Risk Modeling for Decision Support

Abstract

Decision-makers require tools to aid in risky situations. Fundamental to this is a need to model uncertainty associated with a course of action, an alternative’s uncertainty profile. In addition we need to model the responsible agents decision function, their attitude with respect to different uncertain risky situations. In the real world both these kinds of information are ill defined and imprecise. Here we look at some techniques arising from the modern technologies of computational intelligence and soft computing. The use of fuzzy rule based formulations to model decision functions is investigated. We discuss the role of perception based granular probability distributions as a means of modeling the uncertainty profiles of the alternatives. We suggest a more intuitive and human friendly way of describing uncertainty profiles is in terms of a perception based granular cumulative probability distribution function. We show how these perception based granular cumulative probability distributions can be expressed in terms of a fuzzy rule based model.