A semantic model for computing with flexible linguistic expressions and the application in MCGDM

The flexible linguistic expression (FLE) is a generalization of the hesitant fuzzy linguistic term set and linguistic distribution. An FLE expresses a preference in the form of hesitant fuzzy linguistic term sets with the symbolic proportions. The existing studies mainly utilize a symbolic way to compute with FLEs. To provide a semantic model for computing with FLEs, this paper proposes to represent an FLE with a triangular fuzzy number. Two approaches based on the maximum entropy principle are introduced to compute the triangular fuzzy membership functions for complete and incomplete FLEs. A multi-criteria group decision making (MCGDM) model based on FLEs is then proposed, in which the criteria are divided into different subsets of sub-criteria which are selected for different decision makers. A consensus reaching process is introduced to improve consensus of the decision makers by modifying weights of the sub-criteria. Finally the MCGDM model is applied to an autonomous driving evaluation problem.