Wiley Handbook of Science and Technology for Homeland Security, 4 Volume Set

EXPERIENCE WITH EXPERT JUDGMENT: THE TU DELFT EXPERT JUDGMENT DATA

ROGER M. COOKE

Department of Mathematics, Delft University of Technology, Delft, The Netherlands

LOUIS L.H.J. GOOSSENS

Department of Safety Science, Delft University of Technology, Delft, The Netherlands

1 INTRODUCTION

The pros and cons of different weighting schemes remain a subject of research [1, 4]. The European Union (EU) contracted the TU Delft to review its applications both within EU projects, and elsewhere, in which experts assessed variables in their field of expertise for which the true values are known, in addition to variables of interest [3–6]. These are called seed, or calibration, variables. Since then, the TU Delft expert judgment database has nearly doubled. We now have studies involving over 67,000 experts' subjective probability distributions. The main sectors and summary information are given in Table 1.

The authors believe that this database represents a unique source from which much can be learned regarding the application of structured expert judgment in quantitative decision support. The entire data, appropriately anonymized, may be obtained from the first author. It is hoped that others will use this data to further develop methods for using structured expert judgment.

TABLE 1 Summary of Applications per Sector

We assume that uncertainty is represented as subjective probability and concerns ...