Reliability, validity, and imprecision in fuzzy multicriteria decision-making

Abstract

There has been a great deal of interest in the application of fuzzy sets for representing uncertainty in management decision models. One of the important steps in the decision process is the elicitation of fuzzy judgments from the decision-makers. There are two elicitation procedures used in the existing literature: (1) direct numerical assessment and (2) linguistic variables. We introduce a third procedure, which we term the "fuzzy spatial instrument." We examine some characteristics of judgment elicitation procedures as measurement instruments. In particular, we conduct an empirical test of fuzzy measurement instruments using a set of subjects and evaluate the instruments on the criteria of reliability, validity, and imprecision. We conclude that the use of fuzzy instruments will introduce some level of imprecision into the decision-making process due to the nature of the instruments themselves.