EDAS-Sort-B: An Extension of the EDAS Method for Sorting Problems through Boundary Profiles

Multiple Criteria Decision Analysis (MCDA) sorting models are highly relevant for solving real-world problems. Thus, in the literature, the great majority of MCDM methods tackled the choice or ranking problems unlike the sorting approaches although assigning alternatives to predefined homogeneous categories (classes) presents a complex problem. Thus, in this paper, we tackled the sorting problematic using the EDAS “Evaluation based on Distance from Average Solution” method. It is used for ranking alternatives, in a decreasing order, according to their Appraisal Scores (AS). Nevertheless, the current version of EDAS method cannot deal with sorting problems. Since a great majority of real-world decision-making problems are modeled as sorting ones, we proposed a new sorting MCDM method called EDAS-Sort to cope with decision problems requiring assigning alternatives to predefined and ordered classes. Given that we dealt with classes defined by their boundary profiles, the proposed method is called EDAS-Sort-B. To demonstrate and underline it, we presented a case study on a bank agency located in Sfax, Tunisia which aims to assign clients requesting loans to three predefined and ordered categories:  very solvent, solvent, and doubtful according to various criteria. Therefore, the head of the bank agency (decision maker) will gain insight on the client’s profile and whether he is trustful or not to repay the loan. Thus, the EDAS-Sort-B is effective for solving problems requiring assigning alternatives to predefined and ordered categories. Thereupon, the main advantage of EDAS-Sort-B is to help the DM “Decision Maker” to take a real-time decision related to alternatives’ assignment.