pop., long-term
(unspecified); high. pop.; low.
pop.; low. pop., long-term
high. pop.; low. pop.
high. pop.; low. pop.; low. pop.,
long-term
(unspecified); high. pop.; low.
pop.; low. pop., long-term
agricultural
agricultural
agricultural
agricultural
agricultural
agricultural
agricultural
agricultural
industrial
(unspecified)
(unspecified); industrial
groundwater, long-term
143

Amount
(kg)

Characterization
Factor(s)
(CTUe per kg)

Impact
(CTUe)

7.78E-08

76,200 to 76,800

0.0059

1.61E-07

16,900 to 17,100

0.0027

4.11E-07

21,200 to 21,300

0.0087

4.95E-07

23,100 to 23,300

0.0115

4.06E-07

6,080 to 6,140

0.0025

7.36E-06

2,960

0.0218

5.41E-07

46,200 to 46,700

0.0250

7.73E-07

16,700 to 16,900

0.0130

1.01E-07
1.92E-06
9.82E-08
1.70E-07
2.60E-07
1.36E-07
1.98E-07
2.78E-07
2.02E-06
5.07E-08
2.35E-07
5.47E-07

95,800
763
57800
26,600
29,200
7,660
0
21,100
763
29,200
21,100
190,000

0.0096
0.0015
0.0057
0.0045
0.0076
0.0010
0
0.0059
0.0015
0.0015
0.0050
0.1039

Table 30A (cont’d)
21
22
23
24
25
26
27
28
29
30
31
32

Barium
Beryllium
Chromium VI
Cobalt
Copper, ion
Lead
Silver, ion
Thallium
Tin, ion
Vanadium, ion
Zinc, ion
Antimony

water
water
water
water
water
water
water
water
water
water
water
water

33

Arsenic, ion

water

34

Barium

water

35

Cadmium, ion

water

36
37

Chromium VI
Copper, ion

water
water

38

Mercury

water

39

Nickel, ion

water

40
Silver, ion
water
41
Tin, ion
water
42 Vanadium, ion
water
43
Zinc, ion
water
Total CTUe For Items In Table
Reported CTUe

groundwater, long-term
groundwater, long-term
groundwater, long-term
groundwater, long-term
groundwater, long-term
groundwater, long-term
groundwater, long-term
groundwater, long-term
groundwater, long-term
groundwater, long-term
groundwater, long-term
(unspecified); groundwater; river
groundwater; groundwater, longterm; river
river
(unspecified); groundwater;
groundwater, long-term; river
groundwater; river
(unspecified); groundwater; river
(unspecified); groundwater;
groundwater, long-term; river
(unspecified); groundwater;
groundwater, long-term; river
(unspecified); groundwater; river
groundwater; river
groundwater; ocean; river
(unspecified); groundwater; river

144

1.06E-05
3.06E-07
2.45E-06
5.26E-06
1.35E-05
1.35E-05
2.65E-08
1.16E-07
1.37E-06
4.75E-06
6.34E-05
4.49E-07

1,530
3,720
105,000
0
0
375
0
35,400
0
0
0
190,000

0.0163
0.0011
0.2568
0
0
0.0051
0
0.0041
0
0
0
0.0853

3.12E-06

40,400

0.1262

4.76E-06

1,530

0.0073

7.38E-07

0

0

9.03E-07
3.17E-07

105,000
0

0.0948
0

6.22E-08

22,100

0.0014

2.49E-05

0

0

5.22E-09
3.70E-07
3.30E-08
1.07E-06

0
0
0
0

0
0
0
0
0.837
0.850

Table 30A (cont’d)
1
2

Items in bold text differ in characterization factors between SimaPro and GaBi.
Characterization factors vary with sub-compartments, maximum variation does not exceed 2%.

Table 30B: GaBi / TRACI 2 ecotoxicity for glass
55.5% Recycled Content
Item

Substance

1
2

Category
Heavy metals to air
Heavy metals to air

3

Heavy metals to air

4
5
6
7
8
9
10
11

Heavy metals to air
Heavy metals to air
Heavy metals to air
Heavy metals to air
Heavy metals to air
Heavy metals to agricultural soil
Inorganic emissions to agricultural soil
Pesticides to agricultural soil

12

Heavy metals to agricultural soil

13
14
15
16
17

Heavy metals to agricultural soil
Heavy metals to agricultural soil
Heavy metals to agricultural soil
Heavy metals to agricultural soil
Inorganic emissions to industrial soil

Name
Antimony
Arsenic (+V)
Chromium
(unspecified)
Copper (+II)
Nickel (+II)
Selenium
Vanadium (+III)
Zinc (+II)
Antimony
Barium
Chlorothalonil
Chromium
(unspecified)
Copper (+II)
Nickel (+II)
Tin (+IV)
Zinc (+II)
Barium
145

Amount
(kg)

Characterization
Impact
Factor
(CTUe)
(CTUe per kg)

7.78E-08
1.61E-07

491
17,000

0.0000
0.0027

4.11E-07

0

0

4.95E-07
4.06E-07
7.36E-06
5.42E-07
7.73E-07
1.01E-07
1.92E-06
9.82E-08

23,200
6,110
2,980
46,400
16,800
615
0
57,800

0.0115
0.0025
0.0219
0.0251
0.0130
0.0001
0
0.0057

1.70E-07

0

0

2.60E-07
1.36E-07
1.98E-07
2.78E-07
2.02E-06

29,200
7,660
1,710
21,100
0

0.0076
0.0010
0.0003
0.0059
0

Table 30B (cont’d)
18
Heavy metals to industrial soil
19
Heavy metals to industrial soil
20
ecoinvent long-term to fresh water
21
ecoinvent long-term to fresh water
22
ecoinvent long-term to fresh water
23
ecoinvent long-term to fresh water
24
ecoinvent long-term to fresh water
25
ecoinvent long-term to fresh water
26
ecoinvent long-term to fresh water
27
ecoinvent long-term to fresh water
28
ecoinvent long-term to fresh water
29
ecoinvent long-term to fresh water
30
ecoinvent long-term to fresh water
31
ecoinvent long-term to fresh water
32
Heavy metals to fresh water
33
Heavy metals to fresh water
34
Inorganic emissions to fresh water
35
Heavy metals to fresh water
36
Heavy metals to fresh water
37
Heavy metals to fresh water
38
Heavy metals to fresh water
39
Heavy metals to fresh water
40
Heavy metals to fresh water
41
Heavy metals to fresh water
42
Heavy metals to fresh water
43
Heavy metals to fresh water
Total CTUe For Items In Table
Reported CTUe
1

Copper (+II)
Zinc (+II)
Antimony
Barium
Beryllium
Chromium (+VI)
Cobalt
Copper (+II)
Lead (+II)
Silver
Thallium
Tin (+IV)
Vanadium (+III)
Zinc (+II)
Antimony
Arsenic (+V)
Barium
Cadmium (+II)
Chromium (+VI)
Copper (+II)
Mercury (+II)
Nickel (+II)
Silver
Tin (+IV)
Vanadium (+III)
Zinc (+II)

5.10E-08
2.35E-07
5.47E-07
1.06E-05
3.06E-07
2.45E-06
5.26E-06
1.35E-05
1.35E-05
2.65E-08
1.16E-07
1.37E-06
4.75E-06
6.34E-05
4.49E-07
3.16E-06
4.77E-06
7.38E-07
9.22E-07
3.17E-07
6.22E-08
2.49E-05
5.22E-09
3.70E-07
3.30E-08
1.07E-06

Items in bold text differ in characterization factors between SimaPro and GaBi.
146

29,200
21,100
1,222
0
3,720
105,000
4,100
55,200
375
194,000
35,400
2,980
113,000
38,600
1,220
40,400
0
9,710
105,000
55,200
22,100
14,900
194,000
2,980
113,000
38,600

0.0015
0.0050
0.0007
0
0.0011
0.2573
0.0216
0.7452
0.0051
0.0051
0.0041
0.0041
0.5368
2.4472
0.0005
0.1277
0
0.0072
0.0968
0.0175
0.0014
0.3710
0.0010
0.0011
0.0037
0.0413
4.801
4.811

11.3.2 Global Warming
TRACI 2 data for the global warming category when obtaining and disposing of 1
kg of corrugated board with 50% recycled content is given in Table 31A for SimaPro,
and Table 31B for GaBi. SimaPro reported 0.984 kg CO2 equivalent while GaBi
reported 0.734 kg CO2 equivalent, differing by a factor of 1.34. The 6 items in each
table account for 99.8% of the impact reported by SimaPro, and 99.5% of the impact
reported by GaBi.
The differences for this impact category came from GaBi including biotic carbon
dioxide, biotic methane, and a credit for carbon sequestration in its reported impact,
while SimaPro did not. The credit was taken in item 1, where carbon dioxide removed
from the air and sequestered in the biomass used to produce the corrugated board got
an effective characterization factor of -1.00 from GaBi, and a factor of zero from
SimaPro. Item 2 covered biotic carbon dioxide where GaBi assigned a characterization
factor of 1.00, and SimaPro a zero. Item 5 covered biotic methane where GaBi assigned
a characterization factor of 25.0, and SimaPro another zero.
The spreadsheet version of TRACI 2.1 had characterization factors of 1.00 for
carbon dioxide, and 25.0 for methane; however, there were no references for biotic
carbon dioxide, biotic methane, or providing a credit for carbon sequestration. The only
guideline for including biotic carbon, biotic methane, and carbon sequestration credit
given in the paper discussing the TRACI 2 methodology (Bare 2011) was a reference to
a 100-year time horizon for global warming potential.

147

Table 31A: SimaPro / TRACI 2 global warming for corrugated board
50% Recycled Content

Name

Compartment

Sub-compartment(s)

Amount
(kg)

Carbon dioxide, in air
Carbon dioxide,
biogenic
Carbon dioxide, fossil
Dinitrogen monoxide
Methane, biogenic
Methane, fossil

air

(combined)

-1.36E+00

Characterization
Factor
(kg CO2 Equiv.
per kg)
0

air

(combined)

9.53E-01

0

0

air
air
air
air

(combined)
(combined)
(combined)
(combined)

9.16E-01
6.09E-05
7.13E-03
1.95E-03

1.00
298
0
25.0

Total kg CO2 Equiv. For Items In Table

0.916
0.018
0
0.049
0.983

Reported kg CO2 Equiv.

0.984

Substance
Item
1
2
3
4
5
6

1

Items in bold text differ in characterization factors between SimaPro and GaBi.

148

Impact
(kg CO2
Equiv.)
0

Table 31B: GaBi / TRACI 2 global warming for corrugated board
50% Recycled Content
Characterization
Factor
(kg CO2 eq per kg)

Impact
(kg CO2
eq)

Carbon Dioxide 1.38E+00

-1.00

-1.380

Substance

Item

Category

Name

Amount
(kg)

1

Renewable Resources

2

Inorganic emissions to air

Carbon Dioxide
(biotic)

9.79E-01

1.00

0.979

3

Inorganic emissions to air

Carbon Dioxide

9.16E-01

1.00

0.916

4

Inorganic emissions to air

6.10E-05

298

0.018

5

Organic emissions to air (group VOC)

5.94E-03

25.0

0.149

6

Organic emissions to air (group VOC)

1.95E-03

25.0

Total kg CO2 eq For Items In Table

0.049
0.730

Reported kg CO2 eq

0.734

1

Nitrous Oxide
(laughing gas)
Methane
(biotic)
Methane

Items in bold text differ in characterization factors between SimaPro and GaBi.

149

11.3.3 Non Carcinogens
TRACI 2 data for the non carcinogens category when obtaining and disposing of
1 kg of PET with 0% recycled content is given in Table 32A for SimaPro, and Table 32B
for GaBi. SimaPro reported 4.43E-07 CTUh (Comparative Toxic Units, human toxicity
potential) while GaBi reported 7.96E-07 CTUh, differing by a factor of 1.80. The 17
items in each table account for 99.0% of the impact reported by SimaPro, and 99.1% of
the impact reported by GaBi.
SimaPro characterization factors for chromium emissions to air in item 3 appear
to be the average of the TRACI 2.1 spreadsheet factors for chromium (III) and
chromium (VI), GaBi had no factor assigned for this item. SimaPro impact factors for
barium emissions to water in items 9 and 14 were the same as the TRACI 2.1
spreadsheet factor for barium (II) to freshwater; GaBi had no factor assigned for these
items. All 3 of these items contributed to the impact reported by SimaPro, but not to the
impact reported by GaBi.
The GaBi characterization factor for vanadium ion, or vanadium (+III), emissions
to water in item 10 matched the TRACI 2.1 spreadsheet factor for vanadium (V) to
freshwater. No factor for vanadium (+III) was found in the TRACI 2.1 spreadsheet.
SimaPro did not have a factor assigned for this item, so it did not contribute to the
impact reported by SimaPro.
GaBi characterization factors for zinc ion, or zinc (+II), emissions to water in
items 11 and 17 matched the TRACI 2.1 spreadsheet factor for zinc(II). The GaBi
characterization factor for cadmium ion, or cadmium (+II), emissions to water in item 15
matched the TRACI 2.1 spreadsheet factor for cadmium(II). SimaPro did not have

150

factors assigned for these items, so they did not contribute to the impact reported by
SimaPro.
.

151

Table 32A: SimaPro / TRACI 2 non carcinogens for PET
0% Recycled Content
Substance
Item

Name

Compartment
2

1

Arsenic

2

Cadmium

3

Chromium

4

Lead

5

Mercury

6

Zinc

air

2

air

3

air

2

air
2

2

air

air

7
8
9
10
11
12

Zinc
Antimony
Barium
Vanadium, ion
Zinc, ion
Antimony

soil
water
water
water
water
water

13

Arsenic, ion

water

Sub-compartment(s)
(unspecified); high.
pop.; low. pop.; low.
pop., long-term
(unspecified)l high.
pop.; low. pop.
(unspecified)l high.
pop.; low. pop.
(unspecified); high.
pop.; low. pop.; low.
pop., long-term
(unspecified); high.
pop.; low. pop.; low.
pop., long-term
(unspecified); high.
pop.; low. pop.; low.
pop., long-term
agricultural
groundwater, long-term
groundwater, long-term
groundwater, long-term
groundwater, long-term
groundwater; river
groundwater;
groundwater, longterm; river

152

Amount
(kg)

Characterization
Factor(s)
(CTUh per kg)

Impact
(CTUh)

6.14E-07

1.65E-02 to 1.71E-02

1.02E-08

2.38E-07

4.45E-02 to 4.65E-02

1.09E-08

8.05E-06

4.15E-5 to 2.08E-4

1.55E-09

1.91E-06

9.32E-03 to 9.57E-03

1.82E-08

5.92E-08

8.15E-01 to 8.55E-01

4.98E-08

3.55E-06

1.52E-02 to 1.60E-02

5.60E-08

4.33E-08
4.64E-05
6.57E-05
2.41E-04
2.41E-04
2.05E-05

4.35E-02
3.64E-04
9.82E-05
0
0
3.64E-04

1.88E-09
1.69E-08
6.46E-09
0
0
7.47E-09

9.31E-06

2.74E-02

2.55E-07

Table 32A (cont’d)
14

Barium

water

15

Cadmium, ion

water

16

Mercury

water

17

Zinc, ion

water

(unspecified);
groundwater; river
groundwater;
groundwater, longterm; river
groundwater;
groundwater, longterm; river
(unspecified);
groundwater; river

5.21E-06

9.82E-05

5.11E-10

4.55E-06

0

0

2.99E-07

1.42E-02

4.25E-09

2.67E-06

0

0

Total CTUh For Items In Table
Reported CTUh
1
2
3

4.39E-07
4.43E-07

Items in bold text differ in characterization factors between SimaPro and GaBi.
Characterization factors vary with sub-compartments, maximum variation does not exceed 5%.
Characterization factors vary with sub-compartments by a multiple of 5.

153

Table 32B: GaBi / TRACI 2 non carcinogens for PET
0% Recycled Content
Item

Substance

1
2

Category
Heavy metals to air
Heavy metals to air

3

Heavy metals to air

4
Heavy metals to air
5
Heavy metals to air
6
Heavy metals to air
7
Heavy metals to agricultural soil
8
ecoinvent long-term to fresh water
9
ecoinvent long-term to fresh water
10
ecoinvent long-term to fresh water
11
ecoinvent long-term to fresh water
12
Heavy metals to fresh water
13
Heavy metals to fresh water
14
Inorganic emissions to fresh water
15
Heavy metals to fresh water
16
Heavy metals to fresh water
17
Heavy metals to fresh water
Total CTUh For Items In Table
Reported CTUh
1

Name
Arsenic (+V)
Cadmium (+II)
Chromium
(unspecified)
Lead (+II)
Mercury (+II)
Zinc (+II)
Zinc (+II)
Antimony
Barium
Vanadium (+III)
Zinc (+II)
Antimony
Arsenic (+V)
Barium
Cadmium (+II)
Mercury (+II)
Zinc (+II)

Amount
(kg)

Characterization
Factor(s)
(CTUh per kg)

Impact
(CTUh)

6.14E-07
2.39E-07

1.68E-02
4.55E-02

1.03E-08
1.09E-08

8.05E-06

0

0

1.91E-06
5.92E-08
3.55E-06
4.33E-08
4.64E-05
6.57E-05
2.41E-04
2.41E-04
2.05E-05
9.31E-06
5.21E-06
4.55E-06
3.00E-07
2.67E-06

9.44E-03
8.35E-01
1.56E-02
4.35E-02
3.64E-04
0
1.94E-04
1.28E-03
3.64E-04
2.73E-02
0
4.27E-04
1.42E-02
1.28E-03

1.80E-08
4.94E-08
5.54E-08
1.88E-09
1.69E-08
0
4.68E-08
3.08E-07
7.46E-09
2.54E-07
0
1.94E-09
4.26E-09
3.42E-09
7.89E-07
7.96E-07

Items in bold text differ in characterization factors between SimaPro and GaBi.

154

11.3.4 Respiratory Effects
TRACI 2 data for the respiratory effects category when obtaining and disposing
of 1 kg of aluminum with 50% recycled content is given in Table 33A for SimaPro, and
Table 33B for GaBi. SimaPro reported 0.0120 kg PM10 equivalent while GaBi reported
0.0210 kg PM10 equivalent, differing by a factor of 1.75. The 5 items in each table
account for 100% of the impacts reported by SimaPro and GaBi.
Perhaps the most interesting aspect of how this category was handled in
SimaPro and GaBi isn’t a difference that occurred between them, but rather it is a
similarity; both software programs reported the impact for this category in terms of kg
PM10 equivalent. This is surprising because the TRACI 2.1 spreadsheet only provides
characterization factors that use kg PM2.5 equivalent as the reference substance. A
check of the user’s manual for TRACI 2 (Bare 2012) confirmed that the expected
reference is kg PM2.5 equivalent. After further investigation, a comment was found in
the paper introducing TRACI 2 (Bare 2011) that stated the original TRACI methodology
for this impact category was changed for TRACI 2, going from PM10 based to PM2.5
based.
GaBi assigned a characterization factor of 1.00 to item 1, particles > 10 μm, while
SimaPro had no factor assigned, so this item only contributed to the impact reported by
GaBi. Item 2, particles between 2.5 μm and 10μm, is assigned a characterization factor
of 1.00 by SimaPro, and a factor of 1.67 by GaBi. This means the mass for this item had
proportionally more effect on the impact reported by GaBi than the impact reported by
SimaPro. SimaPro and GaBi agreed the characterization factor for item 3, particles <
2.5 μm, should be 1.67. Since the spreadsheet for TRACI 2.1 listed a different reference

155

substance, it did not provide any insight regarding the discrepancies in characterization
factors between SimaPro and GaBi for this impact category.

156

Table 33A: SimaPro / TRACI 2 respiratory effects for aluminum
50% Recycled Content
Substance
Item

Name

Compartment

Sub-compartment(s)

Amount
(kg)

0

Impact
(kg PM10
Equiv.)
0

1.00

3.80E-03

1.67
0.0265
0.167

4.39E-03
2.99E-04
3.51E-03
0.0120
0.0120

Characterization Factor
(kg PM10 Equiv. per kg)

Particulates > 10 μm
air
(combined)
6.45E-03
Particulates > 2.5 μm
2
air
(combined)
3.80E-03
and < 10 μm
3
Particulates < 2.5 μm
air
(combined)
2.63E-03
4
Nitrogen oxides
air
(combined)
1.13E-02
5
Sulfur dioxide
air
(combined)
2.11E-02
Total kg PM10 Equiv. For Items In Table
Reported kg PM10 Equiv.
1
Items in bold text differ in characterization factors between SimaPro and GaBi.
1

Table 33B: GaBi / TRACI 2 respiratory effects for aluminum
50% Recycled Content
Item

Substance

Category
Name
1
Particles to air
Dust (>PM10)
2
Particles to air
Dust (PM2,5 - PM10)
3
Particles to air
Dust (PM2,5)
4
Inorganic emissions to air
Nitrogen oxides
5
Inorganic emissions to air
Sulphur dioxide
Total kg PM10 Equiv. For Items In Table
Reported kg PM10 Equiv.
1

Amount
(kg)

Characterization Factor
(kg PM10 Equiv. per kg)

6.47E-03
3.80E-03
2.63E-03
1.13E-02
2.11E-02

1.00
1.67
1.67
0.0265
0.167

Items in bold text differ in characterization factors between SimaPro and GaBi.
157

Impact
(kg PM10
Equiv.)
6.47E-03
6.35E-03
4.39E-03
2.99E-04
3.52E-03
0.0210
0.0210

11.4 Summary Of Underlying Causes Of Dissimilar Results
A total of 14 combinations of basic materials and impact categories from 3
different impact assessment methods were examined, with the end result that all
significant discrepancies between impact assessments from GaBi and SimaPro were
traced to differences in the characterization factors used to implement the assessment
methods. Table 34 summaries the 98 differences in characterization factors that were
found. No differences were found in the LCI of the two programs for the type and
amount of significant contributors to the impact assessments examined, and all
calculations performed by the programs to convert LCI data to impact assessments
were verified as being correct.

Table 34: Summary of differences in GaBi and SimaPro characterization factors
Characterization Factor Data

Impact 2002+

ReCiPe

TRACI 2

Number of impact categories/material
combinations examined

5

5

4

Instances where GaBi had a
characterization factor of zero while
SimaPro had non zero.

9

21

9

Instances where GaBi had a non zero
characterization factor while SimaPro
had zero.

22

4

22

Instances where GaBi and SimaPro both
had non zero characterization factors that
differed by more than 10%.

2

0

5

Instances where SimaPro had spelling
differences between substance and
characterization factor names, resulting
in zero being used.

4

0

0

158

12. RESEARCH SUMMARY & CONCLUSIONS
Through a series of tests, several life cycle assessment software programs were
compared to each other to establish how consistent the results are from one program to
the next. Comparisons were made involving complete packaging systems and basic
packaging materials, as well as an examination of how individual substances contribute
to reported environmental impact estimates. It was found that impact assessments can
vary widely between software programs.
While an argument might be made that variation is to be expected when dealing
with programs based on different impact assessment methodologies, comparisons done
here show that even when the same methodology is used and the inputs are matched
as closely as possible, implementations of a supposedly common methodology in
different software can provide different results. One example of this is the global
warming category in the Impact 2002+ version 2.1 assessment method; GaBi included
biotic carbon dioxide and a credit for carbon sequestration while SimaPro did not. A
similar issue was found between GaBi and SimaPro for the global warming category in
the TRACI 2 methodology. Another example is the water depletion category in ReCiPe;
GaBi includes water going through electrical generation turbines and SimaPro does not,
causing GaBi to report a result that is several orders of magnitude larger than that
reported by SimaPro.
For the small set of basic materials and impact categories examined, the most
common cause of differences between implementations of impact assessment methods
in SimaPro and GaBi is one software including characterization factors for substances
that the other software excludes. There is no consistency as to which software includes

159

a substance and which excludes it; this varies with impact category and substance.
There are also instances where both SimaPro and GaBi have non-zero characterization
factors for a substance, but the factors differ significantly. Whether or not a difference in
characterization factors has a significant effect on reported impacts depends on the
amount of the substance in the life cycle inventory. The aforementioned issue with
global warming was much more noticeable when the comparison included corrugated
board with a low percent of recycled fiber, meaning higher virgin fiber content, than
when the percent of recycled fiber was high. The water depletion issue comes into play
more as hydroelectric power use increases. Whenever there is a difference in
characterization factors between software, there is the inherent possibility that impacts
reported by the software will significantly disagree for some conditions and not for
others.
Life cycle assessment software has the potential to simplify the often complex
and time consuming task of doing life cycle analysis. For it to fulfill that potential there
needs to be consistency in results that users can rely on. This is particularly true when
using a common impact assessment methodology. If the implementations of an impact
assessment methodology vary, then expectations of consistent results are lost,
diminishing the usefulness of the software.
Providers of life cycle assessment software have commercial interests in selling
their own programs, so the task of identifying differences in assessment methods
between software and promoting consistency in results falls on the users. It is
recommended that as new versions of software are released they should be subjected
to comparison testing. For fully ISO 14040/14044 compliant software such as SimaPro

160

and GaBi, the use of basic materials and a common database, when combined with the
practice of accounting for all the mass of each substance that makes a significant
contribution to the reported results, can provide a basis for such comparisons.
This paper has used a broad study of impact categories to establish that the
choice of LCA software can significantly affect results. What is needed in future work on
this subject is to focus on how these inconsistencies affect different areas of research
that rely on LCA. Many assessment methods cover multiple environmental impact
categories, yet only a subset of impact categories may be used in research on a given
subject. For example, researchers on carbon footprint would be far more interested in
the Global Warming impact category of Impact 2002+ than either the Ozone Layer
Depletion or Terrestrial Ecotoxicity categories. Key items of interest in LCA for different
subject areas need to be identified, and studies conducted to identify and rectify any
inconsistencies between LCA software programs that can affect research in each
subject area.

161

APPENDICES

162

APPENDIX A: Population, MSW, and Oil Consumption Data
Table A1: World population estimates
Population
(billions of
people)
1
2
3
4
5
6
7
8
9

Year
1804
1922
1959
1974
1987
1999
2013
2028
2048

Source: U.S. Census Bureau (U.S. Census 2002)

Table A2: U.S. Population and waste generation estimates

Year

1960
1970
1980
1990
2000
2010
Percent
Change
1960 to 2010

Total Generated
Municipal Solid
Waste (millions
of tons)

Total Containers
Population
& Packaging In
(millions
Waste Stream
of people)
(millions of tons)

88.12
121.06
151.64
208.27
242.54
249.86

27.37
43.56
52.67
64.53
75.84
75.64

180.67
205.05
227.73
250.13
282.39
309.33

184%

176%

71%

Sources: U.S. Census Bureau (U.S. Census 2012, 2013)
U.S. Environmental Protection Agency (EPA 2011)

163

Table A3: World, U.S., and China oil consumption in thousands of barrels per day
Year
1980
1985
1990
1995
2000
2005
2010
Percent
Change
1960 to 2010

World
63,120
60,085
66,550
70,132
76,788
84,089
87,314

U.S.
17,056
15,726
16,989
17,725
19,701
20,802
19,180

China
1,765
1,885
2,296
3,363
4,796
6,695
9,330

38.3%

12.5%

429%

Source: United States Energy Information Administration (EIA 2013)

164

APPENDIX B: Container Modeling Assumptions
Process Assumptions
1. Liners in caps were omitted from the study.
2. Label adhesives were omitted from the study.
3. Printing of labels and containers was included for selected products.
4. Filling of containers was omitted from the study.
5. Distribution center handling was omitted from the study.
6. Retail sales handling was omitted from the study.
7. Consumer use (end use) was omitted from the study.
8. Production scrap and shipping/handling damage was omitted from the study.
9. Affect of varying transport distances was included for selected products.

Functional Units and Units Of Measure
1. Functional unit for tuna in steel can and tuna in pouch was 1kg of product.
2. Functional unit for PLA bottle (water), aseptic carton (juice), glass bottle (beer),
PET bottle (carbonated beverage), and aluminum Can (carbonated beverage)
was 1 liter of fluid.
3. Functional unit for plastic (PP) crate (flowers) and corrugated box (flowers) was
½ box dry pack flowers.
4. Analysis was done in metric units.

Reuse
1. PP crates were modeled for 1, 10, and 100 uses.

165

Product Specific Assumptions
1. Refrigerated truck transport of flower boxes/crates was modeled as truck
transport combined with a separate refrigeration function based on volume of
material and time for transport. A transport speed of 100km per hour (62.1 miles
per hour) was assumed when determining time for refrigerated truck transport.
The volume of material (PP or corrugated board) used to produce the box/crate
was used as shipping volume, not the volume of the container. The interior
volume of the container was allocated to the product (flowers), not the package.
2. Air transport distance used for flower boxes/crates on the out going trip was
500km and 2500km. Ship transport distance used for flower crates on the return
trip was 500km and 2100km. Standard truck transport distance used for flower
crates on the return trip was 1200km. Air, ship, and truck transport distances
between cities came from an unpublished report on distribution of cut flowers
done at MSU School of Packaging. Air transport distance from Bogota, Colombia
to Miami, FL is 2446km. Ship transport distance from Port Everglade, FL to
Cartagena, Colombia is 2043km.
3. Rail transport distance for beer bottles was 0, 500km, and 4000km. Chicago to
St. Louis by rail is 459km, or 285 miles, and Chicago to San Francisco by rail is
3927km, or 2440 miles (Rail Passenger USA 2011).

166

Recycling and Waste Disposal
1. At least two recycling rates, or composting rates in the case of PLA, were used
with various products to see if different LCA software produced results that
tracked each other (correlated).
2. Where available, typical recycling rates in US for a product, rounded to the
nearest 10%, were used as one reference point.
3. What was not recycled or composted went to landfill/incineration, with the U.S.
averages of 82% to landfill and 18% to incineration being used (EPA 2009).

Recycled Content In United States
1. Aluminum can (beverage):

68% (Aluminum Association 2011).

2. Aseptic carton (juice):

No recycled content data found.

3. Corrugated box (flowers):

41.85% (Corrugated Packaging Alliance 2010).

4. Flexible pouch (tuna):

No recycled content data found.

5. Glass bottle (beer):

27% (Glass Packaging Institute 2010).

6. PET bottle (beverage):

3% (EPA 2010).

7. PLA bottle (water):

No recycled content data found.

8. PP crate (flowers):

No recycled content data found.

9. Steel can (tuna):

25% (American Iron and Steel Institute 2010).

Recycling and Composting Rates For United States
1. Aluminum can (beverage):

50.7% Recycled (EPA 2009).

2. Aseptic carton (juice):

6.5% Recycled (EPA 2009).

167

3. Corrugated box (flowers):

81.3% Recycled (EPA 2009).

4. Flexible pouch (tuna):

0.0% Recycled (Bumble Bee Foods 2011).

5. Glass bottle (beer):

39.0% Recycled (EPA 2009).

6. PET bottle (beverage):

28.0% Recycled (EPA 2009).

7. PLA bottle (water):

No composting or recycling data found.

8. PP crate (flowers):

7.4% Recycled (EPA 2009).

9. Steel can (tuna):

66.0% Recycled (EPA 2009).

168

APPENDIX C: Container Compositions and Material Weights
Aluminum Beverage Can
Weight information for a 12oz (354.9ml) aluminum beverage can taken from an
unpublished study done at MSU School of Packaging.

Can dimensions: 12.1cm high x 6.5cm diameter (at widest point).
Cube utilization: 69.4% (based on volume occupied by fluid).
Average weight of can based on 5 samples: 13.018g
Functional unit is 1 liter of fluid (beverage).
There are 2.818 cans per functional unit.
Weight of aluminum for functional unit: 36.68g.
No area estimate for printing.

169

Aseptic Carton
6 cartons of juice, 200ml each, were acquired for this project.
Carton dimensions: 10.6cm high x 5.6cm long x 3.8cm wide.
Cube utilization: 88.7% (based on volume occupied by fluid).
Table C1: Weight of aseptic carton components
Carton

1
2
3
4
5
6

Unopened
Weight
(g)
219.9
220.1
220.1
219.8
219.8
220.0

Weight Of
Empty Carton
(g)
9.333
9.286
9.407
9.306
9.352
9.363

Weight Of
Straw
(g)
0.372
0.373
0.374
0.371
0.375
0.373

Weight of
Straw Bag
(g)
0.159
0.140
0.156
0.148
0.148
0.137

Average
St. Dev.

220.0
0.1

9.341
0.043

0.373
0.001

0.148
0.009

Functional unit is 1 liter of fluid (juice).
There are 5.000 cartons per functional unit.
Table C2: Packaging material used for aseptic carton
Item

Type of
Material

Percent
By Weight

Box

SBS
PE
aluminum foil
PP
PP

75
20
5
100
100

Straw
Straw Bag

Weight Per
Container
(g)
7.006
1.868
0.467
0.373
0.148

Weight Per
Functional Unit
(g)
35.029
9.341
2.335
1.865
0.740
2

Carton area for printing estimate: 0.200m long x 0.155m wide = 0.0310m .

170

Corrugated Box
Weight information for corrugated box taken from an unpublished study on
distribution of cut flowers done at MSU School of Packaging.

Cube utilization: assume 100% for the purposes of this study.
Average weight of box based on 2 samples: 700g
Functional unit is ½ flower box (common unit in flower industry).
There is 1 corrugated box per functional unit.
Weight of corrugated board for functional unit: 700g.
No area estimate for printing.

171

Flexible Pouch
6 pouches of 74g net weight tuna were acquired for this project.
Pouch dimensions: 16cm long x 11.5cm wide, thickness varies.
Cube utilization: assume 100% for the purposes of this study.
Table C3: Weight of flexible pouches
Pouch

1
2
3
4
5
6

Unopened
Weight
(g)
80.42
83.03
82.16
78.11
77.77
83.04

Emptied
Weight
(g)
6.223
6.460
6.378
6.384
6.245
6.433

Average
St. Dev.

80.76
2.38

6.354
0.098

Functional unit is 1kg net weight of tuna.
There are 13.514 pouches per functional unit.
Table C4: Packaging material used for flexible pouch
Item

Type Of
Material

Percent
By Weight

Pouch

PET
PP
aluminum foil
nylon

40
40
15
5

Weight Per
Container
(g)
2.542
2.542
0.953
0.318

Weight Per
Functional Unit
(g)
34.345
34.345
12.879
4.293
2

Pouch area for printing estimate: 2 x 0.160 long x 0.115m wide = 0.0368m .

172

Glass Beer Bottle
6 bottles of beer, 12oz (354.9ml) each, were acquired for this project.
Bottle dimensions: 23cm high x 5.9cm diameter (at base).
Cube utilization: 44.3% (based on volume occupied by fluid).
Table C5: Weight of bottle components
Bottle

1
2
3
4
5
6
Average
St. Dev.

Unopened
Weight
(g)
530.0
529.7
529.9
531.0
530.2
529.9

Weight of
Empty Bottle
(g)
187.362
187.012
187.231
188.292
187.493
187.218

Total Weight
Of Labels
(g)
0.385
0.391
0.340
0.393
0.410
0.407

Weight of
Cap
(g)
2.106
2.152
2.103
2.122
2.099
2.126

530.1
0.451

187.4
0.450

0.388
0.025

2.118
0.020

Functional unit is 1 liter of fluid (beer).
There are 2.818 bottles per functional unit.
Table C6: Packaging material used for beer bottle
Item

Type of
Material

Bottle
Label
Cap

glass (brown)
bi-axially oriented PP
steel

Weight Per
Container
(g)
187.434
0.388
2.118

Weight Per
Functional Unit
(g)
528.133
1.093
5.969
2

Surface area of sheet steel needed for cap: 0.00168m .

173

Table C7: Label areas for printing estimate
Item
Front label
Back label
Neck label

Length
(m)
0.064
0.048
0.038

Total area

Width
(m)
0.076
0.052
0.017

Area
(m2)
0.004864
0.002496
0.000646
0.008006

174

PET Carbonated Beverage Bottle
Weight information for a 12oz (354.9ml) PET bottle taken from an unpublished study
done at MSU School of Packaging. Bottle dimensions and cube utilization for 12oz
bottle estimated from dimensions of 20oz bottle by assuming height as the only
variable.

Bottle dimensions: 15.4cm high x 7.0cm diameter (at base).
Cube utilization: 47.0% (based on volume occupied by fluid).
Functional unit is 1 liter of fluid (carbonated beverage).
There are 2.818 bottles per functional unit.
Table C8: Packaging material used for PET bottle
Item

Type of
Material

Bottle
Label
Cap

PET
PP
PP

Weight Per
Container
(g)
24.221
2.707
2.870

Weight Per
Functional Unit
(g)
68.247
7.628
8.087
2

Label area for printing estimate: 0.064m long x 0.029m wide = 0.00186m .

175

PLA Water Bottle
6 bottles of water, 500ml per bottle, were acquired for this project.
Bottle dimensions: 21.3cm high x 6.4cm diameter (at base).
Cube utilization: 57.3% (based on volume occupied by fluid).
Table C9: Weight of PLA bottle components
Bottle

Unopened
Weight
(g)
473.4
474.1
474.3
478.1
473.0
473.6

Weight Of
Empty Bottle
and Label
(g)
24.399
24.540
24.619
24.550
24.471
24.491

Weight Of
Cap and Tamper
Evident Ring
(g)
2.207
2.183
2.161
2.200
2.191
2.163

1
2
3
4
5
6
Average
St. Dev.

474.4
1.9

24.51
0.076

2.184
0.019

Functional unit is 1 liter of fluid (water).
There are 2.000 bottles per functional unit.
Table C10: Packaging material used for PLA bottle
Item

Type Of
Material

Bottle and Label
Cap

PLA
HDPE

Weight Per
Container
(g)
24.512
2.184

Weight Per
Functional Unit
(g)
49.023
4.368
2

Label area for printing estimate: 0.185m long x 0.076m wide = 0.0141m .
Label weight per container assuming 1 mil thick PLA at 1.24g/cm3 density: 0.443g.
Estimated weight of PLA per container without label: 24.1g.

176

Plastic (PP) crate
Weight information for plastic crate taken from an unpublished study on distribution
of cut flowers done at MSU School of Packaging.

Cube utilization: assume 100% for the purposes of this study.
Weight of plastic crate: 1.8kg.
Functional unit is ½ flower box (common unit in flower industry).
There is 1 plastic crate per functional unit.
Weight of PP for functional unit: 1.8kg.
No area estimate for printing.

177

Steel Food Can
6 cans of 142g net weight tuna were acquired for this project.
Can dimensions: 3.4cm high x 8.5cm diameter (at top).
Cube utilization: 78.5%.
Table C11: Weight of steel food can components
Can

Unopened
Weight
(g)

Emptied
Weight Of
Top Lid
(g)
6.830
6.818
6.841
6.841
6.833
6.778

Weight Of
Paper Label
(g)

169.1
170.6
171.5
174.1
173.2
172.5

Emptied
Weight Of
Can Without Lid
(g)
21.818
21.903
21.607
21.606
21.882
21.855

1
2
3
4
5
6
Average
St. Dev.

171.8
1.8

21.78
0.136

6.824
0.024

0.683
0.006

0.679
0.693
0.687
0.682
0.679
0.678

Functional unit is 1kg net weight of tuna.
There are 7.042 cans per functional unit.
Table C12: Packaging material used for steel can
Item

Type Of
Material

Can
Lid
Label

steel
steel
paper

Weight Per
Container
(g)
21.779
6.824
0.683

Weight Per
Functional Unit
(g)
153.370
48.053
4.810
2

Label area for printing estimate: 0.275m long x 0.027m wide = 0.00743m .

178

APPENDIX D: Product LCA Flow Diagrams
Aluminum
Sheet For Can
Bodies (3004/
3104)

Truck

Draw and Iron
Process to
Form Can
Bodies

Interior
Coating (Liner)
Process

Printing
Process

Landfill
Recycled
Content
Aluminum
Sheet For Tab
(5082/5182/
5042)

Truck

Truck

Stamping
Process To
Form Stay-Tab

Attachment Of
Stay-Tab To
End

Truck

Cleaning,
Filling, and
Sealing

Warehousing,
Distribution,
Sales, and Use

Truck

Incineration

Truck

Recycled
Content

Aluminum
Sheet For Can
Ends (5052/
5082/5182)

Truck

Not modeled in this study.
Recycling

Truck

Coating (Liner)
Process

Stamping
Process To
Form Ends

May Be Modeled In Some LCA Software As One Process

Notes: 1. Ink and sealing compound are not included in this study.
2. Cleaning, filling, and sealing are not modeled in this study.
3. Warehousing, distribution, sales, and use functions are not modeled for this study.
4. Truck transportation distances to be determined.

Figure D1: Flow diagram for aluminum can used for carbonated beverages.
179

PP Resin

Truck

Truck

LDPE Resin

Truck

Cast Film
Process For
Film To
Contain Straw
Extrusion
Process To
Form Straw

LLDPE Resin

Truck

Aluminum

Truck

Rolling
Process To
Form Foil

Truck

Flexographic
Printing

Notes: 1.
2.
3.
4.

Form/Fill/Seal
To Create
Straw In
Packet

Truck

Landfill

Extrusion
Process

Extrusion
Process

Paper

Truck

Truck
Lamination To
Form
Multilayer
Material

Truck

Sterilization,
Forming,
Filling, and
Sealing

Warehousing,
Distribution,
Sales, and Use

Truck

Incineration

Truck
Recycling

Not modeled in this study.

Printing ink is not included in this study.
Sterilization, container forming, filling, and sealing are not modeled for this study.
Warehousing, distribution, sales, and use functions are not modeled for this study.
Truck transportation distances to be determined.

Figure D2: Flow diagram for aseptic carton.
180

C-Flute
Corrugated
Board

Truck

Converstion
Process To
Form Box
Landfill
Truck

Filling

Truck

Air

Distribution
Center

Refrigerated
Truck

Destination

Truck

Incineration

Truck
Recycling

Notes: 1.
2.
3.
4.

No printing process or ink is included in this study.
Filling, distribution center, and destination processes are not modeled in this study.
Truck transportation distances to/from airports are rolled into other truck distances for modeling simplicity.
Air and truck transportation distances to be determined.

Figure D3: Flow diagram for corrugated box.

181

PET Resin

Truck

Cast Film
Process
(Printable
Outer Layer)

Nylon Resin

Truck

Cast Film
Process
(Middle Layer)

Truck

Reverse
Gravure
Printing

Truck
Truck

Aluminum

Truck

Rolling
Process To
Form Foil

Truck

Lamination To
Form
Multilayer
Material

Truck

Conversion
Process To
Form Pouches
Landfill

Truck
PP Resin

Truck

Truck

Cast Film
Process (Food
Contact Layer)

Cleaning,
Filling, Sealing,
and Retorting

Adhesives For
Laminating
Layers Together

Truck

Warehousing,
Distribution,
Sales, and Use

Not modeled in this study.

Notes: 1.
2.
3.
4.

Printing ink is not included in this study.
Cleaning, filling, sealing, and retorting are not modeled for this study.
Warehousing, distribution, sales, and use functions are not modeled for this study.
Truck transportation distances to be determined.

Figure D4: Flow diagram for flexible pouch used for packaging tuna.
182

Truck

Incineration

PP Resin

Truck

Blowen Film
Process To
Form Labels

Truck

Flexographic
Printing

Truck

Landfill
Truck

ECCS Sheet
Steel

Truck

Stamping
Process To
Form Cap

Insertion Of
Cap Liner

Truck

Cleaning,
Filling, Sealing,
and Applying
Label

Warehousing,
Distribution,
Sales, and Use

Truck

Incineration

Truck
Silica
(White Sand)

Cap Liner

Recycling

Rail
Not modeled in this study.

Soda
(Sodium
Bicarbonate)

Lime
(Limestone)

Cullet
(Recycled
Glass)

Iron Sulphide
(Colorant)

Rail

Rail

Truck

Glass Furnace

Press and
Blow Process
To Form Bottle

Rail
Notes: 1.
2.
3.
4.

Ink and adhesive are not included in this study.
Cleaning, filling, sealing, and applying label are not modeled for this study.
Warehousing, distribution, sales, and use functions are not modeled for this study.
Truck and rail transportation distances to be determined.

Rail

Figure D5: Flow diagram for glass (amber) beer bottle.
183

PP Resin

Truck

Cast Film
Process To
Form Labels

Truck

Flexographic
Printing

Truck

Landfill
Truck

Truck
Injection
Molding To
Form Closure

Cleaning,
Filling, Sealing,
and Applying
Label

Truck

Warehousing,
Distribution,
Sales, and Use

Truck

Incineration

Truck
PET Resin

Truck

Injection
Stretch Blow
Molding To
Form Bottles

Truck

Not modeled in this study.

Recycled
Content

Notes: 1.
2.
3.
4.

Ink and adhesive used on label are not included in this study.
Cleaning, filling, sealing, and applying label are not modeled for this study.
Warehousing, distribution, sales, and use functions are not modeled for this study.
Truck transportation distances to be determined.

Figure D6: Flow diagram for PET carbonated beverage bottle.

184

Recycling

PLA Resin

Truck

Cast Film
Process To
Form Labels

Truck

Flexographic
Printing

Truck

Landfill
Truck

Truck
Injection
Stretch Blow
Molding To
Form Bottles

Cleaning,
Filling, Sealing,
and Applying
Label

Truck

Warehousing,
Distribution,
Sales, and Use

Truck

Incineration

Truck
HDPE Resin

Notes: 1.
2.
3.
4.

Truck

Injection
Molding To
Form Closure

Truck

Not modeled in this study.

Ink and adhesive used on label are not included in this study.
Cleaning, filling, sealing, and applying label are not modeled for this study.
Warehousing, distribution, sales, and use functions are not modeled for this study.
Truck transportation distances to be determined.

Figure D7: Flow diagram for PLA water bottle.

185

Industrial
Composting

PP Resin

Truck

Injection
Molding To
Form Crate
Landfill
Truck

Truck

Filling

Air

Distribution
Center

Refrigerated
Truck

Destination

Truck

Incineration

Truck

Truck

Recycling
Truck
Cleaning

Notes: 1.
2.
3.
4.

Ship

Port

No printing process or ink is included in this study.
Cleaning, filling, distribution center, destination, and port processes are not modeled in this study.
Truck transportation distances to/from airports and port are rolled into other truck distances for modeling simplicity.
Air, ship, and truck transportation distances to be determined.

Figure D8: Flow diagram for plastic (PP) crate.

186

Paper
For Label)

Truck

Flexographic
Printing

Truck

Stamping
Process To
Form Lid

Landfill
Truck

Truck
Truck

Tin Free Sheet
Steel

Truck

Organic
Coating
Application
Process

Truck

Draw-Redraw
Process To
Form 2 Piece
Can

Truck

Cleaning,
Filling, Sealing,
Retorting, and
Applying Label

Warehousing,
Distribution,
Sales, and Use

Adhesive To
Attach Label

Recycled Content

Not modeled in this study.

Notes: 1. Ink and adhesive for label are not included in this study.
2. Cleaning, filling, sealing, retorting, and applying label are not modeled in this study.
3. Warehousing, distribution, sales, and use functions are not modeled for this study.
4. Truck transportation distances to be determined.

Figure D9: Flow diagram for steel food can used for packaging tuna.

187

Truck

Incineration
(Label)

Truck
Recycling

APPENDIX E: Test Parameter Combinations For Containers
LCA Software Study
Test Parameter Combinations For Aluminum Beverage Can
Test Parameters
Aluminum recycled at end-of- life (EOL):
Aluminum recycled content:
Transport distance, rail:
Transport distance, standard truck:

0, 50, and 100%
0, 10, 70, and 100%.
0 km.
100 and 1000 km.

Note: Doing all combinations would require 3 x 4 x 1 x 2 = 24 combinations.

Parameter Combinations To Be Used
See Table E1 below.
Table E1: Aluminum beverage can test parameter combinations
Parameter
Aluminum Recycled At EOL (%)
Aluminum Recycled Content (%)
Rail (km)
Standard Truck (km)

1
50
70
0
100

188

2
0
70
0
100

3
100
70
0
100

Test
4
50
0
0
100

5
50
10
0
100

6
50
100
0
100

7
50
70
0
1000

LCA Software Study
Test Parameter Combinations For Aseptic Carton
Test Parameters
Aseptic cartons recycled at end-of-life (EOL):
Aseptic carton recycled content:
Transport distance, rail:
Transport distance, standard truck:

0, 10, 50, and 100%
0%.
0 km.
100 and 1000 km.

Note: Doing all combinations would require 4 x 1 x 1 x 2 = 8 combinations.

Parameter Combinations To Be Used
See Table E2 below.
Table E2: Aseptic carton test parameter combinations
Parameter
Aseptic Carton Recycled At EOL (%)
Aseptic Carton Recycled Content (%)
Rail (km)
Standard Truck (km)

189

1
10
0
0
100

2
0
0
0
100

Test
3
50
0
0
100

4
100
0
0
100

5
10
0
0
1000

LCA Software Study
Test Parameter Combinations For Corrugate Box
Test Parameters
Number of uses for each box:
Corrugate recycled at end-of-life (EOL):
Corrugate recycled content:
Transport distance, outgoing air:
Transport distance, outgoing refrigerated truck:
Transport distance, outgoing standard truck:
Transport distance, return ship:
Transport distance, return standard truck:

1.
0, 80, 100%
0, 25, 50, and 100%.
500 and 2500 km.
100, and 1000 km.
100 km.
Not Applicable.
Not Applicable.

Note: Doing all combinations would require 1 x 3 x 4 x 2 x 2 x 1 x1 x 1 = 48
combinations.

Parameter Combinations To Be Used
See Table E3 below.
Table E3: Corrugated box test parameter combinations
Parameter
Number Of Uses
Corrugate Recycled At EOL
(%)
Corrugate Recycled Content
(%)
Outgoing Air (km)
Outgoing Refrigerated Truck
(km)
Outgoing Standard Truck
(km)
Return Ship (km)
Return Standard Truck (km)

1
1

2
1

3
1

4
1

Test
5
1

80

0

100

80

80

80

80

80

50

50

50

0

25

100

50

50

2500 2500 2500 2500 2500 2500

500

2500

100

100

100

100

100

100

100

1000

100

100

100

100

100

100

100

100

0
0

0
0

0
0

0
0

0
0

0
0

0
0

0
0

190

6
1

7
1

8
1

LCA Software Study
Test Parameter Combinations Flexible Tuna Pouch
Test Parameters
Flexible pouch recycled at end-of-life (EOL):
Flexible pouch recycled content:
Transport distance, standard truck:

0%
0%.
100 km.

Note: Doing all combinations would require 1 x 1 x 1 = 1 combination.

Parameter Combinations To Be Used
See Table E4 below.
Table E4: Flexible pouch test parameter combinations
Parameter
Flexible Pouch Recycled At EOL (%)
Flexible Pouch Recycled Content (%)
Standard Truck (km)

191

Test
1
0
0
100

LCA Software Study
Test Parameter Combinations For Glass Beer Bottle
Test Parameters
Glass recycled at end-of-life (EOL):
Glass recycled content:
Transport distance, rail:
Transport distance, standard truck:

0, 40, 100%
0, 25, 50, and 100%.
0, 500, and 4000 km.
100 and 1000 km.

Note: Doing all combinations would require 3 x 4 x 3 x 2 = 72 combinations.

Parameter Combinations To Be Used
See Table E5 below.
Table E5: Glass bottle test parameter combinations
Parameter
Glass Recycled At EOL
(%)
Glass Recycled Content
(%)
Rail (km)
Standard Truck (km)

1

2

3

4

Test
5

40

0

100

40

40

40

40

40

40

25

25

25

0

50

100

25

25

25

500
100

500
100

500
100

500
100

500
100

500
100

0
100

192

6

7

8

9

4000 500
100 1000

LCA Software Study
Test Parameter Combinations PET Bottle
Test Parameters
PET recycled at end-of-life:
PET recycled content:
Transport distance, rail:
Transport distance, standard truck:

0, 30, and 100%
0, 10, 50, and 100%.
0 km.
100 and 1000 km.

Note: Doing all combinations would require 3 x 4 x 1 x 2 = 24 combinations.

Parameter Combinations To Be Used
See Table E6 below.
Table E6: PET bottle test parameter combinations
Parameter
PET Recycled At EOL (%)
PET Recycled Content (%)
Rail (km)
Standard Truck (km)

1
30
10
0
100

2
0
10
0
100

193

3
100
10
0
100

Test
4
30
0
0
100

5
30
50
0
100

6
30
100
0
100

7
30
10
0
1000

LCA Software Study
Test Parameter Combinations PLA Bottle
Test Parameters
PLA composted at end-of-life (EOL):
PLA recycled content:
Transport distance, rail:
Transport distance, standard truck:

0, 10, and 50%
0%.
0 km.
100 and 1000 km.

Note: Doing all combinations would require 3 x 1 x 1 x 2 = 6 combinations.

Parameter Combinations To Be Used
See Table E7 below.
Table E7: PLA bottle test parameter combinations
Parameter
PLA Composted At EOL (%)
PLA Recycled Content (%)
Rail (km)
Standard Truck (km)

194

1
10
0
0
100

Test
2
0
0
0
100

3
50
0
0
100

4
10
0
0
1000

LCA Software Study
Test Parameter Combinations For Plastic (PP) Crate
Test Parameters
Number of uses for each crate:
PP recycled at end-of-life (EOL):
PP recycled content:
Transport distance, outgoing air:
Transport distance, outgoing refrigerated truck:
Transport distance, outgoing standard truck:
Transport distance, return ship:
Transport distance, return standard truck:

1, 10, and 100.
10, 50, and 100%
0%.
500 and 2500 km.
100, and 1000 km.
100 km.
500 and 2100 km.
1200 km.

Note: Doing all combinations would require 3 x 3 x 1 x 2 x 2 x 1 x 2 x 1 = 72
combinations.

Parameter Combinations To Be Used
See Table E8 below.
Table E8: Plastic (PP) crate test parameter combinations
Parameter
Number Of Uses
PP Recycled At EOL (%)
PP Recycled Content (%)
Outgoing Air (km)
Outgoing Refrigerated Truck
(km)
Outgoing Standard Truck
(km)
Return Ship (km)
Return Standard Truck (km)
1

Test
1
2
3
4
5
10
1
100
10
10
10
10
10
50
100
0
0
0
0
0
2500 2500 2500 2500 2500

6
10
10
0
500

7
8
10
10
10
10
0
0
2500 2500

100

100

100

100

100

100

1000

100

100

100

100

100

100

100

100

100

2100

0

1

2100 2100 2100 2100 2100

500

1

1200 1200 1200 1200 1200 1200

1200

0

Distances for Return Ship and Return Standard Truck are set to 0km in Test 2
because it is assumed a single use crate will not be returned.

195

LCA Software Study
Test Parameter Combinations Steel Food Can (Tuna)
Test Parameters
Steel recycled at end-of- life (EOL):
Steel recycled content:
Transport distance, rail:
Transport distance, standard truck:

10, 70, and 100%
25 and 37%.
0 km.
100 km.

Note: Doing all combinations would require 3 x 2 x 1 x 1 = 6 combination.

Parameter Combinations To Be Used
See Table E9 below.
Table E9: Steel food can test parameter combinations
Parameter
Steel Recycled At EOL (%)
Steel Can Recycled Content (%)
Rail (km)
Standard Truck (km)

196

1
70
25
0
100

2
10
25
0
100

Test
3
100
25
0
100

4
70
37
0
100

APPENDIX F: Database Files Used In Software Modeling
Table F1: GaBi library/database files used
File Name
Aluminum Beverage Can
CH: disposal, aluminium, 0% water, to inert material
landfill
CH: disposal, aluminium, 0% water, to municipal
incineration
RER: aluminium, primary, at plant
RER: aluminium, secondary, from old scrap, at plant
RER: cold impact extrusion, aluminium, 2 strokes

Library/Database
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent

RER: sheet rolling, aluminium

US Ecoinvent

US: Transport, combination Truck, average fuel mix
Aseptic Carton
CH: disposal, aluminium, 0% water, to municipal
incineration
CH: disposal, aluminium, 0% water, to sanitary landfill
CH: disposal, packaging paper, 13.7% water, to municipal
incineration
CH: disposal, packaging paper, 13.7% water, to sanitary
landfill
CH: disposal, polyethylene, 0.4% water, to municipal
incineration
CH: disposal, polyethylene, 0.4% water, to sanitary landfill
RER: Aluminium foil
RER: extrusion, plastic film
RER: extrusion, plastic film
RER: extrusion, plastic film
RER: kraft paper, bleached, at plant
RNA: Linear low density polyethylene resin, at plant
RNA: Low density polyethylene resin, at plant
RNA: Polypropylene resin, at plant
US: Transport, combination truck, average fuel mix
Corrugated Box
CH: disposal, packaging cardboard, 19.6% water, to
municipal incineration
CH: disposal, packaging cardboard, 19.6% water, to
sanitary landfill
RER: corrugated board, fresh fibre, single wall, at plant

USLCI/PE

197

US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
PE International
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
USLCI/PE
USLCI/PE
USLCI/PE
USLCI/PE
US Ecoinvent
US Ecoinvent
US Ecoinvent

Table F1 (cont’d)
RER: corrugated board, recycling fibre, single wall, at
plant
RER: packaging box production unit
US: Transport, aircraft, freight
US: Transport, combination truck, diesel powered
US: Transport, combination truck, diesel powered
Flexible Tuna Pouch
CH: disposal, aluminium, 0% water, to municipal
incineration
CH: disposal, aluminium, 0% water, to sanitary landfill
CH: disposal, polyethylene terephthalate, 0.2% water, to
municipal incineration
CH: disposal, polyethylene terephthalate, 0.2% water, to
sanitary landfill
CH: disposal, polyethylene, 0.4% water, to municipal
incineration
CH: disposal, polyethylene, 0.4% water, to sanitary landfill
RER: Aluminium foil
RER: extrusion, plastic film
RER: nylon 6, at plant
RER: polyethylene terephthalate, granulate, amorphous,
at plant
RNA: Polypropylene resin, at plant
US: Transport, combination truck, average fuel mix
Glass Bottle (Beer)
CH: disposal, glass, 0% water, to inert material landfill
CH: disposal, glass, 0% water, to municipal incineration
CH: disposal, polyethylene, 0.4% water, to municipal
incineration
CH: disposal, polyethylene, 0.4% water, to sanitary landfill
CH: disposal, steel, 0% water, to inert material landfill
CH: disposal, steel, 0% water, to municipal incineration
RER: cold impact extrusion, steel, 1 stroke
RER: extrusion, plastic film
RER: packaging glass, brown, at plant
RER: packaging glass, green, at plant
RER: tin plated chromium steel sheet, 2mm, at plant
RNA: Polypropylene resin, at plant
US: Transport, combination Truck, average fuel mix
US: transport, freight, rail, diesel
PET Bottle
198

US Ecoinvent
US Ecoinvent
USLCI/PE
USLCI/PE
USLCI/PE
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
PE International
US Ecoinvent
US Ecoinvent
US Ecoinvent
USLCI/PE
USLCI/PE
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
USLCI/PE
US Ecoinvent

Table F1 (cont’d)
CH: disposal, polyethylene terephthalate, 0.2% water, to
municipal incineration
CH: disposal, polyethylene terephthalate, 0.2% water, to
sanitary landfill
CH: disposal, polyethylene, 0.4% water, to municipal
incineration
CH: disposal, polyethylene, 0.4% water, to sanitary landfill
RER: extrusion, plastic film
RER: injection moulding
RER: polyethylene terephthalate, granulate, bottle grade,
at plant
RER: stretch blow moulding
RNA: Polypropylene resin, at plant
US: Plastic resin secondary (unspecified)
US: Transport, combination Truck, average fuel mix
PLA Bottle
CH: compost, at plant
CH: disposal, polyethylene terephthalate, 0.2% water, to
municipal incineration
CH: disposal, polyethylene terephthalate, 0.2% water, to
sanitary landfill
CH: disposal, polyethylene, 0.4% water, to municipal
incineration
CH: disposal, polyethylene, 0.4% water, to sanitary landfill
RER: extrusion, plastic film
RER: injection moulding
RER: polyethylene, HDPE, granulate, at plant
RER: stretch blow moulding
US: Ingeo Polylactide (PLA) biopolymer production
NatureWorks
US: Transport, combination truck, average fuel mix
Plastic (PP) Crate
CH: disposal, polyethylene, 0.4% water, to municipal
incineration
CH: disposal, polyethylene, 0.4% water, to sanitary landfill
RER: injection moulding
RNA: Polypropylene resin, at plant
US: Transport, aircraft, freight
US: Transport, combination truck, diesel powered
US: Transport, combination truck, diesel powered
US: Transport, ocean freighter, average fuel mix
199

US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
PE International
USLCI/PE
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
US Ecoinvent
PE/NatureWorks
USLCI/PE
US Ecoinvent
US Ecoinvent
US Ecoinvent
USLCI/PE
USLCI/PE
USLCI/PE
USLCI/PE
USLCI/PE

Table F1 (cont’d)
Steel Food Can (Tuna)
CH: disposal, packaging paper, 13.7% water, to municipal
US Ecoinvent
incineration
CH: disposal, packaging paper, 13.7% water, to sanitary
US Ecoinvent
landfill
CH: disposal, steel, 0% water, to inert material landfill
US Ecoinvent
CH: disposal, steel, 0% water, to municipal incineration
US Ecoinvent
RER: cold impact extrusion, steel, 1 stroke
US Ecoinvent
RER: cold impact extrusion, steel, 1 stroke
US Ecoinvent
RER: kraft paper, bleached, at plant
US Ecoinvent
RER: Steel ECCS worldsteel
PE/World Steel
US: Transport, combination truck, average fuel mix
USLCI/PE
Basic Material Comparison - Aluminum
CH: disposal, aluminium, 0% water, to sanitary landfill
US Ecoinvent
RER: aluminium, primary, at plant
US Ecoinvent
RER: aluminium, secondary, from old scrap, at plant
US Ecoinvent
Basic Material Comparison - Corrugated Board
CH: disposal, packaging cardboard, 19.6% water, to
US Ecoinvent
municipal incineration
CH: disposal, packaging cardboard, 19.6% water, to
US Ecoinvent
sanitary landfill
RER: corrugated board, fresh fibre, single wall, at plant
US Ecoinvent
RER: corrugated board, recycling fibre, single wall, at
US Ecoinvent
plant
Basic Material Comparison - Glass
CH: disposal, glass, 0% water, to municipal incineration
US Ecoinvent
CH: disposal, inert material, 0% water, to sanitary landfill
US Ecoinvent
RER: packaging glass, brown, at plant
US Ecoinvent
RER: packaging glass, green, at plant
US Ecoinvent
Basic Material Comparison - PET
CH: disposal, polyethylene terephthalate, 0.2% water, to
US Ecoinvent
municipal incineration
CH: disposal, polyethylene terephthalate, 0.2% water, to
US Ecoinvent
sanitary landfill
RER: polyethylene terephthalate, granulate, bottle grade,
US Ecoinvent
at plant

200

Table F2: openLCA library/database files used
File Name
Library/Database
Basic Material Comparison - Aluminum
Aluminium, primary, at plant/RER
Ecoinvent unit process
Aluminium, secondary, from old scrap, at plant/RER
Ecoinvent unit process
Disposal, aluminium, 0% water, to municipal
Ecoinvent unit process
incineration/CH
Disposal, aluminium, 0% water, to sanitary landfill/CH
Ecoinvent unit process
Recycled Aluminum For CPIS Test
Empty Process
Basic Material Comparison - Corrugated Board
Corrugated board, fresh fibre, single wall, at plant/RER
Ecoinvent unit process
Corrugated board, recycling fibre, single wall, at
Ecoinvent unit process
plant/RER
Disposal, packaging cardboard, 19.6% water, to
Ecoinvent unit process
municipal incineration/CH
Disposal, packaging cardboard, 19.6% water, to sanitary
Ecoinvent unit process
landfill/CH
Recycling Corrugate For CPIS Test
Empty Process
Basic Material Comparison - Glass
Disposal, glass, 0% water, to municipal incineration/CH
Ecoinvent unit process
Disposal, inert material, 0% water, to sanitary landfill/CH
Ecoinvent unit process
Packaging glass, brown, at plant/RER
Ecoinvent unit process
Packaging glass, green, at plant/RER
Ecoinvent unit process
Empty Process
Recycling Glass For CPIS Test
Basic Material Comparison - PET
Disposal, polyethylene terephtalate, 0.2% water, to
Ecoinvent unit process
municipal incineration/CH
Disposal, polyethylene terephtalate, 0.2% water, to
Ecoinvent unit process
sanitary landfill/CH
Polyethylene terephthalate, granulate, bottle grade, at
Ecoinvent unit process
plant/RER
Recycling PET For CPIS Test
Empty Process

201

Table F3: SimaPro library/database files used
File Name
Aluminum Beverage Can
Aluminum can 100% recycled FAL
Aluminum can FAL
Disposal, aluminium, 0% water, to municipal
incineration/CH with US electricity U
Disposal, aluminium, 0% water, to sanitary landfill/CH
with US electricity U
Incineration/CH with US electricity U
Landfill/CH with US electricity U
Recycling aluminium/RER with US electricity U
Transport, combination truck, average fuel mix/US
Aseptic Carton
Aluminium foil B250
Cutting rolls CF
Disposal, aluminium, 0% water, to municipal
incineration/CH with US electricity U
Disposal, aluminium, 0% water, to sanitary landfill/CH
with US electricity U
Disposal, packaging paper, 13.7% water, to municipal
incineration/CH with US electricity U
Disposal, packaging paper, 13.7% water, to sanitary
landfill/CH with US electricity U
Disposal, paper, 11.2% water, to municipal
incineration/CH with US electricity U
Disposal, paper, 11.2% water, to sanitary landfill/CH
with US electricity U
Disposal, polyethylene, 0.4% water, to municipal
incineration/CH with US electricity U
Disposal, polyethylene, 0.4% water, to sanitary
landfill/CH with US electricity U
Extrusion, plastic film/RER with US electricity U
Flexography CF
Incineration/CH with US electricity U
Kraft paper, bleached, at plant/RER with US electricity
U
Landfill/CH with US electricity U
Linear low density polyethylene resin, at plant/RNA
Low density polyethylene resin, at plant/RNA
Polypropylene resin, at plant/RNA
Recycling aluminium/RER with US electricity U
202

Library/Database
Franklin USA 98
Franklin USA 98
US-EI
US-EI
US-EI
US-EI
US-EI
USLCI
BUWAL250
BUWAL250
US-EI
US-EI
US-EI
US-EI
US-EI
US-EI
US-EI
US-EI
US-EI
BUWAL250
US-EI
US-EI
US-EI
USLCI
USLCI
USLCI
US-EI

Table F3 (cont’d)
Recycling paper/RER with US electricity U
Recycling PE/RER with US electricity U
Recycling PP/RER with US electricity U
Transport, combination truck, average fuel mix/US
Corrugated Box
Disposal, packaging cardboard, 19.6% water, to
municipal incineration/CH with US electricity U
Disposal, packaging cardboard, 19.6% water, to
sanitary landfill/CH with US electricity U
Incineration/CH with US electricity U
Landfill/CH with US electricity U
Packaging, corrugated board, mixed fibre, single wall, at
plant/RER with US electricity U
(Note: Modified into two new files, one using FRESH
fiber, and one using RECYCLED fiber.)
Recycling cardboard/RER with US electricity U
Refrigerator, big, A
Transport, aircraft, freight/US
Transport, combination truck, diesel powered/US
Flexible Tuna Pouch
Aluminium foil B250
Disposal, aluminium, 0% water, to municipal
incineration/CH with US electricity U
Disposal, aluminium, 0% water, to sanitary landfill/CH
with US electricity U
Disposal, polyethylene terephtalate, 0.2% water, to
municipal incineration/CH with US electricity U
Disposal, polyethylene terephtalate, 0.2% water, to
sanitary landfill/CH with US electricity U
Disposal, polyethylene, 0.4% water, to municipal
incineration/CH with US electricity
(Note: Also used for disposal of polypropylene and
nonspecific plastics)
Disposal, polyethylene, 0.4% water, to sanitary
landfill/CH with US electricity U
(Note: Also used for disposal of polypropylene and
nonspecific plastics)
Extrusion, plastic film/RER with US electricity U
Gravure printing CF
Incineration/CH with US electricity U

203

US-EI
US-EI
US-EI
USLCI
US-EI
US-EI
US-EI
US-EI
US-EI
US-EI
LCA Food DK
USLCI
USLCI
BUWAL250
US-EI
US-EI
US-EI
US-EI

US-EI

US-EI
US-EI
BUWAL250
US-EI

Table F3 (cont’d)
Laminating solvent free
(Note: Modified to use US electricity.)
Landfill/CH with US electricity U
Nylon 6, at plant/RER with US electricity U
Polyethylene terephthalate, granulate, amorphous, at
plant/RER with US electricity U
Polypropylene resin, at plant/RNA
Production of pouch 100 g
Transport, combination truck, average fuel mix/US
Glass Bottle (Beer)
Crown caps (1 million)
(Note: File modified to use US electricity.)
Disposal, glass, 0% water, to municipal incineration/CH
with US electricity U
Disposal, polyethylene, 0.4% water, to municipal
incineration/CH with US electricity U
(Note: Also used for polypropylene.)
Disposal, polyethylene, 0.4% water, to sanitary
landfill/CH with US electricity U
(Note: Also used for polypropylene.)
Disposal, steel, 0% water, to municipal incineration/CH
with US electricity U
ECCS steel sheet
Extrusion, plastic film/RER with US electricity U
Flexography CF
Glass bottles FAL
Glass bottles recycled FAL
Incineration/CH with US electricity U
Landfill ECCS steel B250(1998)
Landfill Glass B250 (1998)
Landfill/CH with US electricity U
Linear low density polyethylene resin, at plant/RNA
Polypropylene resin, at plant/RNA
Recycling glass/RER with US electricity U
Recycling PE/RER with US electricity U
Recycling PP/RER with US electricity U
Recycling steel and iron/RER with US electricity U
Transport, combination truck, average fuel mix/US
Transport, freight, rail, diesel/US with US electricity U
PET Bottle

204

BUWAL250
US-EI
US-EI
US-EI
USLCI
BUWAL250
USLCI
BUWAL250
US-EI
US-EI

US-EI
US-EI
BUWAL250
US-EI
BUWAL250
Franklin USA 98
Franklin USA 98
US-EI
US-EI
BUWAL250
US-EI
USLCI
USLCI
US-EI
US-EI
US-EI
US-EI
USLCI
US-EI

Table F3 (cont’d)
Disposal, polyethylene terephtalate, 0.2% water, to
municipal incineration/CH with US electricity U
Disposal, polyethylene terephtalate, 0.2% water, to
sanitary landfill/CH with US electricity U
Disposal, polyethylene, 0.4% water, to municipal
incineration/CH with US electricity U
(Note:Also used for polypropylene.)
Disposal, polyethylene, 0.4% water, to sanitary
landfill/CH with US electricity U
(Note:Also used for polypropylene.)
Extrusion, plastic film/RER with US electricity U
Flexography CF
Incineration/CH with US electricity U
Landfill/CH with US electricity U
PET bottles FAL
PET bottles recycled FAL
Polypropylene resin, at plant/RNA
PP caps FAL
Recycling PET/RER with US electricity U
Recycling PP/RER with US electricity U
Transport, combination truck, average fuel mix/US
PLA Bottle
Composting organic waste/RER with US electricity U
Disposal, polyethylene terephtalate, 0.2% water, to
municipal incineration/CH with US electricity U
(Note: Polyethylene terephtalate files used for PLA.)
Disposal, polyethylene terephtalate, 0.2% water, to
sanitary landfill/CH with US electricity U
(Note: Polyethylene terephtalate files used for PLA.)
Disposal, polyethylene, 0.4% water, to municipal
incineration/CH with US electricity U
Disposal, polyethylene, 0.4% water, to sanitary
landfill/CH with US electricity U
Extrusion, plastic film/RER with US electricity U
Flexography CF
Incineration/CH with US electricity U
Injection moulding/RER with US electricity U
Landfill/CH with US electricity U
Polyethylene, HDPE, granulate, at plant/RER with US
electricity U

205

US-EI
US-EI
US-EI

US-EI
US-EI
BUWAL250
US-EI
US-EI
Franklin USA 98
Franklin USA 98
USLCI
Franklin USA 98
US-EI
US-EI
USLCI
US-EI
US-EI

US-EI
US-EI
US-EI
US-EI
BUWAL250
US-EI
US-EI
US-EI
US-EI

Table F3 (cont’d)
Polylactide, granulate, NatureWorks Nebraska/US with
US electricity U
Recycling PE/RER with US electricity U
Stretch blow moulding/RER with US electricity U
Transport, combination truck, average fuel mix/US
Plastic (PP) Crate
Disposal, polyethylene, 0.4% water, to municipal
incineration/CH with US electricity U
Disposal, polyethylene, 0.4% water, to sanitary
landfill/CH with US electricity U
Incineration/CH with US electricity U
Injection moulding/RER with US electricity U
Landfill/CH with US electricity U
Polypropylene resin, at plant/RNA
Recycling PP/RER with US electricity U
Refrigerator, big, A
Transport, aircraft, freight/US
Transport, combination truck, diesel powered/US
Transport, ocean freighter, average fuel mix/US
Steel Food Can (Tuna)
Cold impact extrusion, steel, 1 stroke/RER with US
electricity U
Disposal, packaging paper, 13.7% water, to municipal
incineration/CH with US electricity U
Disposal, packaging paper, 13.7% water, to sanitary
landfill/CH with US electricity U
Disposal, paper, 11.2% water, to municipal
incineration/CH with US electricity U
Disposal, paper, 11.2% water, to sanitary landfill/CH
with US electricity U
Disposal, steel, 0% water, to municipal incineration/CH
with US electricity U
ECCS steel 100% scrap
ECCS steel sheet
Flexography CF
Incineration/CH with US electricity U
Kraft paper, bleached, at plant/RER with US electricity
U
Landfill ECCS steel B250(1998)
Landfill/CH with US electricity U
Recycling paper/RER with US electricity U
206

US-EI
US-EI
US-EI
USLCI
US-EI
US-EI
US-EI
US-EI
US-EI
USLCI
US-EI
LCA Food DK
USLCI
USLCI
USLCI
US-EI
US-EI
US-EI
US-EI
US-EI
US-EI
BUWAL250
BUWAL250
BUWAL250
US-EI
US-EI
BUWAL250
US-EI
US-EI

Table F3 (cont’d)
Recycling steel and iron/RER with US electricity U
US-EI
Transport, combination truck, average fuel mix/US
USLCI
Basic Material Comparison - Aluminum
Aluminium, primary, at plant/RER U
Ecoinvent unit process
Aluminium, secondary, from old scrap, at plant/RER U
Ecoinvent unit process
Disposal, aluminium, 0% water, to municipal
Ecoinvent unit process
incineration/CH U
Disposal, aluminium, 0% water, to sanitary landfill/CH U Ecoinvent unit process
Incineration/CH U
Ecoinvent unit process
Landfill/CH U
Ecoinvent unit process
Recycling aluminium/RER U
Ecoinvent unit process
Basic Material Comparison - Corrugated Board
Corrugated board, fresh fibre, single wall, at plant/RER
Ecoinvent unit process
U
Corrugated board, recycling fibre, single wall, at
Ecoinvent unit process
plant/RER U
Disposal, packaging cardboard, 19.6% water, to
Ecoinvent unit process
municipal incineration/CH U
Disposal, packaging cardboard, 19.6% water, to
Ecoinvent unit process
sanitary landfill/CH U
Incineration/CH U
Ecoinvent unit process
Landfill/CH U
Ecoinvent unit process
Recycling cardboard/RER U
Ecoinvent unit process
Basic Material Comparison - Glass
Disposal, glass, 0% water, to municipal incineration/CH
U
Ecoinvent unit process
Disposal, inert material, 0% water, to sanitary landfill/CH
U
Ecoinvent unit process
Incineration/CH U
Ecoinvent unit process
Landfill/CH U
Ecoinvent unit process
Packaging glass, brown, at plant/RER U
Ecoinvent unit process
Packaging glass, green, at plant/RER U
Ecoinvent unit process
Recycling glass/RER U
Ecoinvent unit process
Basic Material Comparison - PET
Disposal, polyethylene terephtalate, 0.2% water, to
Ecoinvent unit process
municipal incineration/CH U
Disposal, polyethylene terephtalate, 0.2% water, to
Ecoinvent unit process
sanitary landfill/CH U
Incineration/CH U
Ecoinvent unit process
Landfill/CH U
Ecoinvent unit process

207

Table F3 (cont’d)
Polyethylene terephthalate, granulate, bottle grade, at
plant/RER U
Recycling PET/RER U

208

Ecoinvent unit process
Ecoinvent unit process

APPENDIX G: Packaging Systems Comparison Data
Table G1: Beverage container comparison data
Impact Category
Aluminum Can
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
PET Bottle
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
Glass Bottle
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
Aseptic Carton
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
PLA Bottle
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion

Unit

COMPASS

SimaPro

GaBi

Package
Modeling

kg CO2 equiv

0.1826

0.1849

0.2437

0.3044

MJ Energy

2.3295

2.9253

3.4024

kg PO4 equiv
Liters

7.494E-05
9.569E-01

2.775E-05
1.075E-02

3.819E-04
4.201E+03

kg CO2 equiv

0.4050

0.3313

0.3218

MJ Energy

8.1255

7.6478

8.1523

kg PO4 equiv
Liters

3.886E-04
9.274E-01

4.738E-06
4.218E-02

5.069E-04
1.234E+03

kg CO2 equiv

0.5072

0.4780

0.5946

MJ Energy

8.1489

6.9304

10.1460

kg PO4 equiv
Liters

3.555E-04
5.216E+00

4.220E-06
1.346E-01

5.898E-04
4.180E+03

kg CO2 equiv

0.0958

0.1252

0.0498

MJ Energy

1.9329

2.7602

1.8314

kg PO4 equiv
Liters

0.3149

0.0926

2.190E-04 2.676E-05 5.449E-05
3.570E+00 2.277E+00 3.767E+02

kg CO2 equiv

0.1988

0.2416

0.0644

MJ Energy

3.9037

3.8169

1.7235

kg PO4 equiv
Liters

0.1448

6.955E-04 7.797E-05 1.086E-04
1.082E+00 1.022E+00 2.802E+02

209

0.0964

Table G2: Truck and rail transport comparison data for glass bottle
Transport
Mode

Distance
(km)

Unit

COMPASS

SimaPro

GaBi

Truck

100

kg CO2 equiv

0.5072

0.4780

0.5946

Truck

1000

kg CO2 equiv

0.6004

0.5218

0.6386

Rail

0

kg CO2 equiv

0.5014

0.4640

0.5814

Rail

500

kg CO2 equiv

0.5072

0.4780

0.5946

Rail

4000

kg CO2 equiv

0.5473

0.5765

0.6868

Table G3: Steel can and flexible pouch comparison data
Impact Category
Steel Can - 25% to
28% recycled
1
content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
Steel Can - 37%
recycled content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
Flexible Pouch
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
1

SimaPro

GaBi

Package
Modeling

kg CO2 equiv

0.8221

0.7096

0.6512

MJ Energy

10.5972

7.4115

Unit

COMPASS

7.588E-05 2.100E-04
6.413E+00 5.021E+02

kg PO4 equiv
liters

kg CO2 equiv

0.4796

0.7768

MJ Energy

6.4595

10.1941

kg PO4 equiv
liters

3.470E-04 7.352E-05
4.524E+00 6.413E+00

kg CO2 equiv

0.4073

0.3889

0.3466

MJ Energy

8.0531

9.4005

7.8355

4.904E-04
1.014E+00

8.805E-05
6.866E-01

1.719E-04
1.459E+03

kg PO4 equiv
liters

25% recycled content for SimaPro and GaBi, 28% for Package Modeling.

210

0.2248

Table G4: Reuse comparison data on a per use basis
Impact Category
PP crate
1 use
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
PP crate
10 uses
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
PP crate
100 uses
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
Corrugated Box
1 use
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion

Unit

COMPASS

SimaPro

GaBi

Package
Modeling

kg CO2 equiv

6.6544

5.9228

5.3822

2.5706

186.9295

200.8648

181.6887

kg PO4 equiv
liters

9.040E-03
1.259E+01

1.865E-04
8.590E+00

4.466E-03
1.119E+04

kg CO2 equiv

1.3747

1.1062

1.0102

MJ Energy

28.6723

27.4957

24.9602

kg PO4 equiv
liters

1.630E-03
2.252E+00

2.292E-05
8.593E-01

4.466E-04
1.119E+03

kg CO2 equiv

0.8467

0.6245

0.5995

MJ Energy

12.8465

10.1588

9.6540

kg PO4 equiv
liters

8.892E-04
1.219E+00

6.560E-06
8.619E-02

4.466E-05
1.119E+02

kg CO2 equiv

0.8642

1.0803

0.4868

MJ Energy

13.6383

16.5080

12.5507

1.379E-03
1.788E+01

2.027E-04
1.879E+01

1.065E-03
2.215E+03

MJ Energy

kg PO4 equiv
liters

211

0.6095

Table G5: Recycled content comparison data for corrugated box
Impact Category
0% Recycled Content
1
(12% for COMPASS )
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
25% Recycled Content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
50% Recycled Content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
100% Recycled Content
(87% for COMPASS 1)
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
1

Unit

COMPASS

SimaPro

GaBi

kg CO2 equiv

0.5405

1.1821

0.0276

MJ Energy

13.9916

17.6705

13.0990

kg PO4 equiv
liters

1.616E-03
2.920E+01

3.356E-04
3.113E+01

1.467E-03
3.559E+03

kg CO2 equiv

0.6512

1.1312

0.2572

MJ Energy

13.8707

17.0892

12.8248

kg PO4 equiv
liters

1.535E-03
2.533E+01

2.692E-04
2.496E+01

1.266E-03
2.887E+03

kg CO2 equiv

0.8642

1.0803

0.4868

MJ Energy

13.6383

16.5080

12.5507

kg PO4 equiv
liters

1.379E-03
1.788E+01

2.027E-04
1.879E+01

1.065E-03
2.215E+03

kg CO2 equiv

1.1793

0.9784

0.9460

MJ Energy

13.2944

15.3454

12.0024

1.148E-03
6.855E+00

6.978E-05
6.447E+00

6.635E-04
8.704E+02

kg PO4 equiv
liters

COMPASS limited recycled content of corrugated board to range of 12% to 87%.

212

Table G6: Air and ship transport comparison data for corrugated box and PP crate
Container
Corrugated
Box
1 use
Corrugated
Box
1 use

Transport Distance
Mode
(km)

1,2

Unit

COMPASS

3

SimaPro

3

GaBi

3

Air

500

kg CO2
equiv

0.7805

0.9919

0.4019

Air

2500

kg CO2
equiv

0.8642

1.0803

0.4868

PP Crate
10 uses

Air

500

1.1600

0.8878

0.7921

PP Crate
10 uses

Air

2500

kg CO2
equiv
kg CO2
equiv

1.3747

1.1062

1.0102

PP Crates
10 uses

Ship

500

1.3285

1.0594

0.9634

PP Crate
10 uses

Ship

2100

kg CO2
equiv
kg CO2
equiv

1.3747

1.1062

1.0102

1
2
3

Air distances are out going transport on per use basis.
Ship distances are return transport on per use basis, except no return on 10th use.
kg CO2 equivalent values are average per use.

213

APPENDIX H: Basic Materials Comparison Data
Table H1: Basic material comparison data
Impact Category
Aluminum
50% recycled content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
Corrugated Board
50% recycled content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
Glass
55.5% recycled content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
PET
0% recycled content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion

Unit

COMPASS

SimaPro

GaBi

kg CO2 equiv

5.7539

5.8449

6.9588

MJ Energy

71.5823

91.6934

88.3941

kg PO4 equiv
liters

2.600E-03

1.523E-03

9.270E-03

3.171E+01

3.174E+01

1.651E+05

kg CO2 equiv

0.6838

0.9455

1.0262

MJ Energy

12.2068

15.2014

13.2740

kg PO4 equiv
liters

1.464E-03
2.371E+01

3.955E-04
2.462E+01

9.804E-04
1.484E+03

kg CO2 equiv

0.7831

0.8655

0.8020

MJ Energy

13.2333

15.5838

14.3383

kg PO4 equiv
liters

1.066E-03
8.377E+00

1.494E-04
8.641E+00

6.620E-04
1.652E+03

kg CO2 equiv

3.1510

3.2656

3.2698

MJ Energy

70.1882

82.3845

76.1822

5.800E-03
1.151E+01

5.818E-04
1.476E+01

3.040E-03
6.990E+03

kg PO4 equiv
liters

214

Table H2: Basic corrugated board recycled content comparison data
Impact Category
12% recycled content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
50% recycled content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion
87% recycled content
Global Warming
Fossil Fuel/NonRenewable Energy
Eutrophication
Water Depletion

Unit

COMPASS

SimaPro

GaBi

kg CO2 equiv

0.0927

0.9457

0.1405

MJ Energy

11.2234

15.2289

14.3315

kg PO4 equiv
liters

1.765E-03
4.018E+01

5.628E-04
3.703E+01

1.755E-03
4.077E+03

kg CO2 equiv

0.6838

0.9455

0.5892

MJ Energy

12.2068

15.2014

13.7957

kg PO4 equiv
liters

1.464E-03
2.371E+01

3.955E-04
2.462E+01

1.363E-03
2.763E+03

kg CO2 equiv

1.2594

0.9453

1.0262

MJ Energy

13.1644

15.1745

13.2740

1.172E-03
7.669E+00

2.325E-04
1.254E+01

9.804E-04
1.484E+03

kg PO4 equiv
liters

215

Table H3: openLCA basic material comparison data
Impact Category
Aluminum
50% recycled content
Global Warming
Non-Renewable Energy
Eutrophication
Water Depletion
Corrugated Board
50% recycled content
Global Warming
Non-Renewable Energy
Eutrophication
Water Depletion
Glass
50% recycled content
Global Warming
Non-Renewable Energy
Eutrophication
Water Depletion
PET
0% recycled content
Global Warming
Non-Renewable Energy
Eutrophication
Water Depletion

Unit

SimaPro

openLCA

kg PO4 equiv
liters

5.903E-04
6.033E-04
1.523E-03
3.174E+01

4.603E-04
3.733E-04
2.524E-04
1.600E+01

points
points

9.549E-05
1.000E-04

5.283E-05
5.522E-05

kg PO4 equiv
liters

1.464E-03
2.371E+01

2.908E-04
2.137E+01

points
points

8.781E-05
1.029E-04
1.512E-04
8.995E+00

2.506E-05
4.213E-05
5.920E-05
7.580E-02

3.298E-04
5.421E-04
5.818E-04
1.476E+01

1.545E-04
3.737E-04
1.511E-04
4.898E+00

points
points

kg PO4 equiv
liters

points
points
kg PO4 equiv
liters

216

Table H4: openLCA corrugated board recycled content comparison data
Impact Category
12% recycled content
Global Warming
Non-Renewable Energy
Eutrophication
Water Depletion
50% recycled content
Global Warming
Non-Renewable Energy
Eutrophication
Water Depletion
87% recycled content
Global Warming
Non-Renewable Energy
Eutrophication
Water Depletion

Unit

SimaPro

openLCA

points
points

9.551E-05
1.002E-04

4.812E-05
5.084E-05

kg PO4 equiv
liters

1.765E-03
4.018E+01

4.522E-04
3.255E+01

points
points

9.549E-05
1.000E-04
1.464E-03
2.371E+01

5.283E-05
5.522E-05
2.908E-04
2.137E+01

9.547E-05
9.985E-05
1.172E-03
7.669E+00

5.743E-05
5.949E-05
1.337E-04
1.048E+01

kg PO4 equiv
liters
points
points
kg PO4 equiv
liters

217

BIBLIOGRAPHY

218

BIBLIOGRAPHY

Aissani L, M. Mollaret, and M Pradel. 2009. Rapport de Synthese, Comparaison de
deux logiciels d’ACV GaBi 4 et SimaPro 7.1.8. [Synthetic Report, Comparison of
LCA Software GaBi 4 and SimaPro 7.1.8.] Paper presented at Atelier CemACV2,
15-16 September. http://cemadoc.cemagref.fr/exl-php/util/documents/accede
_document.php. Accessed March 2013.
ALCAS (Australian Life Cycle Assessment Society). 2013. History. http://www.alcas.
asn.au/about-lca/history. Accessed August 2013.
Aluminum Association. 2011. Aluminum Can Recycling Rate Reaches Highest Level in
More than a Decade. www.aluminum.org/AM/Template.cfm?Section=2011&CON
TENTID=31569&TEMPLATE=/CM/ContentDisplay.cfm. Accessed May 2013.
American Iron and Steel Institute. 2010. North America Steel Industry Honors the Steel
Food Can. www.steel.org/sitecore/content/Global/Document%20Types/News/
2010/Container%20-%20North%20American%20Steel%20Industry%20Honors%
20the%20Steel%20Food%20Can.aspx. Accessed May 2013.
Bare, J., G. Norris, D. Pennington, and T. McKone. 2002. TRACI: The Tool for the
Reduction and Assessment of Chemical and Other Environmental Impacts.
Journal of Industrial Ecology 6 (3-4): 49-78.
Bare, J. 2011. TRACI 2.0: the tool for the reduction and assessment of chemical and
other environmental impacts 2.0. United States Environmental Protection
Agency. Published on line January 21, 2011. Copy provided by Jane Bare, email: bare.jane@epa.gov.
Bare, J. 2012. Tool for the Reduction and Assessment of Chemical and other
Environmental Impacts (TRACI), Version 2.1, User’s Manual. United States
Environmental Protection Agency, ORD/NRMRL/Sustainable Technology
Division, Systems Analysis Branch, SOP No. S-10637-OP-1-0.
Bloomberg Markets. 2013. The Top 20 Emerging Markets. www.bloomberg.com/slide
show/2013-01-30/the-top-20-emerging-markets.html#slide1. Accessed April
2013.
Bumble Bee Foods. 2011. FAQ – Current Topics – Recycling (can vs. pouch).
www.bumblebee.com/About/FAQS/CurrentTopics.aspx. Accessed May 2013.
Ciambrone, D. F. 1997. Environmental Life Cycle Analysis. Boca Raton: Lewis
Publishers.

219

Corrugated Packaging Alliance. 2010. Corrugated Packaging Life-cycle Assessment
Summary Report. www.corrugated.org/upload/LCA%20Summary%20Report%
20FINAL%203-24-10.pdf. Accessed May 2013.
CNW (Canada Newswire). 2005. Refillable Beer Bottle Contributes to Cleaner Air.
www.container-recycling.org/assets/pdfs/CNW-RefillableBeerBottle-4-28-05.pdf.
Accessed April 2013.
EC-JRC (European Commission-Joint Research Centre). 2013. Life Cycle Thinking and
Assessment. http://lct.jrc.ec.europa.eu/. Accessed April 2013.
EC-JRC-IES (European Commission - Joint Research Centre - Institute for Environment
and Sustainability). 2010. International Reference Life Cycle Data System (ILCD)
Handbook – General guide for Life Cycle Assessment – Detailed guidance.
Luxembourg: Publications Office of the European Union.
Ecoinvent Center (Swiss Centre for Life Cycle Inventories). 2013. The ecoinvent
Database. www.ecoinvent.org/database/ . Accessed May 2013.
EEA (European Environment Agency). 1997. Life Cycle Assessment - A guide to
approaches, experiences and information sources. Environmental Issues Series
no. 6. Copenhagen: EEA.
EIA (United States Energy Information Administration). 2013. International Energy
Statistics.www.eia.gov/cfapps/ipdbproject/iedindex3.cfm?tid=5&pid=5&aid=2&cid
=ww,&syid=1980&eyid=2012&unit=TBPD. Accessed April 2013.
EPA (United States Environmental Protection Agency). 2006. Life Cycle Assessment:
Principles And Practice.www.epa.gov/nrmrl/std/lca/lca.html. Accessed April 2013.
EPA (United States Environmental Protection Agency). 2009. Municipal Solid Waste In
The United States, 2009 Facts and Figures. http://epa.gov/wastes/nonhaz/
municipal/pubs/msw2009rpt.pdf. Accessed May 2013.
EPA (United States Environmental Protection Agency). 2010. Recycled Content
(ReCon) Tool. www.epa.gov/wastes/conserve/tools/warm/ReCon_Online.html.
Accessed May 2013.
EPA (United States Environmental Protection Agency). 2011. Municipal Solid Waste
Generation, Recycling, and Disposal in the United States Tables and Figures for
2010. www.epa.gov/wastes/nonhaz/municipal/pubs/2010_MSW_Tables_and_
Figures_508.pdf. Accessed April 2013.
EPA (United States Environmental Protection Agency). 2013. What is sustainability?
www.epa.gov/sustainability/basicinfo.htm#sustainability. Accessed April 2013.

220

Glass Packaging Institute. 2010. Environmental Overview – Complete Life Cycle
Assessment of North American Container Glass. www.gpi.org/sites/default/
files/N-American_Glass_Container_LCA.pdf. Accessed May 2013.
Goedkoop, M., and R. Spriensma. 2001. The Eco-indicator 99: A damage oriented
method for Life Cycle Impact Assessment, Methodology Report, Third Edition.
PRé Consultants bv. www.pre-sustainability.com/download/misc/EI99_
methodology_v3.pdf. Accessed April 2013.
Goedkoop, M., M. Oele, A. de Schryver, and M. Vieira. 2008. SimaPro 7 Database
manual – Methods library. The Netherlands: PRé Consultants. http://www.presustainability.com/download/manuals/DatabaseManualMethods.pdf. Accessed
July 2013.
Goedkoop, M., R. Heijungs, M. Huijbregts, A. Schryver, J. Struijs, and R. van Zelm.
2013. ReCiPe 2008: A life cycle assessment method which comprises
harmonized category indicators at the midpoint and the endpoint level, First
edition (revised), Report I: Characterization. http://www.lcia-recipe.net/. Accessed
May 2013.
Gonzalez-García, S., D. García-Rey, and A. Hospido. 2013. Environmental life cycle
assessment for rapeseed-derived biodiesel. International Journal Of Life Cycle
Assessment 18 (1): 61-76.
Guinée, J. ed. 2002. Handbook on Life Cycle Assessment: Operational Guide to the
ISO Standards. Dordrecht, The Netherlands: Kluwer Academic Publishers.
Heijungs, R. and S. Suh. 2002. The Computational Structure of Life Cycle Assessment.
Dordrecht, The Netherlands: Kluwer Academic Publishers.
Humbert, S., M. Margni, O. Jolliet. 2005. IMPACT 2002+: USER GUIDE – Draft for
Version 2.1. Lausanne, Switzerland: Industrial Ecology & Life Cycle Systems
Group, GECOS, Swiss Federal Institute of Technology Lausanne (EPFL).
Hunt, R. G. and W. E. Franklin. 1996. LCA – How it Came About. International Journal
Of Life Cycle Assessment 1 (1): 4-7.
ISO (International Organization for Standards). 2006. Environmental management – Life
cycle assessment – Principles and framework. ISO 14040. Geneva: International
Organization for Standardization.
ISO (International Organization for Standards). 2006. Environmental management – Life
cycle assessment – Requirements and guidelines. ISO 14044. Geneva:
International Organization for Standardization.

221

Jolliet, O., M. Margni, R. Charles, S. Humbert, J. Payet, G. Rebitzer, R. Rosenbaum.
2003. IMPACT 2002+: A New Life Cycle Impact Assessment Methodology.
International Journal of Life Cycle Assessment 8(6): 324-330.
Karp, D. 2011. Market Watch: Santa Monica’s ban on plastic bags – what it means. Los
Angeles Times, 4 February. http://www.latimes.com/features/food/la-fomarketwatch-20110204,0,4142620.story. Accessed April 2013.
Kendal, A., J. Yuan, and Sonja B. Brodt. 2013. Carbon footprint and air emissions
inventories for US honey production: case studies. International Journal Of Life
Cycle Assessment 18 (2): 392-400.
Lewis, H., K. Verghese, and L. Fitzpatrick. 2010. Evaluating the sustainability impacts of
packaging: the plastic carry bag dilemma. Packaging Technology And Science
23: 145-160.
Lee Y and J. Shan. 2000. A preliminary comparison and integration of lifecycle impact
assessment models. Journal of Chinese Institute of Environmental Engineering
10: 323-335.
Lifset, R. 2013. Raising the Bar for Symbiosis, Life Cycle Assessment, and Material
Flow Analysis Case Studies. Journal of Industrial Ecology 17 (1): 1.
Lin, L. 2012. China Slowing Auto Sales Still Eclipse U.S.-Japan-Germany: Cars.
Bloomberg News, 4 September. www.bloomberg.com/news/2012-09-04/chinaslowing-auto-sales-still-eclipse-u-s-japan-germany-cars.html. Accessed April
2013.
Lin, R. 2010. L.A. County passes sweeping ban on plastic bags. Los Angeles Times, 16
November. http://latimesblogs.latimes.com/lanow/2010/11/la-county-passessweeping-ban-on-plastic-bags.html. Accessed April 2013.
Mistry, M. 2010. COMPASS – Manual – Part 3 – Background and Technical Guide –
Version 2.0. Charlottesvill,VA: Sustainable Packaging Coalition.
Mourad, A., E. Garcia, G. Vilela, F. von Zuben. 2008. Environmental effects from a
recycling rate increase of cardboard of aseptic packaging system for milk using
life cycle approach. International Journal Of Life Cycle Assessment 13 (2): 140146.
NREL (National Renewable Energy Laboratory). 2013. U.S. Life Cycle Inventory
Database. www.nrel.gov/lci/. Accessed May 2013.
openLCA. 2011. openLCA Life Cycle Assessment software. http://openlca.org/web/
guest. Accessed May 2013.

222

Pawelzik, P., M. Carus, J. Hotchkiss, R. Narayan, S. Selke, M. Wellisch, M. Weiss, B.
Wicke, and M. Patel. 2013. Critical aspects in the life cycle assessment (LCA) of
bio-based materials – Reviewing methodologies and deriving recommendations.
Resources, Conservation, and Recycling 73: 211-228.
PE International . 2011. GaBi. www.gabi-software.com/software/gabi-5/. Accessed May
2013.
PRé Consultants. 2011. SimaPro. www.pre-sustainability.com/software. Accessed May
2013.
Rail Passenger USA. 2011. Route Atlas. http://railpassengerusa.com/. Accessed July
2013.
ReCiPe. 2013. ReCiPe methodology for Life Cycle Impact Assessment. www.lciarecipe.net/. Accessed July 2013.
Romero-Hernández, O., S. Hernández, D. Muñoz, E. Detta-Silveira, A. Palacios-Brun,
A.Laguna.. 2009. Environmental implications and market analysis of soft drink
packaging systems in Mexico. A waste management approach. International
Journal Of Life Cycle Assessment 14 (2): 107-113.
SAIC (Scientific Applications International Corporation), 2006. Life Cycle Assessment:
Principles and Practice (referred to as LCA 101 document on EPA website).
EPA. www.epa.gov/nrmrl/std/lca/lca.html. Accessed April 2013.
Sand, C. 2010. The Packaging Value Chain. Lancaster, Pennsylvania: DEStech
Publications, Inc.
Saunders, C., A. Barber, and G. Taylor. 2006. Food Miles – Comparative
Energy/Emissions Performance of New Zealand’s Agriculture Industry. Lincoln
University Agribusiness and Economics Research Unit research report no. 285.
Christchurch, New Zealand.
Simon, R., E. Rice, T. Kingsbury, and D. Dornfeld. 2012. A Comparison of Life Cycle
Assessment (LCA) Software in Packaging Applications. Berkeley, California:
Laboratory For Manufacturing And Sustainability, University Of California,
Berkeley, College Of Engineering.
Sonneveld, K. 2000. The Role Of Life Cycle Assessment as a Decision Support Tool for
Packaging. Packaging Technology And Science 13: 55-61.
SPC (Sustainable Packaging Coalition). 2011. COMPASS (Comparative Packaging
Assessment). www.design-compass.org. Accessed March 2013.

223

UN ESA (United Nations, Department of Economic and Social Affairs). 2013.
Population. http://esa.un.org/wpp/unpp/panel_population.htm. Accessed 2013.
U.S. Census (United States Census Bureau). 2002. Global Population Profile.
www.census.gov/population/international/files/wp02/wp-02003.pdf. Accessed
April 2013.
U.S. Census (United States Census Bureau). 2012. Population – U.S. Census Bureau,
Statistical Abstract of the United States: 2012. www.census.gov/prod/2011pubs/
12statab/pop.pdf. Accessed April 2013.
U.S. Census (United States Census Bureau). 2013. U.S. and World Population Clock.
www.census.gov/popclock/. Accessed April 2013.
Verghese, K., H. Lewis, and L. Fitzpatrick, ed. 2012. Packaging for Sustainability.
London: Spinger-Verlag.
Villanueva, A., and H. Wenzel. 2007. Paper Waste - Recycling, Incineration Or
Landfilling? A Review of Existing Life Cycle Assessments. Waste Management
27(8): S29-S46.
WCED (World Commission on Environment and Development). 1987. Our Common
Future. http://www.un-documents.net/our-common-future.pdf. Accessed
November 8, 2013.
Wever, Renee. 2009. Thinking-about-the-Box, A Holistic Approach to Sustainable
Design Engineering of Packaging for Durable Consumer Goods. Delft, The
Netherlands: VSSD.

224