Multi-Attribute Decision Making using Competitive Neural Networks

Multi Attribute Decision Making (MADM) methods are widely used for making the optimal decision. Different approaches have been presented to solve decision-making problems. The aim of MADM is ranking of feasible alternatives. In this paper, a new approach to solve MADM problems using an artificial neural network has been presented. The competition among alternatives is modeled by a competitive network. The ordered list of the alternatives is achieved in two phases: partial ranking and fine ranking. The results of this approach are compared with Simple Additive Weighting (SAW) and Technique for Order Preference by Similarity to the Ideal Solution (TOPSIS).