Linguistic Distribution Assessment Social Network Large-Scale Group Decision Making Method And Their Application In Intercity Railway PPP Projects

With the complexity of decision making problems and the development of society,social network large-scale group decision making(SN-LSGDM)has become a hot research topic.At the same time,it is difficult to provide accurate assessment information by considering the vagueness of peoples judgments and the complexity of decision-making problems.Linguistic distribution assessment(LDA)can reflect peoples hesitation and uncertainty in the decision-making process.Therefore,this paper discusses the large group decision-making method of LDA under social networks and applies it to the selection of intercity railway PPP projects.The main work includes the following aspects:(1)Firstly,according to the complete trust relationship,a density peak clustering algorithm based on trust constraint is proposed.Secondly,in order to obtain stable trust-relationships within subgroups,the concept of trust consensus is proposed and the minimum trust consensus adjustment model is established.On this basis,the weights of subgroups and experts within subgroups are obtained,and the collective decision matrix is also obtained.Finally,regarding subgroups as the decision-making units,the single-layer minimum adjustment consensus model of the subgroup decision,the alternative,and the element levels are established for different consensus scenarios.(2)Firstly,for the incomplete trust relationship,a trust-based distance partitioning algorithm is proposed,and expert and subgroup weights are calculated according to experts in-degree and trust scores.Secondly,a new LDA cross-entropy is proposed and attribute weights are obtained.Finally,regarding subgroups and experts as decision units,an improved model of hierarchical minimum adjustment consensus is established to adjust both subgroups and experts.(3)To solve the LSGDM problem under social networks with noncooperative behaviors.First,the trust relationship is updated according to the trust relationship and opinion deviation.On this basis,we estimate the incomplete LDA information.Secondly,the non-cooperation degree and punishment coefficient are defined,and the non-cooperative behaviors are divided into complete non-cooperative,partial non-cooperative and complete cooperative.Finally,corresponding weight punishment and opinion punishment strategies are taken for different non-cooperative behaviors.Based on the above research,chapters 3 and 4 respectively discuss the LSGDM methods under social networks based on single-layer and hierarchical consensus adjustment models,and make a comparative analysis of the two methods.In addition,it is applied to the selection of PPP models for intercity railway projects.Chapter 5 studies the social network LSGDM method based on non-cooperative behaviors,and applies it to the social capital selection of intercity railway PPP projects.There are 6 figures,37 tables and 109 references in this paper.