A decision making model for multi-attribute problems incorporating uncertainty and bias measures

Although a particular attribute, in isolation, may be considered as very important, it will not be of much use in differentiating among alternate decisions if all values for that particular attribute are very close. Another problem is in specifying weights indicating the attributes the decision maker considers important. These weights are difficult to obtain because they often reflect an attitude brought about by considering all of the alternatives rather than reflecting the decision maker’s behavior. This article addresses the preceding problems through successive iterations of a process by which the decision maker is presented with sets of alternatives that most closely satisfy his ideal choice. The process incorporates the uncertainty of the decision maker with respect to how much each alternative’s attribute value satisfies his ideal value. Entropy is also used as a measure of uncertainty in obtaining the alternative that is as close as possible to the ideal. The bias of the decision maker is reflected in how much he feels each attribute will make him happy. Fuzzy set theory is used to cope with the vagueness of the decision maker’s feelings with respect to his satisfaction with an alternative and the relative importance of each attribute.