<![CDATA[Background: Evacuation route planning in disaster situations is a complex problem involving dynamic environmental conditions, infrastructure constraints, and real-time traffic changes. Single deterministic or metaheuristic algorithms often face limitations in handling large-scale and dynamic networks. Objective: This study aims to develop and evaluate a hybrid optimization algorithm integrating Dijkstra and Ant Colony Optimization (ACO) to determine the shortest and safest evacuation routes using real-time geospatial data. Method: Road networks were modeled as weighted graphs derived from OpenStreetMap and traffic APIs. The hybrid approach combines Dijkstra for initial shortest-path computation and ACO for global route optimization under dynamic conditions. Performance was evaluated based on travel distance and computational time through simulation scenarios involving 500–2000 nodes. Results: The hybrid algorithm achieved the shortest travel distance (10.7 km) compared to Dijkstra (12.8 km) and ACO (11.9 km), reducing route length by 16.4% relative to the baseline. Computational time remained efficient at 0.96 seconds, making it suitable for real-time applications. Overall: The findings demonstrate that hybrid optimization provides a more adaptive, robust, and efficient solution for disaster evacuation route planning.]]>
Copyrights © 2024
