A New Belief Function Based Approach for Multi-Criteria Decision-Making Support

In this paper, we propose a new approach based on belief functions for multi-criteria decision-making (MCDM) support which is inspired by the technique for order preference by similarity to ideal solution (TOPSIS). This new approach, called BF-TOPSIS (Belief Function based TOPSIS), includes four distinct methods with different computational complexities. BF-TOPSIS offers the advantage of avoiding the problem of the choice of data normalization, of dealing with some missing scores, and of taking into account the reliability of each source (or criterion) that provides the scores of alternatives. We present results of BF-TOPSIS for different MCDM examples and discuss it robustness to rank reversal phenomena.