Improved TOPSIS Method Based On Evidence Theory Under Incomplete Information

Decision making is an unavoidable thing in our daily life.It is a matter of national politics,economy,military affairs and diplomacy,which is small to every aspect of our daily life.With the rapid development of society,the decision-making problems become more and more huge and complex,in the face of such a complex decision-making environment,how to make a reasonable decision,attracted the attention of many scholars,multi-attribute decision making method is presented in this context is used to solve the reality of many complex decision problems in life.The traditional multi attribute decision making method usually assumes that all decision makers understand all the decision attributes,and make decision in the same decision index system.With the increasing number of decision making problems to be considered and the wider knowledge involved,it is very difficult for a single decision maker to understand all aspects of a decision problem,given a set of index,for some of the attributes they do not understand,the decision maker cannot or do not want to give decision information.Such problems are generally referred to as incomplete information multi-attribute group decision making.This is common in the current decision-making environment.So it is of great significance and application value to study this kind of problem.For multiple attribute decision making with incomplete information,the existing methods are usually solved by the method of complement or some methods based on evidence theory.By using heuristic algorithm and machine learning method,the incomplete information is added to complete information.Although this method can deal with this kind of problem,but its subjectivity is too strong,and the decision information given by the decision maker has been destroyed,often cannot truly reflect the decision makers.The existing research based on evidence theory is mainly concerned with incomplete weight information or incomplete attributes,the problem of incomplete information is seldom considered at the same time.This paper draws on previous research results;we both consider the incomplete of weight information and attribute information.Firstly,the weights of attributes and the weights of decision makers are obtained by the combination of evidence theory and group consistency,then,an improved TOPSIS method based on evidence theory is used to get the final evaluation result.Finally,the proposed method is applied to the distribution center location problem to verify the feasibility and rationality of the proposed method.