基于精英选择遗传算法的需求响应公交规划

为提高公交系统服务质量,需求响应公交是近年来被提出的新型运营模式,其可以根据乘客需求定制公交线路。对需求响应公交的规划包括点规划和路径规划两阶段。首先通过K-means算法实现公交零食停靠点规划,其次在路径规划的过程中,提出了一种基于精英选择的遗传算法。最后,通过100名乘客需求的实例试验,需求响应公交最大可以使企业盈利162.30元,并证实了基于精英选择遗传算法相比与基于轮盘赌选择遗传算法具有更快的收敛速度和更好的搜索结果,精英种群的规模设为15%左右时求解结果较好。

Based on Elitist Selection Genetic Algorithm for Demand Responsive Transit Planning

In order to improve the quality of public transportation system services,demand response transit is a new operation mode proposed in recent years.It can plan bus lines according to passenger demand.The plan for demand response transit includes two stages bus stop planning and routing planning.Firstly,the K-means algorithm is used to realize the planning of bus stop.Secondly,in the process of path planning,a kind of genetic algorithm based on elite selection is proposed.Finally,through an experiment of 100 passenger demand,the demand response transit can make the enterprise profitable 162.30 yuan.The experiment also confirms that the elite selection genetic algorithm has faster convergence speed and search results than the roulette wheel selection genetic algorithm.When the scale of the elite population is set to about 15%,the result is more prefer satisfied.