The Methods of Extracting Spatiotemporal Characteristics of Travel Based on Mobile Phone data

Abstract

With the rapid development of urbanization in China, the problem of traffic congestion is mainly due to the rapid increase in traffic demand. Compared with a variety of travel behaviors and origin-destination spatiotemporal distribution, which is helpful for us to explore the cause of traffic congestion. Traditionally, travel surveys are time consuming and huge economic investment. The accuracy of the results were existed large errors. In recent years, data acquisition techniques and storage capabilities are developed rapidly, more and more human travel related data have been collected. These "Big Data" is brought both opportunities and challenges for extracting valid travel information. In this paper, the different trajectories of travel mode are match with traffic analysis zones through using geography information system. And then stay points are identified by clustering spatiotemporal characteristics of trajectories. Moreover, the OD matrix is established by different stay regions. The indices of travel and OD desire lines are chosen to analyze travel behaviors. Meanwhile, the OD volume distribution in rush hours are used to explain traffic demand in different urban area. The findings could be helped government make the appropriate decision of urban traffic system and made residents the better daily travel planning, which has significant reference value.