Research On Probabilistic Linguistic Multiple Attribute Group Decision Making Based On Behavioral Decision Theory

Faced with the problem of multi-attribute group decision making in a complex situation,result of the decision makers often has a certain deviation due to the instability of their decision making attitade.The traditional multiple attribute group decision making method assumes that the decision makers are completely rational,but in the face of complex decision problem,decision makers have limited rational psychological characteristics,at the same time,there may be more than one linguistic term to describe the decision information when the decision maker expresses the decision information,and the decision maker may prefer some linguistic terms.Therefore,this paper chooses probability linguistic term sets to describe the information of attribute preference in the complex fuzzy situation.In the decision making process,the behavioral decision theory is introduced to consider the different psychological behaviors of decision-makers,and the group consensus and alternative optimization of probabilistic linguistic multi-attribute group decision making are studied,it makes the decision making result more practical.The main research work is as follows:Firstly,under the probabilistic linguistic environment,the consensus problem of multi-attribute decision making for small groups is studied.The probability linguistic term sets is introduced into the group decision making environment,the group consensus level of the probability linguistic decision matrix is defined,when the probability linguistic individual decision matrix doesn't meet the consensus,two group consensus feedback adjustment decision strategies are given.Under the cooperative behavior,the group consensus automatic feedback adjustment strategy is established.In the case of incomplete collaboration,an interactive group consensus feedback decision model is established considering the degree to which decision makers change their attitudes.The feasibility and applicability of two strategies are verified by the examples.Secondly,under the probabilistic linguistic environment,the consensus problem of multi-attribute decision making for complex large groups is studied.Using probability linguistic arithmetic rules and probability linguistic Archimedes MM operator to construct difference matrix between decision makers,this paper presents an probability linguistic orthogonal clustering algorithm for large groups.After clustering large groups,the improved prospect theory is used to convert the probabilistic linguistic subgroup decision information,and defines a subgroups consensus measure and large group consensus measure.Considering minority opinions and uncooperative behaviors,a consensus approach and its correction strategy are proposed for probability linguistic multi-attribute decision making in complex large group based on clustering algorithm and prospect theory.Thirdly,considering the decision-maker's loss-aversion behavior,this paper studies the optimal alternative of probabilistic linguistic multi-attribute group decision making under different attribute conditions.On the one hand,the improved TODIM method is used to give the multi-attribute group decision making method under the probabilistic linguistic environment.In the case that the attribute weights are unknown and independent of each other,a model considering both the maximization of negative ideal dissociation difference and the maximization of entropy is established to determine the attribute weights.The fuzzy measure and Choquel integral are used to calculate the weights of related attributes.On the other hand,considering the psychological behavior of decision maker's regret avoidance,a risk-type multi-attribute group decision making method with unknown attribute weight is proposed.When the decision-makers give the multidimensional preference information between alternatives,the LINMAP method under the probabilistic linguistic environment determines the attribute weights and the positive ideal solution of the alternatives.The calculation formula of regret-rejoice utility value under the probabilistic linguistic attribute value is given.Then,we study the three-way decisions making under probabilistic linguistic environment.The loss function of decision rough set is a probabilistic linguistic function,and a probabilistic linguistic decision rough set model is proposed.Bayes theory and the improved TODIM method are used to deduce the three-way decisions rules,and a multi-attribute group decision model based on the decision rough set is established,which expands the application field of the three-way decisions.Finally,taking the green degree of three stainless steel products of Tai Yuan Iron and Steel Company(TISCO)as an example,the paper validates the effectiveness and practicability of the probabilistic linguistic multi-attribute group decision making method based on behavior decision theory.