<p>Flash floods pose a serious threat to Bangladesh, especially in low-lying, disaster-prone areas like Feni district. Despite numerous national disaster management policies, localized vulnerability assessments remain poorly integrated into flood planning. This study addresses this gap by integrating empirical vulnerability driver analysis with policy evaluation, which is a novel approach rarely undertaken in Bangladesh’s flood management research. Using a cross-sectional observational design, the study combines policy evaluation, remote sensing analysis , and household surveys to identify multi-dimensional (economic, social, environmental, and physical) drivers of flash flood vulnerability. Geospatial analysis mapped flood-affected areas, while a structured questionnaire assessed 30 vulnerability indicators across affected communities. Policy document analysis revealed critical gaps in current frameworks. Results indicated pervasive vulnerability across all dimensions (average scores: 3.91–4.06), with economic and social factors being particularly severe. Key policy deficiencies included weak early warning systems, inadequate inter-agency coordination, and limited community participation. The study proposes a comprehensive recommendation framework addressing institutional, community, and structural interventions. By offering a reproducible, empirically grounded approach linking vulnerability assessment with policy reform, this study advances disaster risk reduction discourse and provides actionable guidance for flood resilience planning in Bangladesh and comparable high-risk regions.</p>