An entropy-based approach to faculty assessment with interval numbers

Abstract

In ambiguous decision domain, interval numbers are employed to evaluate the overall value of alternatives (e.g., the faculty candidates) due to the inherent problems related to the uncertainty of the multiple attribute decision making (MADM). However, it is uneasy to compare interval numbers of different alternatives with multiple attributes. In this study, a new approach is proposed, in which the interval decision matrix is transformed into definite one to make the entropy of the attributes maximal. Then, the weights of attributes are determined by setting up a programming model which maximizes the distances among the overall values of the alternatives. An example of assessing the faculty candidates (alternatives) is used to illustrate the proposed approach.