Document Type : Original Research Paper

Authors

• A. Alinezhad
• S. Kashanifar

Department of Industrial Engineering, Faculty of Industrial and Mechanical Engineering, Islamic Azad University, Qazvin Branch, Iran

Abstract

In this research, in order to manage and control the credit risk of customers, from the combination of two models of discriminant analysis and data coverage analysis to detect the presence or absence of an overlap between two groups by means of a separating hyperscreen and assuming the existence of each observation with independent characteristics with The presence of fuzzy data, the observations were categorized into two categories of good customers and bad customers. The variables of this research were selected from the 6C method, and out of the 17 selected indicators, using the Delphi method, 8 influential indicators were included in the research model. These indicators were used for 83 real customers of a leasing company who received facilities during 2013 and 2014. The results show that each of the observations is definitively placed in the category of good customers and bad customers, and with the arrival of each new observation, its credit status is predicted.

Keywords

• Data Envelopment Analysis/ Discriminant Analysis
• Fuzzy Data
• Credit Risk
• Credit Scoring