URBAN TRAVEL TIME RELIABILITY
主讲人：Prof. dr. Henk van Zuylen
邀请人：涂辉招 副教授
时间：2016年10月31日（周一）10:00—11:30
地点：交通运输工程学院102会议室
主讲人简介
Henk van Zuylen is emeritus professor in the faculty of Civil Engineering and Geosciences at the Delft University of Technology. He is visiting professor at Tongji University (Shanghai), the Research Institute of Highway (Beijing), South West JiaotongUniversity, Hunan University (Changsha), South East University, and Changsha University of Science and Technology. He is director of the Chinese-Netherlands ITS training center and the research and consultancy company vZC.
His main research interests are: dynamic traffic management, network reliability, traffic monitoring systems, travelers’ behavior in response to traffic measures, technology assessment and innovation processes, and interactive and creative planning processes. He participated in the FANTASIE project executed for the European Commission to identify and assess new technologies for transport. He also participates in the European research program EDICT, in which the potential of new systems for public transport for European cities are assessed. Henk van Zuylen executed several studies in China, e.g. for Shanghai and Beijing to find ways to reduce congestion on the ring roads, for Changsha the assessment of the SCATS traffic control system and the development of the decision support system iTides for traffic management, for Chengdu for the development of a training system for emergency management.
主讲内容简介
Decisions made in urban traveling are not only based on (average) travel times but also on travel time reliability. Often the reliability is expressed in standard deviation of travel times, but skewness is at least as important. The travel time reliability can be determined from model studies and from empirical data.
Models should be calibrated and validated to produce realistic travel time distributions, which is a stronger requirement than just the calibration for average travel time. The determination of travel time distributions from empirical data requires a careful analysis of the data sources. Probe vehicles with GPS (often taxis) can give a biased travel time distribution. Also travel times from Automated Number Plate Recognition cameras have to be ‘cleaned’ from outliers.
There appears to be a positive correlation between average value of travel times and the standard deviation. The relation between average travel time, standard deviation and skewness also exist, but is more complicated. Traffic control can be optimized to maximize travel time reliability. Because of the correlation between average travel time and the standard deviation, the minimization of reliability also improves the average travel time.
In the presentation empirical data from the city of Changsha (Hunan) will be used as illustration together with results from simulations and signal optimization.
欢迎各位老师、同学参加同路人学术论坛！
交通运输工程学院研究生会
交通运输工程学院青年教师沙龙