A metaheuristic approach to improve consistency of the pairwise matrix in AHP

Abstract

In this paper, we are interested in modifying inconsistent pairwise comparison matrix which is a critical step in the AHP methodology, where decision makers have to improve the consistency by revising the process. To this end, we propose an improved genetic algorithm (GA) to allow decision makers to find an appropriate matrix and adjust the consistency of their judgment without loss of original comparison matrix. Numerical results with different dimensions of matrices taken randomly show the effectiveness of these strategy to improve and identify the consistency of pairwise matrix which mean that GAs are a very good tool to generate the consistent pairwise comparison matrices with different number of criteria.