Joint multi-objective optimization of maritime SAR equipment deployment using high-risk area recognition and SA-APSO

Pre-positioning of search and rescue (SAR) equipment is a practical approach for rapidly and effectively responding to maritime emergencies. In this study, we present a comprehensive method to optimize the location and configuration of SAR equipment (LCSRE), which are critical determinants of emergency response efficiency. First, we use a random forest (RF) model to identify high-risk subareas based on maritime accident data and Automatic Identification System (AIS) data, providing a data-driven foundation for informed LCSRE decisions. Next, the Fuzzy Comprehensive Evaluation Method (FCEM) is employed to calculate a comprehensive impact index for external interference factors at candidate sites, quantifying their suitability. Finally, considering the supportive role of islands, we develop a model aimed at minimizing response times and overall configuration costs while improving service coverage. To solve the model, we propose a Simulated Annealing-enhanced Adaptive Particle Swarm Optimization (SA-APSO) algorithm. The numerical experiments demonstrate that the proposed method achieves an average cost reduction of 69.2% and a coverage improvement of 30.1% compared to the ship-only strategy. Moreover, by integrating the high-risk subarea recognition into multimodal deployment planning, it further reduces total cost by an additional 3.49%–18.36% and increases coverage by 5.23%–11.33% relative to a risk-neutral multimodal baseline. Compare to the actual 2024 configuration plan, the optimized solution increases coverage from 85.47% to 95.41% and reduces total cost by 11.63%. These results demonstrate the practical value and robustness of the proposed method for maritime SAR planning.