M Leonelli and JQ Smith
Bayesian Decision Support for Complext Systems with Many Distributed Experts
Abstract: Complex decision support systems often consist of component modules which, encoding the judgments of panels of domain experts, describe a particular sub-domain of the overall system. Ideally these disparate modules need to be pasted together to provide a comprehensive picture of the whole process. The challenge of building such an integrated system is that, whilst the qualitative features are common knowledge to all, the explicit forecasts and their associated uncertainties are expressed only locally. The structure of the integrated system therefore needs to facilitate the coherent piecing together of these separate evaluations. If such a system is not available there is a serious danger that this might drive to incoherent and so indefensible decision making. In this paper we develop a graphically based framework which embeds a set of conditions that, if satisfied in a given context, are sufficient to ensure the composite system is truly coherent. Furthermore, we develop new message passing algorithms that enable the uncertainties within each module to be fully accounted for in the evaluation of expected utility scores of this composite system.
Keywords: Bayesian decision theory; combination of expert judgment; decision support systems; graphical models; uncertainty handling.