基于公交数据挖掘的时刻表排班协同换乘优化

为实现公交换乘协同排班,减少乘客出行换乘时间,本文对公交信息系统的IC卡数据及车辆GPS数据进行数据挖掘,提取换乘信息并对现有的发车排班进行优化.首先,构建了公交运行状态信息提取模型,提取现有的公交运行状态信息.在此基础上,设计了邻域搜索的公交时刻排班优化算法,得到最佳发车排班时刻表.为验证所提出方法的有效性,选取了成都市的56路和3路公交线路的实际数据进行案例验证.结果表明:通过优化排班的方法,在不改变现有的公交供需条件的前提下,可以有效实现协同换乘;与原有的公交服务相比,优化之后的公交服务能够更加贴近出行需求,提升线路之间的换乘衔接效率,从而提高公交服务质量.

Timetable Transfer-coordination Optimization Based on Transit Data Mining

The existing timetable is extract and optimized through exploiting the data of IC card and vehicle GPS to coordinate the transfer and save the transfer time. First, the model of data mining for transit data is proposed to obtain the existing transit operation information. And then, a neighborhood search algorithm is developed to optimize the timetable and get the best solution. The real data for two routes in Chengdu with the label of 56 and 3 are introduced to test the proposed method. The result shows that existing timetable can be optimized to coordinate the transfer without any change for the exiting supply and demand scenario, which indicates that the optimized transit service is more conform to the practical travel demand, which can lift the efficiency of transfer and improve the transit service level.