改进遗传算法在公交智能调度中的应用

Application of Improved Genetic Algorithm on Intelligent Scheduling for Public Traffic Vehicles

针对基本遗传算法易发生未成熟收敛及收敛速度慢的缺陷,将初始种群均匀化生成、选择算子采取自适应排序、交叉算子采用竞争择优的方式进行改进,并应用到公交调度优化问题中,建立符合实际情况的公交调度数学模型。通过仿真实例分析表明,改进后的遗传算法用于公交车辆智能调度是有效、可靠的。通过公交车排班表和发车频率图的比较可知,该结果是符合客流的实际分布情况的。

Initial population, selection operators and crossover operators are improved in this paper to overcome premature convergence and slow convergent speed of simple genetic algorithm. The improved genetic algorithm is applied to a practical transport dispatch optimizing problem. Through the simulation analysis shows that the improved genetic algorithm is efficient and reliable for intelligent scheduling buses. The comparison of bus scheduling table and frequency graph shows that the result is consistent with the actual distribution of passenger flow.