Particle Swarm Optimization/Greedy-Search Algorithm for Helicopter Mission Assignment in Disaster Relief

Abstract

In the immediate aftermath of a large-scale disaster, optimal helicopter rescue mission assignment is critical to saving many lives. However, the current practice in the field is mostly human centered. The Japan Aerospace Exploration Agency has been developing a decision support system for aircraft operation in order to promptly plan and execute rescue missions. The current research focuses on evacuation missions in particular and investigates the potential of particle swarm optimization with an integrated greedy search in aircraft resource management. A robust particle model is proposed that can reflect various helicopter properties as well as evacuation mission characteristics. Particle swarm optimization parameters are modified and set based on numerical simulations, and the values determined in this way are used in an optimization of disaster relief mission assignments based on real data obtained during the Great East Japan earthquake and tsunami of 2011. The swarm is initialized with a greedy-search algorithm to increase the calculation speed and improve the quality of the results. It is shown that a hybrid particle swarm optimization/greedy-search algorithm can be successfully adapted to disaster relief support systems and provide valuable analysis and decision-making information to the authorities in charge.