A min-max fuzzy location-assignment model for workload-balanced drone-supported emergency logistics under flood disruptions

Flood disruptions can severely restrict ground access to affected communities, creating challenges for timely relief distribution. Drones provide an alternative for last-mile delivery, but their effectiveness depends on coordinated facility deployment, resource allocation, and uncertainty management. This study proposes a min–max fuzzy location–assignment model for workload-balanced drone-supported emergency logistics under flood disruptions. The novelty lies in jointly integrating facility activation, drone deployment, community service responsibility, and fuzzy demand-to-mission conversion within a compact pre-routing framework, thereby explicitly linking uncertain relief requirements to repeated missions and individual drone workloads. Travel time and relief demand are represented using triangular fuzzy numbers and transformed through an α -based approach. The bi-objective model minimizes the maximum drone workload and number of activated facilities, while community-level service outcomes are separately evaluated using post-optimization shortest processing time sequencing. The Chiang Mai case study achieves a 164.12-min makespan with two facilities, reducing makespan by 1.07% and 3.37% relative to comparable two-facility P-Center and K-Means solutions, respectively, while matching the best-performing benchmark configurations. Sensitivity and scalability analyses further demonstrate the effects of uncertainty, decision preferences, and problem size. The framework provides a transparent pre-routing decision-support approach for workload-balanced drone deployment under uncertain flood conditions.