Adaptive DEA for clustering of credit clients

Abstract

Competition among the industrial and service organizations to provide their clients with financial and credit requirements through the banking facilities has considerably increased. On the other hand, the challenge facing these financial and credit resources is that they are limited. Therefore, the optimal allocation of these limited financial resources with the aim of maximizing the investment value is of a great priority for banks and other financial institutes. In this study, first the credit criteria for the applicants for bank facilities have been identified and then based on the improved Data Envelopment Analysis (DEA) technique, an effective method has been proposed for the client clustering. The improved DEA method which is called Golden DEA reduces the calculation time and increases the decision-making operations that ultimately lead to the improvement of the existing method. Also, the improved DEA model provides a short, dynamic and straight path in order to achieve greater efficiency for every institution. The priority provided by the improved DEA method has been compatible with the priority given by the existing DEA method for all of the understudied cases.