A Robust GA/PSO-Hybrid Algorithm in Intelligent Shipping Route Planning Systems for Maritime Traffic Networks

Abstract

The development of intelligent shipping route planning systems is important for maritime traffic networks, and has attracted considerable attention in the field of marine traffic engineering. In practical applications, the traditional experience-based planning scheme has been widely used due to its simplicity and easy implementations. However, the traditional manual procedure is experience-dependent and time-consuming, which may easily lead to unstable shipping route planning in different waters. The purpose of this study automatically and robustly determines that the optimal shipping route is based on artificial intelligence approaches. It is general that Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) are almost the most popular methods in route planning. These two heuristic-based optimization techniques benefit from their specific advantages when solving different optimization problems. In this paper, we proposed a hybrid heuristic scheme by integrating GA and PSO to improve the accuracy and robustness of shipping route planning in restricted waters. The experimental results about both synthetic and real-world problems have demonstrated that our proposed hybrid approach outperforms the existing schemes in terms of both accuracy and robustness, and the approach is helpful for optimizing maritime traffic network for the links of terminals.