The Improved Study On TOPSIS Multi-attribute Decision Making Method

The deepening of the information age has promoted the rapid development of the economic integration process,making the competition between enterprises increasingly intense.In order to cope with the intense competitive pressure,more and more enterprises to focus on the development of core business,to create a unique core competitiveness of enterprises,non-core business or front-end products to take the form of outsourcing or procurement to achieve,as far as possible the enterprise advantage resources used to build the core competitiveness.Therefore,in such circumstances,how to choose the supplier scientifically and reasonably has become the top priority of various enterprises.The choice of suppliers requires the evaluation and sequencing of multiple vendor programs under multiple indexes,which is a typical multiple attribute decision making problem.In the multiple attribute decision making of such suppliers,it is the key to solve the problem by constructing a perfect supplier evaluation index system and choosing effective multiple attribute decision making method.In terms of supplier evaluation index system,this paper summarizes the commonly used indicators on the basis of previous studies to construct the initial index of the supplier selection evaluation index system,and then invited industry experts to use the Delphi method to score the selected indexes,and according to the expert's opinion on the index system to filter,so as to determine the final supplier selection evaluation index system.In the multiple attribute decision making methods selection,based on the deep analysis of traditional TOPSIS method defects,targeted to improve the defects,then an improved TOPSIS method based on Mahalanobis distance and contact vector distance is established.The improved method uses Markov distance to replace the European distance to improve the traditional TOPSIS method,to solve the shortcomings of the European distance failure when there is a linear correlation between attributes;considering the set of opposites and introducing the contact vector distance,possible solutions to solve the distance close positive ideal solution and negative ideal solution is also near the defect.And then through the decision maker preference coefficient will be Mahalanobis distance and contact vector derived from the results of the new relative degree of close,thus overcoming the traditional shortcomings of the TOPSIS method.Finally,the validity of the method is verified by the empirical evidence chosen by the supplier.