A Fuzzy Expert Systems for Bank Real Customers' Credit Rating

Reimbursement delays by the customers leads the banks to suffer major problems including the inability of the central bank to repay loans more than the amount of credits offered to the customers. Importance of credits in the banking industry and its critical role in economic and employment growth has led to the development of several models to evaluate the applicant's credit. But many of these models are classic and they could not evaluate the customer's credit in optimal way. Thus artificial intelligence models were considered in this field. In this paper a model based on fuzzy logic is presented to bank real customers’ credit rating. In this research a structured model was obtained for determination and categorization of input variables for application in the system by factor analysis then an expert fuzzy system was modeled consisting of six steps. In the first step, a fuzzy system was designed with inputs of financial capacity, support, reliability, repayment record and output of customer credit.In the second step input and outputs were partitioned, in the third step these partitioned inputs and outputs were converted into fuzzy numbers. The fuzzy inference was compiled in step four. In step five the de-fuzzy was conducted. Finally the designed model was tested in step six. The model proposed by using fuzzy specialist system has better performance than the studied bank experts credit rating.