A perspective on structured expert judgment

Abstract

Expert judgment has always been used in risk analysis, both in the qualitative and in the quantitative phases of such analyses. Focusing on the use of expert judgment for obtaining quantitative statements about uncertain quantities, it appears that expert judgment is a data source with a number of special characteristics. Consequently, the appropriate use of this source requires a structured approach and the development and introduction of specialized methodological tools. These formalized methods should deal with such aspects as the choice of experts, the encoding of uncertainty, the elicitation of judgments, and the evaluation and combination of judgments. Purpose of this paper is to review the state of the art of the use of expert judgment, in particular in risk analysis. The paper discusses the key characteristics which underline the need for a structured approach to expert judgment, and the procedures which may be used in applying structured expert judgment. Also, an impression is given of a number of practical experiences of the use of expert judgment. The paper concludes with a discussion of “lessons learned” and of unresolved issues which require further attention.