Theoretical foundations and practical issues for using expert judgements in uncertainty analysis of high-level radioactive waste disposal

Estimating the performance of a high-level radioactive waste disposal system inevitably includes uncertainties because of the large temporal and spatial scales involved. These uncertainties come from a variety of sources, some of which, given the present state of knowledge, cannot be readily quantified. Because of the complexity of the system and the issues that needed be addressed, the use of expert judgements to treat uncertainties is unavoidable in performance assessment. This paper covers the mathematical foundations and some of the practical implications in treating uncertainties for performance assessment of high-level radioactive waste disposal systems. The paper begins with a discussion of the use of expert judgments in uncertainty analysis. A mathematical approach for treating uncertainties which includes decision analysis and uncertainty in the state of knowledge (both about the models and the numerical values of parameters) is presented. We conclude the paper with an example that demonstrates the use of a methodology based on the Bayesian approach for combining probability estimates from two or more sources of information into a single estimate. The specific example addresses the estimation of the recurrence interval of a characteristic earthquake. Even though this is a relatively simple example, it is evident that the implementation of the Bayesian approach and use of expert judgements is not simple. (author) 1 fig., 9 tabs., 29 refs