A GIS-based multi-objective optimization strategy for urban automated external defibrillator deployment

Abstract Background Out-of-hospital cardiac arrest (OHCA) remains a critical urban health challenge due to persistently low survival rates and uneven access to automated external defibrillators (AEDs). Existing deployment strategies often rely on heuristic decisions, limiting their ability to address spatial disparities in emergency response. This study aims to develop a spatially informed, data-driven framework to optimize AED placement and improve urban emergency response capacity. Methods Using Qinhuai District in Nanjing as a case study, we applied Geodetector to identify the spatial determinants of OHCA risk. High-risk population density and its interaction with resident density emerged as dominant factors, informing the construction of a high-resolution AED Demand Index for candidate site selection. We then formulated a multi-objective optimization model to balance service coverage, perceived accessibility, and cost-effectiveness. The model was solved using an enhanced NSGA-II algorithm incorporating tabu search and 2-opt mutation. Results The optimized layouts demonstrated substantial improvements over existing AED configurations. The maximum service distance decreased by 55.3%, and overall satisfaction reached a near-optimal level of 0.997. The enhanced algorithm also achieved notable computational gains compared with standard approaches. Conclusions The proposed GIS-based optimization framework provides a scalable and evidence-based tool for improving AED deployment in dense urban environments. By integrating spatial risk detection with multi-objective optimization, this approach supports urban planners in reducing spatial inequities and strengthening emergency response systems.