Towards a Self-Regulating Process of Pairwise Comparison in AHP

The process of pairwise comparison (PC) in AHP requires the decision maker to fill a total number of kn elements in k matrices (n and k are the number of alternatives and criteria, respectively). The decision making process becomes a more time consuming process with a higher possibility of making human errors by the increase in the value of kn. In this paper, a self regulating approach to AHP´s PC, inspired by the similar process in PROMETHEE II, is proposed. In this approach the decision maker does not have to identify all of the kn elements directly; therefore, the production of final results will be faster and even more accurate. The proposed approach is implemented and evaluated with real world data. The results show an outstanding consistency (Cons0.70) between the final output of this modified version of AHP and that of the other three multi-criteria decision making methods, which are used in our experiments.