Real-time Arterial Performance Measurement Using BRT Probe Data and Signal Timing Data

Real-time arterial travel time estimation has been a very challengeable task due to the complexity of urban street traffic behaviors and lack of high resolution and high quality data. However, travel time information is considered as one of the most important tools to manage arterial flow. This study developed a series of data processing methods to estimate arterial travel time and LOS based on real-time BRT bus data and signal timing data. BRT buses have a few advantages over local buses to serve as probe vehicles to measure arterial performance. A few data processing procedures were introduced in this paper to estimate the arterial travel time by converting bus trajectory to general traffic travel time. The arterial travel time is estimated by adding the bus queuing delay, average traffic signal waiting time, and free-flow travel time. To verify the effectiveness of the developed method, a field test was conducted. The Valley Transportation Authority (VTA) BRT buses together with Intelli Drive corridor were selected as the data sources. The testing results demonstrate that the proposed method estimated arterial travel time and level of service well based on BRT probe data and traffic signal data. The root mean square error (RMSE) and root mean square percentage error (RMSPE) for the estimation of travel time are 49 seconds and 9%, respectively. The accuracy for the estimation of LOS is 73%.

摘要由于城市道路交通行为的复杂性以及高质量交通数据的缺乏，实时估计城市主干道的旅行时间具有一定的难度。本文基于实时的快速公交（BRT）以及信号配时数据，作了一系列研究，用以估计主干道旅行时间以及交通服务水平（LOS）。本文通过将公交车的排队延时时间、平均信号灯等待时间和自由流旅行时间综合在一起来实现旅行时间的估计。并以Valley Transportation Authority（VTA）BRT和智能驾驶系统作为数据源进行实验。实验结果表明本文提出的估计主干道性能的方法非常有效。其中根均方误差（RMSE）和根均方百分比误差（RMSPE）分别为49s和9%，LOS估计的精度高达73%。