An Approach to
Dynamic Environmental Life-Cycle Assessment
by Evaluating Structural Economic Sequences
CHAPTER 1 – Introduction
Section 1.1 – Introduction
Life–cycle Assessment (LCA) is a holistic approach to analyzing environmental burdens of a product, process or activity. LCA attempts to analyze each step in the life–cycle by identifying the energy and materials used and in the most comprehensive case, determining impacts to the environment. LCA is currently being implemented by industry and used to establish certification criteria by government. Further, it is on the verge of becoming an established international standard for environmental performance analysis. However, there are several problems that remain in the LCA methodology. The problems are critical, and raise questions regarding overall effectiveness and practicability of implementation.
The conventional approach to LCA has traditionally been a static engineering or technical exercise with little concern towards social, economic and temporal aspects. The following research presents a method to expand the LCA methodology to address these shortcomings. This research presents and applies a dynamic life-cycle emissions methodology based on the philosophy of structural economics and the constructs of dynamic economic Input-Output interindustry modeling. Specifically, emphasis will be on:
· the importance of context in a cross-disciplinary perspective to create an appropriate analysis and presentation of results,
· the importance of a dynamic approach to environmental Life-Cycle Assessment,
· the application of a dynamic economic Input-Output (I–O) model applied to the environmental LCA methodology,
· the use of experimental design constructs to generate scenarios, and,
· the presentation of a case study to assist in the investigation process.
The methodology to be applied will be based on the philosophy of structural economics (Duchin 1998) and the constructs of the Sequential Interindustry Model (SIM) approach (Romanoff and Levine 1976, 1981, 1986, 1990b, 1991; Levine and Romanoff, 1989). Moreover, scenarios will be developed that embrace the constructs of experimental design to minimize efforts to achieve rigorous results (Box, Hunter, and Hunter 1978).
This study will explore one case study under several scenarios to demonstrate the use of LCA in this context. The case study presented is the environmental LCA of a methanol fuel cell electric vehicle (FCEV). To present the methodology with clarity, the study will streamline the analysis by examining one LCA classification criteria, green house gas (GHG) emissions. Thus, the case study seeks to answer the following question:
What is the net decrease or increase in green house gas emissions by introducing methanol FCEVs in the U.S. economy for the period of 20 years, 2000 to 2020?
In summary, the emphasis of this research is to demonstrate the steps necessary to conduct a dynamic LCA study by embracing the constructs of structural economics with emphasis on utilizing a dynamic input-output framework. Emphasis will be on the development and implementation of a methodological approach to conduct an LCA study. The methodological approach will incorporate endogenous, as well as, exogenous changes such as technological change and changes in consumer demand as they affect environmental impacts in a dynamic framework. For reasons of data availability, this research will not result in the presentation of empirical results for the use of actual environmental problems or policy decisions. Emphasis will be on the practicability of the method presented.
Section 1.2 – Hypothesis and Intent of Dissertation
Life Cycle Assessment (LCA) is a comprehensive approach to analyzing environmental burdens of a product, process or activity. The conventional approach to LCA has traditionally been a static engineering or technical approach. However, the potential to substantially reduce all forms of environmental degradation cannot solely rely on engineering studies of new and existing technologies. The reduction of environmental burdens also depends on the examination of economic instruments, government regulation and legislation, education, and ultimately, the initiative, support and cooperation of citizens in their various public and private capacities (Duchin 1998). And as emphasized by Graedel and Allenby (1998):
"A fundamental mistake make by many scientists and technologists studying industrial ecology is to assume that the major issues and problems are technical in nature. This is, in fact infrequently the case. Rather, while elements of new technological systems or scientific data on, for example, the lifecycle impact of specific materials, may certainly need be developed, it is often non-technical issues that prove most unfathomable to industry in general and the industrial ecologist in particular" (Graedel and Allenby 1998, 26).
Thus, the proposed research seeks to demonstrate the application of structural economics, a methodology that presents a contextual framework for integrating these attributes, as an approach to LCA. Structural economics is a method of empirical research "concerned with describing the state, or structure of an economic system and with the quantitative and qualitative changes that take place in that structure with the passage of time" (Duchin 1998, 10). The economic structure is defined in terms of production and consumption activities, both physically and socially. The economy is treated as a comprehensive system, capturing the interdependency of all activities. A by-product of this holistic approach is the broadening of a scope beyond conventional economics that leads to collaboration with other disciplines (Duchin 1998).
Each discipline of analysis, economic, engineering, and the social sciences, has particular strengths to analyze problems within its respective area. However, interdependence among each discipline can bring together complementary elements that create truly holistic solutions to environmental problems. Significant advances in the reduction of burdens on the environment can only be achieved through the combined understanding of technological possibilities, economic incentives and societal preferences (Duchin 1994). The optimal "vector of change," that is, the several indices that make up the web of industrial activities and consumer behavior necessary to change in order to realize an overall reduction of environmental impacts, can be best realized utilizing the complementary analysis of an economic framework combined with engineering expertise.
The intent of LCA is to present a methodology and framework to systematically generate the quantitative criteria to support policy decisions. The quantitative criteria encompasses the collection of materials and energy inputs and outputs of the life of a product, process or activity as shown in Figure 1-1. However, this static perspective is insufficient to meet the demands of contemporary LCA applications. Contemporary applications for LCA studies include: product development, strategic planning, public policy making, marketing predictions and sales projections. Each of these applications requires a method that does not limit the analysis to the status quo, but also includes what will be in the future. A predictive method that includes the intricacies of temporal distinctions is necessary to accomplish these tasks.
Fig. 1-1. General materials flow diagram for a product life cycle (Boguski 1996).
Providing a predictive methodology that examines temporal distinctions is only part of the solution. To support decision-making, policy analysts, ecological economists, and business strategists alike, need to be able to assess the costs and benefits of environmentally preferable choices. Once identified, decision-makers can develop the appropriate incentives to increase the likelihood that the preferable choices will happen. To do this, they need to take on the difficult "how" questions. The costs and the implications of preferable environmental choices depend on "how" those options can be achieved (Duchin 1992). Thus, the strength of the structural economic framework is its integration of the ecological economist's and environmental engineer's concerns in a system-wide perspective that is suitable for comparing the implications of alternative future courses of actions (Duchin 1998).
The intent of this research is to investigate the appropriateness of structural economics to satisfy the holistic contextual perspective of the LCA methodology in a temporal framework. Specifically, this research will focus on the key elements of structural economics using Sequential Interindustry Modeling (SIM). SIM creates a temporal framework by distributing the static I-O modeling activities over time. Thus, SIM provides two views, a temporal view, which is associated with the timing of each industry’s production schedules and a cross-sectional view, where one observes activities at a specific time (Romanoff and Levine 1991). This approach can be readily applied to the investigation of scenarios involving structural change and changes in final demand as they affect resource use.
Input-Output economic modeling, because of its logical structure and consistency with the physical concept of materials balance, provides a promising framework to build models needed to evaluate alternative economic-environmental options for the future (Cumberland and Stram 1974). A basic consideration underlying environmental pollution analysis is the materials balance, i.e. the input of materials and energy in economic activities eventually are disposed of in the environment as residuals (gaseous, liquid, solid, heat and noise) or accumulated within activities as machinery, buildings, and consumer durables (Førsund 1985) (Ayres and Kneese 1969). Moreover, it bridges the gap between economists, plant managers and engineers by establishing a platform on which they can readily communicate (Rose 1995). Further, economic I-O analysis is an integral part of most Computable General Equilibrium models (CGEs). CGEs are macroeconomic models that account for supply, demand and non-linear functions through mechanisms of profit maximization, utility maximization and market clearing. The majority of CGEs are based on a set of social accounts, they explicitly incorporate resource constraints, allow for input substitution, and have a strong price-quantity integration. Further, they can accommodate international and interregional competition through spatial differentiated prices and factor mobility (Rose 1995). The advantage of conducting a LCA within a framework that is intrinsic to policymaking, such as CGEs, is the conveyance of information in a form that is immediately understood by a broader audience.
Illustrative Example
The following is an example that illustrates the importance of conducting an LCA in a dynamic framework.
Consider the life cycle assessment of a product that involves an emerging technology. The problem posed in a life-cycle assessment context is:
What are the environmental impacts/benefits of introducing an emerging vehicle technology in the U.S. economy? Specifically, what are the net increases or decreases in materials use, energy use, Toxic Release Inventory (TRI) emissions, "criteria" air emissions, and green house gases by introducing methanol powered Fuel Cell Electric Vehicles (FCEVs) in the U.S. economy for a period of 20 years, 2000 to 2020?
Why is this question appropriate to ask? In order for environmental decision-makers to choose the environmentally and economically preferable option, they require the ability to comprehensively predict with knowledge of uncertainty, what may occur in the future. A static LCA methodology provides this comprehensive view, but only for one “state” in time. Albeit, a static model is useful to answer a class of static questions, such as the total emissions due to the life-cycle of a class of products at one instant in time. And further, it is useful as a point of departure for predictive analysis. However, in order to capture the important feedback phenomena within industry-to-industry activities, within industry-to-environment activities, and environment-to-environment activities, the analysis requires a dynamic model. For example, fixed capital expenditures have life-spans that may extend several years. However, there are interrelationships between the current production quantities and future expenditures on expansion capital that extend beyond the intra-interval relationships of a static model.
Concluding the environmental superiority of one product, process or activity over another is only part of the system of analysis. The rate at which the transition from an inferior activity to a preferable activity occurs is as critical to the success of implementing the more optimal choice. Dynamic modeling provides insight to the analysis of this transitional path. For example, consider a major change in vehicle technology where internal combustion engine vehicles (ICEVs) are replaced by EVs. The environmental impacts of internal combustion engine vehicles may be so dire, especially those that are near their end-of-life, that incentives to shorten their use stage may be desirable. However, to instantly remove all internal combustion vehicles, as literally suggested by a static model, would be physically impossible and would result in economic catatonia. In contrast, careful dynamic analysis would allow for planning an incentive path that is not only optimal for the environment, but also the economy.
To make the case for a dynamic model in the proposed question, consider those elements that will change over the period of analysis. Over the 20-year period there will be changes in the production phase of automobiles: to a lesser degree changes in materials extraction operations, and to a greater degree changes in the technologies implemented and corresponding production processes. There will be changes in the use phase of the automobiles: changes in the population distribution of vehicles in use, changes in how they are used, and even changes in the composition of the fuels that they burn. There will be changes in the disposition phase of the automobile: the expected useful life will change, there will be changes in the quantities reused, recycled, incinerated and land-filled, including even the abatement processes to process the disposition of automobiles. Within the 20-year period, the activities that make up the life-cycle of the automobile will be very different from the first day of the study to the last.
As a final example, consider the distribution of automobiles that will be received by dismantlers as depicted in Figure 1-2. The distribution of retired automobiles changes the make-up of materials available to the dismantling and shredding industries. As various types of plastics that make up a single automobile increases so will the amount of plastics available to the recovery industry. Thus, it would be inconsistent, as is typical of a static model, to model the recycling and recovery operations based on the same time period as the production period. This is especially the case where the average life-span of an automobile is twelve years (Bustani et al. 1998). A temporal I-O modeling approach differentiates the various time-dependent details and allows for a more accurate analysis of trends and practicable goals. For example, a temporal I-O approach is able to distinguish the past inputs of the system (plastics used in automobiles) as they affect the present state of the system (plastics that enter the recycling stream). And also, a dynamic model is able to account for any anticipatory adjustments that are to be made for future states, such as the ramping up of recycling capacity due to the current production of specific plastics in automobiles and their useful life-expectancy.
Fig. 1-2. Vehicles received by Dismantlers (Bustani et al. 1998).
What has been described to this point are the dynamic aspects of a life-cycle inventory of the LCA methodology. To gain greater insight to potential environmental impact, one must also consider the temporal distinctions of the environment. For example, dynamic considerations are essential when impacts of residuals are not due to current flows, but to stocks of accumulated flows of residuals and the assimilative capacity of the various receptors (Førsund 1985).
In consideration of the complex dynamics of the environmental system, this research will be an intermediate step to creating a dynamic LCA economic-ecological model that will lead to more robust impact analyses. In summary, the dynamic input-output modeling approach facilitates the analysis of trade-offs involving environmental decision-making. This method has the potential to illustrate the following issues:
· The scale of the activities and thus the output mix among activities,
· The input mix in an activity,
· Process techniques of production and consumption,
· The product characteristics, including durability,
· Modification (treatment) of primary residuals, and,
· Recycling/reuse of residuals.
It is important to understand the environmental repercussions of macroeconomic activities, such as the introduction and legislation of a specific automobile technology into a national economy. The strength of the I-O methodology embedded in structural economics is its systematic interpretation of the economic and ecological repercussions.
Section 1.3 – Importance of Topic
"Life Cycle Assessment is an objective process to evaluate the environmental burdens associated with a product, process, or activity by identifying energy and materials used and wastes released to the environment, and to evaluate and implement opportunities to affect environmental improvements” (SETAC 1991, 1). In this context, environmental impacts, including relevant safety, health, and social factors are quantified and summed up across the lifetime of a product, process, or activity being evaluated. LCA is most successful as a tool that takes a broad, system-wide perspective towards resource utilization and emissions issues. The goal is to reduce overall environmental impact by providing directional environmental indicators to be examined by more-focused assessments using other analytical techniques (Owens 1996, 1997). This goal is important. Minimizing the impacts of subsystems does not ensure that the impacts of an entire system are minimized or even reduced (Richards et al. 1994). In many cases, reduction or change in one part of the system typically results in only shifting the burden.
For
example, methyl tertiary butyl ether (MTBE) is a fuel additive, initially
introduced in the late ‘70s as an octane booster. Reformulated gasoline containing MTBE is known to reduce
emissions of carbon monoxide (CO), other products of incomplete combustion, and
evaporative emissions due to reducing the vapor pressure of gasoline.
Reformulated gasoline containing MTBE is especially effective in reducing
emissions in older vehicles when engines are cold and under heavy loads. However, one physical characteristic of MTBE
that is unlike most other petroleum hydrocarbon (HC) additives, is its solubility. MTBE is very soluble in water and its
transport is not limited in groundwater due to soil adsorption. In the likely occurrence of a gasoline
spill, other HCs would tend to stay put and MTBE would travel relatively
quickly, harming nearby lakes, streams and drinking water sources. MTBE is an animal carcinogen with the
potential to cause cancer in humans (Belpoggi et al. 1995, 1998) (Bird et al.
1997) (Chun et al. 1992) (Burleigh-Flayer et al. 1992).
Despite
the potential carcinogenicity and physical characteristics of MTBE, a recent
LCA study concluded reformulated gasoline containing the MTBE additive had a
significant advantage over conventional gasoline to reduce hazardous air
pollutants. Moreover, the study’s conclusion resulted in the recommendation
that “MTBE blended gasoline be considered for use in areas where population
density is relatively high and the concern of hazardous air pollutants exist”
(Raynolds et al. 1998, 130). The
findings and recommendations of this study are not false within the context of
reducing hazardous air pollutants.
However, the study presented in the paper did not consider the scenario
of increased potential harm to humans and impacts to the environment via the
pathway of groundwater. To the defense
of the researchers, it may have been a case of simple myopia. The study was focussed only on the problem
at hand to be solved – reduction of hazardous air emissions. Or, it may have been the influence of a lack
of data pertaining to the other media, i.e. soil and water, a concern intimate
to LCA practitioners (Peereboom et al. 1998). A series of general recommendations were made notwithstanding the omission of
impacts to soil and water. The recommendations were correct within the context
of reducing hazardous air pollutants, however, the recommendations may have
serious environmental repercussions via the introduction of a potential
carcinogen to drinking water supplies.
The boundaries of the Life-cycle Inventory (LCI) and the Life-cycle
Impact Assessment (LCIA) were clearly stated by the study. What remained implicit, was the boundary of
the recommendations. The shortcoming of the study was that the recommendations
did not explicitly describe the context that they were targeted.
Context
is important. Environmental Life-Cycle Assessment is a quantitative method to
facilitate Contextual Environmental
Management (CEM) decision-making.
Here, contextual environmental management is environmental
decision-making that takes into consideration cross-disciplinary interests that
include technical, economic, and societal concerns. CEM emphasizes not only what is included in the field of
influence, but as important, what has been omitted.
This research presents a method that embraces the paradigm of contextual environmental management by utilizing the constructs of Structural Economics. Structural Economics is the basis to define the proper system that best represents real-world scenarios to meet the demands of decision-making applications. Moreover, these scenarios can then be communicated more effectively to economists, policy analysts and decision-makers. Analysis within a system perspective is the essential element of LCA that sets it apart from other environmental decision-making methodologies. LCA gives the practitioner the capability to select a truly more preferable process, product or activity.
Section 1.4 – Research Approach
The approach to the research is as follows:
1. Model Development
This research will involve the presentation and implementation of a dynamic economic I-O[1] LCA model in the form of a time-varying SIM structure. This will involve the introduction and definition of environmental LCA in the context of SIM.
2. Case Study
As an illustration of the methodology presented, the case study will seek to answer the following question:
What are the environmental impacts/benefits of introducing an emerging electric vehicle technology in the U.S. economy? Specifically, what are the net increases or decreases in green house gases (GHGs) by introducing methanol powered Fuel Cell Electric Vehicles (FCEVs) in the U.S. economy for a period of 20 years, 2000 to 2020?
Today's internal combustion engine (ICE) converts only 19% of the useful energy in gasoline to propelling an automobile. Methanol fuel cell vehicles are theoretically able to achieve efficiencies of at least 38%. What are the annual net decreases in GHG emissions due to the introduction of FCEVs and when will they occur? The expected technological maturation of the methanol FCEVs is estimated to be as early as 2004 as Daimler-Chrysler, Toyota, Volkswagen and General motors have all demonstrated prototype vehicles (Nowell 1998). The experiment time period is chosen to be 20 years, from 2000 to 2020. This should allow for the effects of the introduction of new technology to be realized in the simulation as the population of vehicles transitions from older model years to newer model years. In addition to the time period of analysis, the number of time intervals chosen will be twenty intervals, and hence their duration will be one year. This is based on the availability of annual production schedules of vehicles and their average life-span.
3. Experimental Method
Presentation of a method for scenario generation and experimental design involves the following steps:
· Determination of factors and treatments of the experiment,
· Design of an appropriate factorial experiment,
· Application of a Monte Carlo simulation to generate replications,
· Calibration of the model , and,
· An analysis of contrasts to examine the results.
4. Computational Environment
The computation environment will require the ability to develop a mathematical engine that can implement the dynamic I-O model, solve the model, and generate an analysis of the scenarios in a formal experimental design structure.
Section 1.5 – Limitations and Key Assumptions
There are several general limitations and key assumptions to the research:
· The data set that is available for this research does not support an empirical study. Specifically, data regarding the end-of-life contributions of GHGs are not included in the LCA. The LCA will be limited to analyzing emissions at the manufacturing and use phases. Thus, the emphasis of the research will be on the demonstration of a dynamic LCA methodology.
· The basis of the technical coefficient matrix that is used for the research has been determined by monetary value. The main assumption is that the coefficients represent materials and energy use that then approximate environmental impact. The precise determination of environmental impact is beyond the capabilities of the available data. Simply, a 'less is best' analytical approach, i.e. any of amount of reduction in quantities released to the environment or energy expended is preferable.
· Final demand for the economic I‑O model will be limited to personal consumption expenditures (PCE). Although substantial, Government purchases, investments and net exports will be ignored. Further, added value parameters of taxes and rents will not be included in the I‑O table.
Section 1.6 – Contributions to Knowledge
The contributions to knowledge in the field of LCA include:
1. The integration of economic, societal and engineering perspectives that lead to better insight in the discovery of potential environmental problems, finding technical solutions, creating economic incentives and guidance for the resulting policies.
2. The demonstration of an environmental decision-making methodology that embraces the philosophy of Structural Economics.
3. The presentation of a dynamic approach to LCA in the open literature.
4. A contribution to the further study of economic I-O as a systems approach to modeling LCA. This allows for the incorporation of larger data sets and comprehensive normative analysis on multiple scales: regional, national and global.
5. The demonstration of utilizing comprehensive publicly available data to conduct LCAs.