Generating consistent subjective estimates of the magnitudes of causal relationships in fuzzy cognitive maps

Fuzzy cognitive maps (FCM) are useful tools for exploring the impacts of different input states on fuzzy dynamical systems. The development of a FCM requires the specification of the signs and magnitudes of the relevant causal relationships by one or more experts based on subjective estimates of the causal relationships. However, while it is often relatively easy to determine the sign of the causal relationship, determination of the magnitude is often problematic, and so in many cases only the signs are specified. However, if only the signs are specified, then only limited use can be made of the FCM. There is thus the need to provide an effective means for the generation of consistent estimates of the magnitude of each causal relationship. In this paper we present an integrated process for generating consistent subjective estimates of the magnitudes.