Fig. 1. Life cycle analysis
framework.
Katherine L. Yuracko and Michael J. Gresalfi
LMER
Bob Lehrter
FDF
Peter Yerace
DOE-FN
Igor Komarikov
UT
ABSTRACT
During the past five years, a number of U.S. Department of Energy (DOE) funded efforts have demonstrated the technical efficacy of converting various forms of radioactive scrap metal (RSM) into useable products. From the development of accelerator shielding blocks, to the construction of low level waste containers, technology has been applied to this fabrication process in a safe and stakeholder supported manner. The potential health and safety risks to both workers and the public have been addressed. The question remains: do the benefits of fabricating products from RSM outweigh the costs? The DOE Fernald Environmental Management Project (FEMP) has developed a methodology based on life cycle analysis to evaluate the costs and benefits of recycling and reusing RSM, rather than disposing of this RSM in an approved burial site. This paper presents the life cycle analysis conducted to determine whether the DOE Fernald site should recycle some of the radioactively contaminated structural steel from the Plant 4 D&D project.
BACKGROUND
Working in consultation with stakeholders, Fernald has developed the "Decision Methodology for Fernald Scrap Metal Disposition Alternatives" which outlines a methodology to help decision makers to compare and select among competing proposals for the disposition of RSM at the FEMP. The methodology developed takes into consideration both quantitative and qualitative factors in three categories: direct costs and benefits; socio-economic and institutional impacts; and environmental, safety, and health impacts. The methodology includes both the analytical requirements to develop defensible values for a comprehensive set of performance measures, and the structure for using the performance measures to compare and rank alternative proposals. An important purpose of the methodology is to provide a mechanism for public participation in decision-making. In addition, another purpose in developing the methodology was to create a management tool which would enable DOE-FEMP decision makers to continuously track various disposition alternatives and to identify the most advantageous alternatives throughout the life of decontamination and dismantling projects.
A decision on RSM disposition alternatives should be based on two categories of information: 1) the potential impacts of choosing each of the candidate alternatives; and 2) value judgments regarding the relative worth of achievement of different objectives. Correspondingly, the methodology is divided into two phases: the life cycle analysis phase, in which the possible impacts of each of the candidate alternatives are assessed; and the decision phase. In the first phase, the objectives and program scope are defined, the RSM disposition alternatives are identified, performance measures are specified, and the impacts of the alternatives are described in terms of the performance measures. In the second phase, the decision phase, the methodology will aid the decision maker(s) in the comparison of alternatives and the selection of the preferred alternative.
For this initial application of the methodology, disposition of radioactively contaminated structural metal from the FEMP Plant 4 D&D project is being evaluated. This paper presents the life cycle analysis of seven disposition alternatives for the Category A - Accessible Metals (i.e., structural steel) from the demolition of Building 4A/Plant 4. Disposition of the remainder of materials from Plant 4 (transite, concrete, lead, process equipment, light gauge metals, etc.) was not evaluated, although the methodology can be modified to address other materials in the future. Two scenarios were evaluated, with the primary difference being the quantity of steel considered. In the first scenario, only 10% of the Plant 4 structural steel was considered (approximately 150 tons). The second scenario analyzed the disposition of 15,200 tons (100% of OU3 structural steel). In light of regulator and stakeholder input, the 15,200 tons scenario appears to have much greater value in addressing the recycling vs. disposal issue for structural steel, at Fernald and throughout the DOE weapons complex.
The information from this analysis is being used as the focal point for a dialogue involving DOE, Fluor Daniel Fernald (FDF), the regulators, and other stakeholder groups to discuss the pros and cons of the various disposition alternatives. Input received will likely result in changes to some of the results, to more accurately reflect stakeholder views and preferences. This dialogue will in turn lead into the decision phase of the analysis.
LIFE CYCLE ANALYSIS PHASE
Life cycle analysis is the process of identifying and assessing all benefits and costs that result from a course of action over the entire period of time affected by the action and providing the results in a form that will promote sound decision-making. A life cycle analysis provides a logical approach to the comprehensive assessment of alternatives which is mandated by the uncertain, hidden, and at times counterintuitive costs and benefits of alternative proposals. The elements of a life cycle analysis depend on the purpose of the analysis and the availability of specific data. In general, however, elements of a life cycle analysis consist of direct costs and benefits, which derive from the outlays that DOE would expend; socio-economic and institutional impacts; and environmental, safety, and health impacts. The life cycle analysis framework is depicted in Fig. 1. The six steps that make up the life cycle analysis are inter-linked and are described below.
Fig. 1. Life cycle analysis
framework.
Define Nature of Decision and Program Scope
A clear statement is needed of the current system and the nature of the decision that is required. This establishes the boundaries for which viable alternatives can be defined. It also defines the scope for which impact analysis is required. Finally, it helps in identification of possible decision-aiding approaches for use in the decision phase. This step also includes a preliminary assessment of the quality of the information available to perform the analysis; identification of the criteria for the quality and efficacy of the analysis; and a preliminary identification and inventory of assets and resources.
Specify Objectives and Performance Measures
To conduct an effective analysis, it is necessary that a clear statement be made of the program objectives, so that the intents and reasoning behind the program are well understood by the analysts, the decision makers, and the stakeholders who will have a say in the final decision. In order to estimate how well proposed alternatives achieve the identified objectives, measures are needed to quantify that performance. Thus, the identified objectives are translated into attributes and corresponding measurement scales (performance measures) that relate descriptions of impact levels to quantitative or qualitative scores. This is an important step, because the performance measures determine the specific analytical approaches that will be taken in subsequent steps of the methodology and constitute the input to the decision phase. Nine performance measures were defined for this study. (Note that regulatory compliance was not included as a performance measure because it is assumed that all of the alternatives will fully comply with ARARs. Therefore, regulatory compliance will not differentiate among the alternatives and does not need to be included in this comparative evaluation.) The performance measures are summarized in Table I.
Table I Performance Measures Used for the Life Cycle Analysis
Identify Alternatives
This is the step in the methodology where the specific alternatives to be considered are defined. This step forces the decision makers to think through the specific alternatives and identify the potential impacts of each proposed alternative. This step also includes a generic description of the system of activities (the general process) that are involved in carrying out a particular alternative. For example, in a metal melt option, the key steps of metal extraction, packaging, and shipment to a smelter would be outlined, as well as the key decisions and other issues that might be faced in carrying out that alternative. Seven material disposition alternatives were evaluated for this initial application of the methodology. These are presented in Table II. All of the alternatives considered will fully comply with ARARs and are implementable (i.e., are technically and administratively feasible and rely on available services and materials). Note that the methodology is designed to accommodate emerging technologies and changes to key parameters over time; thus the analysis may be updated periodically to include new alternatives and new information.
Table II Summary of Disposition Alternatives
Define Analytical Methods
In this step the analytical models and tools are defined that will be used to evaluate the alternatives on the performance measures. For simplicity, the tools are divided into three categories, however it is important to note that there are substantial interactions between the models in the different categories: 1) direct costs and benefits; 2) socio-economic issues; and 3) environmental, safety, and health impacts. The choice of tools is made in consultation with stakeholders and is a function of the desired size of the analysis and judgments about which performance measures should receive the greatest attention.
Assess the Impacts of the Alternatives
In this step of the life cycle analysis, the analytical tools developed are used to evaluate the impacts of the alternatives on the performance measures. At this stage, the opportunity exists to re-assess the initial assumptions, objectives, and scope that were developed in the initial steps. Although the entire methodology is an iterative process at every step, a feedback mechanism is indicated in Fig. 1 at the end of this step to emphasize that performance measures may be further refined, the system definition and process flow model revised, new strategic alternatives identified, and additional analyses performed.
Summarize Results of Life Cycle Analysis
The results of the life cycle analysis can be summarized in a matrix with the disposition alternatives along the top row of the matrix and the performance measures along the side. Within each cell of the matrix is the value of the performance measure for that particular disposition alternative. Table III presents the results of the life cycle analysis for 15,200 tons of structural steel.
Table III Range of Scores for Each Performance Measure
DECISION PHASE
It can be seen that not all performance measures favor one alternative. When there is no clearly superior alternative across all performance measures, decisions must be made regarding which performance measures are more important and what is the relative value to assign achievement on different performance measures. Much work has been done to develop structured approaches for analyzing tradeoffs between competing objectives. These methods can help inform decision makers on their choices, but they must be recognized solely as tools to assist decision makers, not replace them.
Given the manner in which decision maker preference information can be elicited and used, as well as the characteristics of the problem, it is not surprising that the number of specific methods that have been developed is large. However, there are a number of fundamental or more prominent methods that are particularly relevant to the asset disposition problem. Multiattribute value theory (MAVT) is one of the most widely used methods for dealing with and solving multiattribute decision problems. A number of specific techniques have evolved, but they all share the following operational steps:
The first two steps were conducted as part of the life cycle analysis phase. The next step addresses value judgements about the relative importance of the competing objectives. In this step, preferences are quantified through the assessment of weighting factors. These weighting factors can be thought of as one's "willingness to pay" to achieve benefits or avoid adverse impacts of various types. The weights reflect judgments regarding the relative value of making improvements according to one objective (e.g., reducing adverse public health impacts) relative to making improvements according to another objective (e.g., reducing cost).
In the next step, the scores of each alternative on the individual performance measures are combined into an overall rating for the alternative. A weighted linear additive scoring rule will be used. This scoring rule provides a weighted linear sum of the performance measure scores, normalized by the sum of the weights. The weighted linear additive scoring rule favors alternatives that score best on the criteria with the greatest weights.
The final, and most important step, is to perform sensitivity analyses. Sensitivity analyses analyze the implications of alternative value judgments (weighting factors) or alternative scores on the priority ranking of alternatives. Sensitivity analyses are conducted to determine whether plausible changes in weights or scores have a significant impact on the ranking of alternatives. Weights or scores may be changed individually or simultaneously to examine the following: What is the critical decision criterion? Under what conditions do specific alternatives reverse positions in ordering? What maximum, minimum, or target values result in alternatives reversing positions in ordering?
STAKEHOLDER INPUT
Stakeholder input is an important part of every step of the process outlined in this paper. Stakeholder input to date has included the following: conducted public workshop on June 11, 1996, to introduce draft Recycling Methodology, submitted draft Recycling Methodology to EPA, received EPA and public comments, prepared response package to comments, provide response to stakeholders, and conducted public workshop on November 7, 1996. The draft methodology was presented to the regulators and other stakeholders at a public meeting in June 1996. After all comments received were evaluated, it was determined that no changes to the methodology text were required. Stakeholder comments on the methodology were considered in developing the Plant 4 evaluation summarized in this paper.
At the November public workshop, stakeholder input was solicited on both the scoring of the alternatives and the weighting factors used to express preferences among competing objectives. In particular, Fluor Daniel Fernald developed a questionnaire with two exercises. The first exercise asked respondents to score the alternatives (using a scale from 1 through 5) on 4 performance measures: local economic impacts, institutional preference, local social preferences, and protectiveness of the environment. The second exercise asked respondents to assign weighting factors to performance measures. FDF conducted practice exercises at the public workshop based on a hypothetical "Plant X" that contains 1,500 tons of metal. After stakeholders completed the practice exercises and the results were discussed, questionnaires were distributed for them to complete at home and return by December 1st. The "homework" assignment requested stakeholders to evaluate the alternatives on the local economic, institutional, social, and environmental performance measures and to assign weighting factors to the performance measures for each of 3 cases: 1,500 tons of metal, 150 tons, and 15,200 tons of metal.
SUMMARY AND NEXT STEPS
The Disposition Summary Matrix table (Table IV) provides a compilation of the best data currently available and may be used as a tool to aid decision makers in arriving at an ultimate conclusion on the question of how best to disposition the structural steel from Plant 4 and throughout the FEMP. This table summarizes the results of the life cycle analysis phase of the Fernald Metals Disposition Methodology as it applies to the case of 15,200 tons of steel from OU3.
Table IV Disposition Summary Matrix
.
However, the decision methodology was designed to be a "living" process and the analysis results may be modified and revised as conditions change and new information dictates. Much of the information presented is based on best estimates rather than data generated from completed projects and activities. As the physical work of remediation projects is undertaken and "hard" data and better information become available, the Disposition Summary Matrix tables will be updated to reflect changes which could impact the comparison of alternatives. Additionally, as new technologies and approaches become available in the future, they will be evaluated and included in the Disposition Summary Matrix table as appropriate.
The information presented in the Matrix table, and the corresponding text, is being used as the focal point for a dialogue between DOE decision makers, FDF, the regulators, and stakeholder groups to discuss the issues surrounding disposition of scrap steel from OU3 remediation. The Matrix will be updated to incorporate a better understanding of stakeholder preferences, DOE policies, or EPA requirements as these develop.
ACKNOWLEDGMENTS
The "Decision Methodology for Fernald Scrap Metal Disposition" was generated, in a cooperative manner, by both Oak Ridge National Laboratory and the U.S. Department of Energy's Fernald Environmental Management Project. The authors of this report recognize that this FEMP decision methodology was generated in full partnership with the following members of the ORNL Life Cycle Analysis Team: Katherine L. Yuracko (lead), T. Randall Curlee, Rafael G. Rivera, Stanton W. Hadley, and Robert D. Perlack. Joe Jacoboski of Fluor Daniel Fernald and Igor Komarikov of the University of Tennessee were major contributors to the life cycle analysis of scrap metal disposition alternatives.
REFERENCES