基于浮动车数据的上海市主干道车速特征分析.docVIP

基于浮动车数据的上海市主干道车速特征分析 王雪松 刘浩冰 （同济大学交通运输工程学院，上海 201804） 中文摘要：上海市45000辆以上的出租车拥有GPS数据，基于出租车浮动车数据可以获取城市主干道的运行状况。论文材料了177个主干道路段的657辆出租车的浮动车数据，分析其高峰、平峰的速度特征。利用分层模型分析了速度的影响因素。研究发现流量是速度的一个主要的影响因素，其它道路几何设计的指标同时也影响路段的速度特征，如路段长度、车道数、平曲线。 关键词：城市主干道；路段；车辆速度；浮动车数据；分层模型 Analyze Speed Features for Shanghai Urban Arterials using Floating Car Data Xuesong Wang, Haobing Liu School of Transportation Engineering, Tongji University, Shanghai, 201804, China Abstract: In Shanghai an opportunity exists to study urban arterial speed by accessing continuously updated GPS data from Shanghai’s taxis, of which more than 45,000 are equipped with GPS systems. This method is called the Floating Car Data (FCD) method. While the FCD method is currently limited to data from taxi (i.e., professional) drivers, it is still possible to use it acquire a comprehensive picture of city wide arterial speeds. GPS data from 657 taxis were captured during the morning Peak and Off Peak hours over one week period on the selected 177 segments of the arterial network. Multilevel regression analyses were then conducted to examine factors affecting the average speeds on the selected arterial segments. Results showed that amount of traffic significantly influenced speeds. However, we also found that roadway design variables including segment length, number of lanes, and horizontal curvature interact with saturation to influence speeds differently at different saturation levels. Keywords: Urban Arterials, Road Segment, Vehicle Speed, Floating Car Data, Multilevel Regression Model 1 Introduction The operational characteristics of urban arterials are usually characterized using measures such as travel time, vehicle speed, and saturation degree. These data are often recorded using static spot measurement methods such as laser guns, video, and loop detectors. Sometimes probe vehicles are used, and these can provide a more dynamic assessment of speed, but usually only a few are deployed at a time, and then only for short pe