关键词: 车辆派遣, 时变, 动态, 遗传算法

Abstract: In order to accord with the actual traffic status in road networks，vehicle routing problem with time dependent travel time is researched.The authors modify the traditional vehicle scheduling model，total vehicle expense，total delay penalty，total vehicle use expenses are considered in the objective，so as to improve the applicability and universality of model.Since vehicle scheduling problem is NP-hard，parallel genetic algorithm is proposed to solve this problem.In order to verify the effectiveness of the parallel genetic algorithm，stand VRP problems are used.The simulation results show that the algorithm is very easy and feasible and compared with BC-Saving algorithm，it has rapid convergence ability and searching ability.

Key words: vehicle routing problem, time dependent, dynamic, genetic algorithm