A combined ahp-entropy method for deriving subjective and objective criteria weights

Abstract

This paper investigates the problem of deriving subjective and objective criteria weights, by combining the AHP and Shannon's entropy method. The paper outlines the challenge of making preferential judgments based on non-homogeneous decision data and variant decision knowledge. Such decision complexity often leads to inaccurate assessment of criteria weights, and consequently reducing the credibility of decisions. The combined AHP-entropy method conforms to the type of decision data (qualitative or quantitative; deterministic or probabilistic) and to the degree of decision knowledge (none, partial, or full preferential judgments). An easy-to-apply spreadsheet-based application program of the unified AHPentropy method is developed for deriving criteria weights, synthesizing decision elements, and ranking decision alternatives. A numerical example is used to clarify the method’s application.