A Graph-Based Adaptive Routing Optimization Algorithm for Emergency Evacuations

Emergency evacuation routing is a critical component of disaster management, ensuring individuals’ safe and timely relocation from hazardous areas to safe zones. Traditional evacuation planning relies on static, preplanned strategies that often lack adaptability to evolving disaster scenarios. This study presents a novel dynamic graph-based routing method that optimizes evacuation efficiency by adjusting road segment directions in real-time. The proposed algorithm iteratively distributes traffic flow across the network using a depth first search (DFS)-based approach, dynamically reconfiguring road topology to maximize the flow of vehicles. Implemented in the road network of Hoboken, New Jersey, the method demonstrated a 74% improvement in evacuation efficiency compared to benchmark methods that lack adaptive road reconfiguration. The findings highlight the transformative potential of this approach in supporting real-time decision-making during emergencies, offering a robust tool for adaptive evacuation planning.