Research On Linguistic Multiple Attribute Group Decision Making Methods Under Risk Preferences And Their Applications

As an indispensable activity in people’s life and production,decision-making is also the premise and foundation of people’s actions.Due to the complexity of decisionmaking problems,the intensification of individual differences and the limited knowledge of decision makers,in order to achieve accurate decision-making,multiattribute decision making emerges as the times require.However,multi-attribute decision making sometimes fails to meet people’s needs,and group decision making has the advantages of knowledge gathering and brainstorming,it is gradually changing to multi-attribute group decision making and then to linguistic multi-attribute group decision making.In fact,in the actual linguistic multi-attribute group decision making problem,different decision makers usually have different risk preferences,and risk preferences often affect the semantic value of linguistic decision information,resulting in distorted decision results.One the one hand,in order to cover the different risk preferences of decision makers,linguistic multi-attribute group decision making under group risk preferences is studied.On the other hand,in order to futher improve the accuracy of decision results and the practicability of decision methods,this paper maximizes the advantags of group decision making and the heterogeneity of decision makers,prevents group interference,assuming that decision makers make independent decisions,and no unified decision standard in the deicison making process(types of decision information,attribute sets,weight information,types of risk preferences,etc.).In order to save the time to reach decision consensus,linguistic multi-granular group decision making and heterogeneous multi-attribute decision making under different individual risk preferences are studied.Although domestic and foreign scholars have made great achievements in the theoretical and methodological research of linguistic multi-attribute group decision making,multi-granular linguistic multi-attribute group decision maing and heterogeneous multi-attribute group decision making.There are still some problems that need to be addressed:(1)The research on linguistic multi-attribute group decision making under group risk preference needs to be further explored;(2)The research outcomes of multi-granular linguistic multi-attribute decision making methods under different individual risk preferences are relatively limited;(3)There is room for research and further expansion of heterogeneous multi-attribute group decision making methods under different individual risk preferences.In order to make up for the shortcomings of the above research,from the perspectives of group and individual risk preferences,this paper focuses on several types of linguistic multi-attribute group decision making problems.The main research work and innovation achievements are as follows:(1)A linguistic multi-attribute group decision making method is proposed in which the weight information of attributes and sub-attributes is completely unknown under group risk preference.In this paper,a new two-stage optimization model is constructed by combining the group risk preference of decision makers with multi-level evaluation index system(EIS).By introducing a group generalized linguistic term set with two group risk preference parameters,semantic values of linguistic decision information under group risk preference can be quantified.In linguistic multi-attribute group decision making based on multi-level EIS,the two-stage optimization model can not only capture the influence of group risk preference on semantic values of linguistic terms in linguistic decision matrices,but also quickly solve weights of attributes and sub-attributes in the case that decision makers cannot provide their judgement matrices for comparising attributes in pairs.In addition,a linguistic multi-attribute group decision making method is proposed in which the weight information of attributes and sub-attributes is completely unknown under the group risk preference,aiming to solve the decision making proble that the decision-making group with risk preference use multi-level EIS to evaluate and then select the best alternative.It is more practical by outputting partial sorting and comprehensive sorting results of alternatives.Finally,the proposed decision-making method is applied to the decision making of teacher selection in enterprise online teaching and training.The effectiveness and feasibility of the proposed method are verified by calculation and comparative analysis.(2)A multi-granular linguistic multi-attribute group decision making method with different attribute sets under different individual risk preferences is proposed.In order to demonstrate the advantages of group decision making and the heterogeneity of decision makers,decision makers often make independent decisions.To adapt to the decision makers’ independent decision making environment,individual risk preference is introduced,and the multi-granular triangular fuzzy generalized linguistic term set under individual risk preference is redefined.Then,a multi-granular linguistic conversion model under individual risk preference is constructed to solve the conversion of multi-granular linguistic information under individual risk preference.Secondly,for different attribute sets selected by decision makers,based on the principle of maximizing dispersion,an optimization model is established to determine the attribute weights of individual attribute sets selected by different decision makers.In addition,a multi-granular linguistic multi-attribute group decision making method with different attribute sets under different individual risk preferences is proposed,and the decision making process of multi-granular linguistic multi-attribute group decision making based on different individual risk preferences and attribute sets is designed.Finally,the proposed decision method is applied to the decision making of enterprise green suppliers.The effectiveness and feasibility of the proposed method are verified by calculation and method comparision.(3)A heterogeneous multi-attribute group decision making method with unknown weight information under different individual risk preferences is proposed.Considering that the behavior of deiciosn makers may use different types of decision information,this paper divides heterogeneous information into numerical and linguistic mixed type and linguistic mixed type,and studies heterogeneous multiattribute group decision-making problems and heterogeneous linguistic multi-attribute group decision-making problems.For the first problem,the trapezoidal fuzzy set theory is used to construct a multi-granular interval linguistic representation model under individual risk preference,which is used to represent the semantic values of multigranular interval linguistic terms.In order to aggregate heterogeneous decision information,the heterogeneous modular Technique for Order Preference by Similarity to an Ideal Solution(TOPSIS)method is extended to the decision-making situation considering different individual risk preferences.Aiming at the situation of incomplete attribute weight information of decision makers and individual attribute sets,based on the principle of maximization of dispersion and the principle of group compatibility,two optimization models for weight solution are established.In addition,a heterogeneous multi-attribute group decision-making method with incomplete weight information of individual attributes and decision makers is proposed,aiming to solve the heterogeneous multi-attribute group decision-making problem under different individual risk preferences.For the second problem,in order to aggregate the multigranular interval linguistic terms given by decision makers with different individual risk preferences,a multi-granular interval linguistic conversion model is constructed.Since the attribute weight information in the individual attribute set is completely unknown,the Best-Worst Method(BWM)is extended to the decision-making situation of multiple attribute sets,and an optimization model is constructed to solve the attribute weights of different attribute sets.Similarly,a heterogeneous linguistic multi-attribute group decicision-making method with completely unknown individual attribute weights is proposed,aiming to solve the heterogeneous linguistic multi-attribute group decision-making problem under different individual risk preferences.Finally,the proposed two decision methods are applied to the enterprise patent conversion decision based on quality evaluation.The effectiveness and feasibility of the proposed method are verified by calculation and method comparision.The research of this paper aim to solve linguistic multi-attribute decision-making problems in different decision-making situations,and enrich the theory and method of linguistic multi-attribute group decision-making under risk preferences.In addition,the relevant decision-making problems of enterprises are selected to verify the validity and feasibility of the decision-making methods,and also provide theoretical and methodological support for enterprises to solve practical problems.