Multiobjective Genetic Algorithm for Pareto-Optimal Route Selection
Feng Wen, Mitsuo Gen · ASME Press eBooks · 2009
The multi-objective route selection problem (m-RSP) is a hot research topic in the car navigation system (CNS) for ITS (Intelligent Transportation System). In this paper, we propose a two-stage genetic algorithm (tsGA) to solve it. The purpose of the proposed approach is to create enough Pareto-optimal routes with good distribution for the car driver depending on his/her preference. In the first stage of the proposed solution approach, the accurate route searching ability of Dijkstra algorithm and the exploration ability of Genetic algorithm (GA) are effectively combined together to find enough Pareto solutions that are the initial population of the second stage. In the second stage, we adopt route-based genetic algorithm to calculate more Pareto solutions. Solutions provided by the proposed approach are compared with the current research to show the effectiveness and practicability of the solution approach proposed.