Research On Consensus Sorting Method For Multi-criteria Group Decision Making Considering Heterogeneous Information

Consensus-based multicriteria group decision making methods have been a hot topic of research in recent years and are developing rapidly.However,most of the multi-criteria group decision making methods target the ranking of alternatives,and relatively little research has been done on the problem of alternatives sorting,so it is necessary to study such problems.Due to the difference between individuals,decision makers may express different types of preference information,such as preferences for comparing the advantages and disadvantages of solutions,the degree of importance of criteria,and so on.Therefore,it is an important research topic to describe the multi-criteria group sorting consensus process considering heterogeneous preference information.To address the above issues,this thesis proposes multi-criteria group decision consensus sorting methods considering heterogeneous information,which mainly includes the following two aspects.First,the study of multi-criteria consensus sorting model based on utility additive methods under heterogeneous information.This part of the study extends the multi-criteria sorting problem to the context of group decision making considering heterogeneous information,and through a series of methods such as preference adjustment and robustness analysis,the group consensus reaching process is obtained,and then the results of group sorting are sought.First,using different types of individual preference information and utility additive methods,we construct an individual preference model considering heterogeneous information,verify whether the decision maker preferences are compatible,and if not,perform preference adjustment;then,perform robustness analysis to find a representative set of solutions from the solution space of compatible decision maker preferences,construct the utility functions of decision makers and obtain the sorting results of individual decision makers;finally,we integrate the sorting results of all decision makers and measure the group consensus level.If the consensus level is low or not reached to a satisfactory consensus level,we update the sorting results of individual decision makers according to the proposed updated strategy,and then improve the group consensus level.Iterating the process until the consensus reaches a satisfactory consensus level to get the final group sorting results.Second,a multi-criteria consensus sorting model considering criterion interaction under heterogeneous information is studied.Based on the first part,this part of the study further considers the interaction between criteria while considering heterogeneous information,and constructs a multi-criteria consensus sorting model considering the interaction of criteria under heterogeneous information.First,the individual preference model of decision makers is constructed by using Choquet integral idea,and whether the decision makers’ preferences are compatible is also verified,and if not,preference adjustment is performed;then,the category acceptability of individual decision makers is obtained by using SMAA Monte Carlo simulation;finally,according to the results of SMAA,the sorting of all decision makers is integrated,the degree of group consensus is calculated,and the final group sorting result is obtained according to the designed group consensus mechanism.For the above research,the paper gives a corresponding example.This example collects and collates the QS ranking data of 58 universities in China,and adopts the proposed method in two parts to sort the universities in China,and verifies the validity and rationality of the proposed method.It is confirmed that the proposed method can effectively and reasonably solve the consensus-based group sorting problem in the face of heterogeneous preference information of decision makers,and can provide effective decision support to group decision makers in the realistic group decision-making environment.