基于手机信令大数据的城市居民出行OD预测

Prediction of Urban Residents’ OD Matrix Based on Mobile Phone Big Data

为得到具有客观性、动态性的居民出行起止点间交通出行量(OD trips),以中国移动手机信令大数据为基础,阐述了手机信令数据的来源与构成,针对城市居民出行目的,通过数据分析,设置10分钟为划分出行活动的阈值,提出基于手机信令数据获得的OD矩阵的原理与方法,应用数据挖掘技术,对贵阳市居民出行OD矩阵进行了预测.为进一步加强大数据与城市智能交通系统的融合,将得到的城市居民出行OD矩阵应用到真实的城市公交线网中.最后通过MicroCity平台,将数据结果实现可视化.分析结果表明:与传统OD调查方式相比,利用手机信令大数据获得的OD矩阵客观性、动态性较强,信息采集分析周期短,更容易与城市智能交通系统融合,应用可视化平台,可实时反映出城市公交运营状态,为城市智能规划和调度提供重要参考.

In order to obtain the objective and dynamic residents’ Orgin-destination trips(OD trips), this paper is based on the mobile phone big data of China Mobile, and expounds the source and composition of mobile phone data. For the purpose of urban residents’ trips,set 10 minutes as the threshold to divide the travel activity with the data analysis. This paper studies the principle and method of OD matrix based on mobile phone data. With the data mining techniques, the OD matrix of Guiyang city residents is predicted. In order to strengthen the integration of big data and urban intelligent transportation system, the OD matrix is applied to the urban public transport network in Guiyang. Finally the results are shown with the software MicroCity. Analysis result shows that the OD matrix based on the big data mining technology is more objective and dynamic than the traditional OD survey method. Meanwhile, the information collection and analysis cycle is short. It is easier to integrate with the urban intelligent transportation system. With the visual platform, the urban bus operation status is shown in time. It can provide reference for the urban intelligent planning and scheduling.