AN EXPONENTIAL UTILITY APPROACH TO EIGENVECTOR METHOD IN THE ANALYTIC HIERARCHY PROCESS: AN IDEA INTRO

Abstract

It is crucial for a credible decision making theory to provide unique answers for the alternatives of a decision. However, different methods and algorithms devised in order to elicit true priority vectors from intuitive judgments give different priority vectors, especially when judgments are inconsistent what constantly takes place in the real life. One could deduce that such variety of results that a potential decision maker can obtain violates the uniqueness requirement mentioned above and therefore is seemed unacceptable. On the other hand, it is known that Eigenvalue Method, commonly applied in the Analytic Hierarchy Process, captures transitivity in matrices that are not consistent in the unique way. That could lead to a conclusion that maybe the Eigenvalue Method is necessary and sufficient to facilitate credible decision making based on priority weighting followed by inconsistent matrices comprising pairwise comparison judgments. However, the Eigenvalue Method, despite of obvious advantages, has also few disadvantages that cannot be neglected. A research described in this article gives rise to a new method for priority vectors deriving which coincides with the Eigenvalue Method but avoids its drawbacks. Keywords: Analytic Hierarchy Process, eigenvalue method, deriving priority vector, optimization models, condition of order preservation.