ABSTRACT Floods are among the most frequent and damaging natural hazards in India, particularly affecting low‐lying urban areas in the eastern regions such as Cuttack, Odisha. This study aims to develop a flood hazard susceptibility map for the Cuttack district in Odisha, India, using a combined Analytical Hierarchy Process (AHP) and fuzzy logic framework. Eight spatial parameters were used to generate weighted flood susceptibility surfaces. The AHP method is used for assigning weights to various flood hazard criteria: slope (16%), elevation (17.7%), drainage density (14%), rainfall (15.7%), distance from roads (2.8%), distance from rivers (6.2%), Topographic Wetness Index (TWI) (16.5%), and Land Use Land Cover (LULC) (10.2%). The AHP‐derived weights were integrated with fuzzified inputs using a fuzzy gamma overlay, resulting in a composite flood hazard map, which was classified into five susceptibility zones. Model validation using 125 ground truth flood points and Receiver Operating Characteristic (ROC) analysis yielded an Area Under the Curve (AUC) value of 0.85, indicating strong predictive accuracy. Additional performance metrics, including an accuracy of 0.82, F1‐score of 0.83, precision of 0.84, and recall of 0.81, further confirmed the reliability of the generated map. Sensitivity analysis, performed by varying key parameter weights by ±10%, showed less than 5% variation in high and very high hazard zones, confirming the model's robustness. The results show that approximately 44% of the study area falls under high or very high flood susceptibility zones, including densely populated and rapidly urbanizing regions such as Cuttack Sadar, Naraj, and Mundali. This approach demonstrates the utility of semi‐quantitative spatial modeling techniques for flood risk assessment in data‐constrained urban watersheds.
Pradhan et al. (Tue,) studied this question.