互联网定制公交的网络调度模型及其求解算法

针对互联网定制公交服务模式中存在的路线规划不合理、缺乏成熟的通用模型等问题,本文首先采用K-mean算法对居民的实时出行需求进行聚类,在此基础上构建了面向互联网定制公交,以最大需求服务率和最小费用为双层目标,以最大载客量和乘客时间阈值等因素为约束条件的动态网络调度模型,进而针对模型的快速求解设计了基于Hadoop平台的并行蚁群算法.以广州市为例,对模型与算法的有效性、实时性进行检验,结果表明,所提出模型和算法可应用于实际互联网定制公交业务中,对推进"互联网+交通运输"具有积极的意义.

The Network Scheduling Model and Its Solution Algorithm of Internet Customized Shuttle Bus

For Internet customized bus service model in the presence of route planning is not reasonable, lack of common model problems, K- mean algorithm is used to cluster the residents of real- time travel demand, based on constructed for Internet customized bus with the greatest demand service rate and minimum cost for the double target, its maximum capacity and passenger time threshold factors for the constraints of the dynamic scheduling model and for the fast solution of the model design parallel ant colony algorithm based on Hadoop platform. Taking Guangzhou city as an example, the real-time test validity of the model and algorithm. The results show that the proposed model and algorithm can be applied to the Internet customized bus service, has the positive significance to promote the "Internet plus transportation".