Research On Multi-attribute Group Decision-making Method Based On Preference Approval Structure And Its Application

In the multi-attribute group decision making(MAGDM)problems,preference ordering is often used to express the ranking of multiple individuals for a given alternative.In the realistic decision-making situation,individuals can easily provide preference information about the acceptance of decision-making objects.The preference approval structure(PAS)as an extension of the ordinal preference model,by merging the two categories,ie.acceptable(good)and unacceptable(bad),makes the ranking and approval preference information to be combined organically,which can more easily simulate the preference expression of decision makers(DMs).Considering the complexity and uncertainty of realworld decision-making problems,the probabilistic language term set(PLTS)can better characterize uncertain information and compensate for the shortcomings of PAS in portraying the degree of decision makers’ hesitation in evaluating values.Therefore,this thesis uses PAS to characterize individual preferences information and extends it to the probabilistic language environment,while introducing behavioral theory to make the decision-making method more realistic.The research content and results of this thesis include:(1)A consensus method of preference approval structure multi-attribute group decision making based on regret theory.For a decision problem where the attribute value is a preference approval structure,this thesis proposes an individual reference point acquisition method based on regret theory.According to the regret-joy value,the individual preference approval structure is obtained,and the axiomatic distance preference approval structure aggregation model is used to calculate group preference.Based on this,this thesis proposes a consensus measurement method and designs the corresponding adjustment mechanism,and the feasibility of the proposed method is verified by a specific example.(2)A preference approval probabilistic language group decision-making method based on prospect theory and VIKOR.This chapter introduces PAS into the probabilistic language environment and proposes a new PLTS score function and a method of converting probabilistic language to preference approval.Based on TOPSIS ideas,an optimization model for attribute weights is constructed.The prospect theory and the VIKOR method are combined to determine the ranking of the alternatives,and the effectiveness of the proposed method is verified by an example.(3)A preference approval probabilistic language group decision consensus reaching method based on PROMETHEE.This chapter puts forward an individual support information acquisition method and an expert weight determination method.An adjustment feedback mechanism is designed through the PAS distance deviation between individuals and groups.On this basis,a preference approval probabilistic language group decision consensus model based on PROMETHEE is constructed and the effectiveness of the proposed method is verified by relevant example.The preference approval probabilistic language group decision consensus reaching method based on PROMETHEE proposed in this thesis is applied to the site selection problem of compressed air energy storage power station project.The index evaluation system for the site selection of energy storage power station projects is constructed from six dimensions.On this basis,the comprehensive evaluation and classification of various alternative sites are carried out,which provides theoretical support for the site selection of compressed air energy storage power station projects.