Application of the fuzzy analytical hierarchy process for flood vulnerability assessment in Sylhet District, Bangladesh

Floods are among the most frequent and destructive natural hazards, and their frequency and intensity have increased under recent climatic and environmental changes. Sylhet District is one of the most flood-vulnerable regions of Bangladesh, where riparian and low-lying communities face recurrent flood-related challenges. This study assesses flood vulnerability in Sylhet District using the Fuzzy analytical hierarchy process (Fuzzy AHP), a multi-criteria decision-making technique that improves conventional AHP by incorporating uncertainty and ambiguity in expert judgment. Sixteen physical flood-conditioning indicators, including rainfall, elevation, slope, drainage density, land use and land cover, geology, flow accumulation, soil texture, hydrological soil group, topographic wetness index, permeability, curvature, water table depth, stream power index, distance from river, and normalized difference vegetation index, were integrated and weighted within a geographic information system environment. The consistency of the Fuzzy AHP pairwise comparison matrix was assessed using the Consistency Ratio, which was 0.0592, indicating acceptable consistency as it is below the threshold of 0.10. The generated flood vulnerability map categorized the district into four vulnerability levels: low, moderate, high, and very high. The findings reveal that 149.508 km2, representing 4.48% of the study area, falls within the low vulnerability zone, while 597.884 km2, accounting for 17.90%, falls within the moderate vulnerability zone. The high vulnerability class occupied the largest area, covering 1306.765 km2 or 39.12% of the district, while the very high vulnerability class covered 1285.870 km2 or 38.50%. Overall, high and very high vulnerability zones together accounted for 77.62% of Sylhet District, indicating widespread flood susceptibility across the study area. The final vulnerability map was validated using Receiver Operating Characteristic-Area Under the Curve analysis, producing an AUC value of 0.809, which indicates very good predictive performance. This spatial assessment provides critical insights for disaster preparedness, targeted flood mitigation, drainage improvement, land-use planning, wetland conservation, and evidence-based decision-making for sustainable regional development.