Pythagorean Fuzzy Sets for Credit Risk Assessment‎: ‎A Novel Approach to Predicting Loan Default

The incorporation of Pythagorean fuzzy sets into credit risk assessment represents a relatively innovative approach for predicting loan defaults‎, ‎ offering a more precise and adaptable tool for financial institutions‎. ‎ Key customer information—such as credit history‎, ‎credit mix‎, ‎credit utilization‎, duration of ‎credit history‎, ‎income level, and employment stability—is obtained as linguistic variables‎. ‎These linguistic assessments are then transformed into Pythagorean fuzzy numbers‎. ‎ The combined Pythagorean fuzzy information is subsequently processed using the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS)‎‎. This approach employs a modified accuracy function to determine the Pythagorean fuzzy positive ideal solution and the Pythagorean fuzzy negative ideal solution‎. ‎For distance calculations within the TOPSIS framework‎, ‎spherical distance measurements are utilized. ‎ Alternatives are ranked based on the relative closeness coefficient and an adjusted index, collectively facilitating decision-making. The practical applicability of the proposed model is demonstrated through an illustrative numerical example.