A partial binary tree DEA-DA cyclic classification model for decision makers in complex multi-attribute large-group interval-valued intuitionistic fuzzy decision-making problems

This paper proposes the idea of combining “interest groups” with the practical decision information to classify the decision makers (DMs) in complex multi-attribute large-group decision-making (CMALGDM) problems in interval-valued intuitionistic fuzzy (IVIF) environment. It constructs a partial binary tree DEA-DA cyclic classification model to achieve the multiple groups’ classification of DMs. Not only does this method provide references for the classification of DMs when the decision information is known, but it also lays foundations for DMs’ effective weight determination and the aggregation of decision information. First, this paper normalizes all of the cost attributes into benefit attributes to avoid the wrong decision result. Second, it employs the C-OWA operator to transform IVIF number (IVIFN) samples into single-valued samples. With respect to this transformation, this paper provides the corresponding BUM functions of DMs according to their risk attitudes; therefore, the preference information of DMs can be more objectively aggregated. Third, this paper adopts the partial binary tree DEA-DA cyclic classification model to present an accurate classification of DMs. Thus, for each interest group, group members with different interest preferences can be distinguished and distributed to the appropriate groups. Finally, to show the feasibility and validity of the model, we give an illustrative example.