Three-Way Group Decision Making Based on Evidential Reasoning with Probabilistic Linguistic Term Assessments

In consideration of the different importance degrees that may be assigned to all possible linguistic terms, this paper investigates a novel three-way group decision-making method based on the probabilistic linguistic term set (PLTS) information systems. We first construct PLTS information systems based on multiple attributes. Considering the reliabilities of the experts, we determine the weights of the experts by the similarities of the information provided by the expert with regard to other experts. Subsequently, using the evidential reasoning (ER) method, we aggregate the information provided by all experts and obtain the conditional probability of each object. The introduction of the ER rules and the weights of experts successfully solve the problem of conflict between the evaluation information. Then an approach is presented to calculate loss functions and thresholds, which reduces the subjectivity of the decision-making process. Next, the decision result of each object is deduced based on the minimum-loss principle. Finally, a case study about the selection of mask foundries during the COVID-19 is used to demonstrate the effectiveness of our proposed method. And the superiority of our proposed method are proved by comparative analysis.