A novel method to estimate incomplete PLTS information based on knowledge-match degree with reliability and its application in LSGDM problem

In recent years, large-scale group decision making (LSGDM) has been researched in various fields. Probabilistic linguistic term set (PLTS) is an useful tool to describe evaluation information of experts when solving the LSGDM problem. As decision-making becomes more complex, in most cases, decision makers are unable to give complete evaluations over alternatives, which leads to the lack of evaluation information. To estimate missing information, this paper proposes a new method based on knowledge-match degree with reliability that knowledge-match degree means the matching level between evaluation values provided by individual and ones from group. The possession of reliability associated with evaluation information depends on fuzzy entropy of PLTS. Compared with previous methods, this approach can enhance accuracy and reliability of estimated values of missing evaluation information. Based on this method, we develop a complete decision process of LSGDM including information collection, subgroup detecting, consensus reaching process (CRP), information aggregation and ranking alternatives. Subsequently, a case about pharmaceutical manufacturer selection is used to illustrate the proposed decision method. To verify effectiveness and superiority, we make a comparative analysis with other methods and finally draw a conclusion.