Abstract

There are two mainstreams in the use of the analytic network process (ANP) and analytic hierarchy process (AHP). One is the standard applications of crisp distributive and ideal mode versions. The other is characterised by fuzzification of the AHP/ANP methodology and by attempts to tackle better inherently uncertain and imprecise decision processes with quantitative and qualitative data. This paper presents modification of the AHP/ANP method, in which fuzzy numbers have been used for determining weight values of criteria and alternatives.Unlike the papers describing the procedure of fuzzification of the AHP/ANP method, the method described here takes into account the level of uncertainty of the decision maker. After application of the AHP/ANP method in this way, the values of the functions criteria for each considered alternative are obtained. Certain values of the level of certainty are corresponding to the obtained values of the functions criteria. It is possible to generate various sets of the values of criterion functions. Since large number of experts often participate in decision making, the model deals with possibility of synthesis of the optimality of criterion values in case of group decision making. The proposed methodology has been used for the assessment of management plans in West Serbia.

Key words: Fuzzy logic, fuzzy multicriteria decision making, fuzzy ANP, fuzzy AHP.