关键词: 城市交通, 高速公路收费流水数据, 通勤识别, 特征聚类

Abstract:

The use of high- speed commuter vehicles has also increased especially during major holidays, and traffic problems such as high-speed slow-moving and congestion have occurred. At present, the main method to solve the above traffic problems is traffic demand management, and the realization of targeted traffic demand management requires mining and analysis of highway toll ticket data, and grasping the running state and spacetime distribution characteristics of vehicles on the expressway. Based on the highway toll ticket data, this paper uses the K-means++ clustering method to identify the commuter vehicles which is using the highway, and further analyzes the time and space distribution characteristics of the commuter vehicles. From the perspective of commuting, it is of great significance to explore the distribution of urban commuter vehicles’rapid travel corridors and study the relationship between highway network and urban road network, which is to improve the efficiency of urban transportation system and alleviate traffic problems.

Key words: urban traffic, highway toll ticket data, commuting identification, feature clustering