A Consensus Model Based Dynamic Probabilistic Linguistic Multi-attribute Group Decision Making Method And Application Research

With the development of society and the popularization of the Internet,people’s daily decision-making scenarios are complex and changeable,and the amount of relevant information generated is also increasing.It is often difficult for decision makers to accurately express their personal opinions based on numerical evaluation information alone,so they need to describe their personal opinions with the help of uncertain linguistic evaluation information.Probabilistic linguistic term set(PLTSs)is a new linguistic form proposed by researchers in recent years to describe uncertain linguistic evaluation information.Currently,the research on PLTSs group decision making method has gradually become the focus of many scholars.Therefore,this paper focuses on the consensus model of PLTSs multi-criteria group decision making method and group consistency problem,and proposes a PLTSs group decision making model based on consensus model and a dynamic PLTSs multi-criteria group decision making model based on EDAS(Evaluation based on Distance from Average Solution)and consensus model in social network environment,and applies them to the selection problem of prevention and control scheme in universities.The main research works are as follows.(1)Considering the influence of the social network relationship existing among decision makers on the decision outcome,this paper proposes a new,probabilistic linguistic multi-attribute large group decision making method based on group consistency based on social network theory.To address the situation that decision makers belong to multiple community clusters in real decision making,the concept of community sub-module affiliation is proposed and preprocessed,the decision makers are clustered and analyzed by improving the Louvain community detection algorithm,and the maximum consensus sequence mining algorithm is extended to the PLTSs multi-criteria group decision making domain,and a consensus measurement method based on group consistency and consensus reaching model.(2)Considering the characteristics of imperfect rationality of group decision makers in real decision making,this paper proposes a new,multi-attribute regret theory decision making method based on EDAS and consensus model in the evaluation information environment of probabilistic linguistic terms.For the probabilistic linguistic multi-attribute group decision problem with unknown attribute weights,the regret theory is combined with the group decision method,and the EDAS method is combined with probabilistic linguistic information processing for solution ranking.Further,consensus measures are applied to the initial solutions to obtain the final solutions that meet the wishes of most decision makers.(3)In order to get the decision result with the highest possible group consensus degree,this paper defines the best solution preference degree and the best solution consensus degree,constructs a consensus measurement model based on probabilistic linguistic information,and generates the maximum consensus comparison sequence that reaches the group consensus threshold by dynamic feedback adjustment on the basis of calculating the consensus degree of the comparison sequence,so as to get the final decision solution,the method does not need to assemble the final priority vector and compare the ranking,and the method is relatively simpler and easier to understand.Finally,the model proposed in this paper is applied to the selection of epidemic prevention and control schemes in universities,and is compared and analyzed with other methods to illustrate the reasonableness and feasibility of this method.