Improving Analytic Hierarchy Process Expert Allocation Using Optimal Computing Budget Allocation

The analytic hierarchy process (AHP) has been widely applied to multicriteria decision making problems. The AHP aids decision makers to determine the priorities of multiple criteria, and make reasonable decisions. In the AHP, evaluating candidate alternatives requires multiple experts' evaluations to avoid personal subjectivity. Although having more experts can improve selection quality, inviting more experts also results in higher recruitment cost and longer evaluation time. In this paper, we introduce the idea of optimal computing budget allocation (OCBA) and propose a method, named the AHP_OCBA method, to improve the efficiency of expert allocation. The proposed method optimizes the allocation of experts to maximize the probability of correctly selecting the best alternative in the AHP. This method also can minimize the required number of experts to meet the probability of correct selection. An illustrative example is provided to indicate the implementation of the AHP_OCBA method. We show the improvement of the proposed method compared to proportional and equal allocation rules in the numerical result.