A Novel Conflicting Evidence Discounting Method Based on TOPSIS Decision-Making

Following the idea of TOPSIS method, which is ordering evaluation targets in terms of similarities among them and the ideal solution, a new method to determine the discounting factors for evidences is proposed in this paper. New method combines the main ideas of Interval Arithmetic(AI), belief functions and discounting D-S method, extending the applicability of TOPSIS method to fuse highly conflicting evidences. Two numerical cases in target recognition are given to illustrate the effectiveness of new method. The results show that the evidences can be fused effectively and correctly.