Developing an integrated fuzzy credit rating system for SMEs using fuzzy-BWM and fuzzy-TOPSIS-Sort-C

Financial institutions employ credit rating systems to classify new borrowers, set loan terms, and determine collateral requirements. Small and medium-sized businesses (SMEs) are frequently found to be unorganised when it comes to financial data management. Credit risk assessment using unstructured financial data is multifaceted for financial institutions. Most extant credit rating systems are data-intensive and have been criticised for breaching their data distribution and structural assumptions. Such approaches may struggle to estimate SME's credit ratings accurately with sparse and missing data. To address the finance gap for SMEs, this study extends the application of expert systems by proposing a multi-criteria credit rating system. The proposed method utilises existing fuzzy-BWM to determine the weight of criteria and a newly proposed fuzzy-TOPSIS-Sort-C approach to classify SMEs according to their characteristic profiles. The suggested decision support system is expert-driven and built on sparse financial and non-financial data. BWM and TOPSIS were integrated with fuzzy set theory to overcome uncertainty when making decisions. This paper extends the prior TOPSIS-Sort-C method to fuzzy TOPSIS-Sort-C as a theoretical contribution. The suggested system efficacy is demonstrated by a case study that compares it with a commercial rating model using real-world data. The system performs with considerable effectiveness and a true positive rate when applied to credit rating. It could assist financial institutions in identifying potential SMEs for lending.