In order to improve the quality of group decision making and obtain more reliable decision results,this paper es-tablishes a hierarchical feedback mechanism and multi-level measurement model for the reliability problem of heterogeneous multi-attribute group decision making considering self-confidence in linguistic environment.First,experts in the subgroups under each attribute of each alternative are divided into three categories according to the high,medium and low self-confi-dence by combining the three-way decision-making ideas.Then,the decision reliability is quantitatively analyzed from the three levels of expert composition,self-confidence and consensus of subgroups successively.A feedback mechanism is es-tablished from coarse to fine and refined layer by layer,providing a way for reliability adjustment and control of decision-making process.Considering the proportion of high and medium confidence experts in the sub-groups as well as factors such as confidence level and consensus level,a multi-level reliability measurement model is constructed,and two-level reliability judgment rules for scheme-level decision-making and overall decision-making are formed to effectively improve the interpret-ability of decision-making results.On this basis,a reliability driven heterogeneous multi-attribute group decision-making method is proposed.Finally,the feasibility and effectiveness of the proposed method are analyzed and verified by the case of supplier selection.
关键词
多属性群决策/专家自信度/决策可靠性/群体共识度/三支决策
Key words
multi-attribute group decision making/self-confidence of expert/reliability of decision-making/group consen-sus degree/three-way decision making
维普
万方数据