4th International Conference on Intelligent and Fuzzy Systems (INFUS), Bornova, Turkey, 19 - 21 July 2022, vol.505, pp.505-513
The assessment of the client's solvency is an integral part of the work of any commercial bank or microfinance organization to determine the possibility of issuing a microloan to a particular borrower. A preliminary analysis of the solvency and credit history of a potential microcredit borrower allows to assess in advance the risks of non-repayment on time or the probabilities of timely repayment of a bank loan. Even though numerous scientific studies (including scoring analysis methods) are devoted to solving such problems, this article discusses an unconventional approach to the multi-criteria assessment of the solvency of potential microcredit borrowers, which is based on a fuzzy analysis of their solvency indicators. In particular, the proposed fuzzy inference system in combination with existing statistical methods for data analysis can serve as an additional effective option for an information system for supporting credit decision making. This approach was tested on the example of credit histories of ten arbitrary microcredit borrowers and was compared with the corresponding solvency assessments obtained using scoring analysis, Pareto and Bord methods. The practice of bank lending has shown that the combined approach to assessing the solvency of potential borrowers of microcredits is more balanced, that is, it allows to identify a group of people more reliably with high credit discipline and those who create an area of increased credit risk.