A Multi-attribute Utility Decision Analysis for Treatment Alternatives for the Doe-sr Aluminum-based Spent Nuclear Fuel

The current mission of the Department of Energy Savannah River Operations (DOE-SR) with respect to aluminum-based spent nuclear fuel (Al-SNF) is “to identify and implement appropriate actions for the safe and efficient management of spent nuclear fuel ...including placing these materials in forms suitable for ultimate disposition.” as stated in the Savannah River Site Spent Nuclear Fuel Management: Preliminary Draft Environmental Impact Statement. The proximate goal for DOE-SR is to identify the treatment alternatives for Al-SNF that achieve this mission with optimal efficiency and effectiveness. A multi-attribute utility decision analysis (MUA) was performed to address this decision. The results of the MUA indicate that, for DOE-SR, the utility for the melt and dilute treatment alternative is greater than the utility for direct disposal. This result is due largely to the DOE-SR perception of the acceptability of the treatment alternatives to peer reviewers, and to the weight that DOE-SR puts on this acceptability. Sensitivity analyses were performed on the model to examine the impact of different decision-maker values on the outcome. The relatively higher utility for melt and dilute was present throughout the sensitivity analyses, except when extreme changes were forced into the model. In particular, the higher utility for melt and dilute could be eliminated by removing “acceptability” from the analysis, and could be reversed by reversing the acceptability values. INTRODUCTION The current mission of the Department of Energy Savannah River Operations (DOE-SR) with respect to aluminum-based spent nuclear fuel (Al-SNF) is “to identify and implement appropriate actions for the safe and efficient management of spent nuclear fuel ...including placing these materials in forms suitable for ultimate disposition.” The proximate goal for DOE-SR is to identify the treatment alternatives for Al-SNF that achieve this mission with optimal efficiency and effectiveness. DOE-SR initially considered ten treatment alternatives for the management of Al-SNF, and has narrowed the choice to two of these: the direct disposal and melt and dilute alternatives. The decision analysis presented in this document focuses on a decision between these two remaining alternatives. WM'99 CONFERENCE, FEBRUARY 28 MARCH 4, 1999 MULTI-ATTRIBUTE UTILITY ANALYSIS The decision analysis is a multi-attribute utility analysis (MUA). In this analysis the objectives of the decision-makers, within the context of the decision to be made, are identified, the attributes that affect the decision are defined, values for those attributes are estimated and then normalized to a scale common to all attributes. A single metric, in the form of a multi-attribute utility, is provided as input to the decision-maker. The analysis addresses uncertainties in the data by modeling uncertain parameter values, using the modeling tool Analytica , . Uncertainty is expressed in the model results as a cumulative distribution. Distributions are combined by randomly sampling on each distribution of parameter uncertainty and combining the samples to generate aggregate uncertainty distributions; these resulting distributions are the multi-attribute utilities. The method can be readily modified to account for changes in the state of knowledge regarding data, treatment technologies, decision-maker values, or regulatory environment. The multi-attribute decision analysis process employed includes the following steps: 1. Identification of the decision objectives and hierarchy. 2. Identification of the attributes. 3. Modeling the physical processes of the alternatives. 4. Identification and quantification of appropriate uncertainties. 5. Elicitation and construction of single-attribute utility functions. 6. Calculation of single attribute utilities. 7. Combining single attribute utilities using an appropriate MUA model; verification of assumptions. 8. Exercise of the model, computing expected utilities for each alternative. 9. Performing sensitivity and importance analyses. The ultimate decision-maker is the Department of Energy (DOE). However, the DOE management at the Savannah River Site (DOE-SR) will recommend a decision to DOE Headquarters. Thus, DOE-SR provides the substantive expertise and is acting as the de facto decision-maker for this analysis. Therefore, critical input was elicited from DOESR staff DECISION ATTRIBUTES Multi-attribute utilities are calculated for both treatment alternatives within the context of six decision attributes. These attributes are defined by identifying the objectives for the decision “Minimize Cost” is an objectives hierarchy showing the five major objectives that DOE-SR believes must be met. Each objective must be measured on a suitable scale in the MUA, and that scale is referred to as the attribute of the objective. The decision attributes considered in this decision are: ∗ The objective “Minimize Cost” defines two separate cost attributes to reflect the decision-makers’ different utilities towards capital costs and operational costs) WM'99 CONFERENCE, FEBRUARY 28 MARCH 4, 1999 • Capital cost. • Other costs: maintenance and operational (M&O) cost, including transportation cost. • Public radiological health. • Acceptability to two major peer review panels. • Secondary waste. • Likelihood of proliferation. Other issues such as worker safety and schedule were considered as concepts for potential decision objectives. However, it was determined by DOE-SR that these issues, while very important to the successful completion of DOE-SR’s mission, do not represent issues that would discriminate between the two alternatives. THE DECISION ANALYSIS Figure 1. Process Flow Schematics for Direct Disposal and Melt and Dilute Treatment Alternatives The Attributes Discussion with DOE-SR suggested that only capital costs would figure separately in their recommended decision; all capital cost could be considered together as a single attribute. Capital cost is singled out because funding for capital improvements requires congressional consideration and action separate from the operational budget. As a result of this separate consideration, the weighting of capital cost could be different from the Pool Storage