|M.Sc Student||Keilani Moua'taz|
|Subject||Evaluation of Traffic Flow Characteristics Based on Traffic|
Data Collected by Probe Vehicle Equipped with
|Department||Department of Civil and Environmental Engineering||Supervisors||Professor David Mahalel|
|Professor Gilad Even-Tzur|
|Full Thesis text - in Hebrew|
This research presented a traffic data collection technique by probe vehicle equipped with GPS, to evaluate the traffic flow characteristics.
The data collection was done during 4 days in June and July 2003. The research was done in Ebn-Gavirol street in Tel-Aviv, a central urban street, 3100 meters long, a flat street with typical section of two tracks and two lanes. There are 14 signalized junctions along the street. All the traffic lights are coordinated along the street by a green wave program. The data were collected in various times during the day, in order to characterize the various activity hours. The GPS data were collected in the first three days in 5 seconds interval and on the fourth day in one-second interval. 45 runs were done from south to north, and 44 runs from north to south. The data collecting was done with the accuracy of few centimeters in determining the vehicle location in space.
The research presented a detailed methodology for preparing a program for collecting traffic data, determining the sample size, planning and performing the survey by GPS, the probe vehicle's driving technique, data reduction, quality control and processing and summering the data.
According to the collected information an analysis of sequent information about the location of a vehicle in space and in time was done. We evaluate the traffic flow characteristics along the street including time-space diagram, travel speed profile, average speed, speed variability, accelerations, acceleration’s noise, travel times, stopping times at traffic lights, control delay, and queues.
The data collected by probe vehicle equipped with GPS has a remarkable advantage with the possibility to calculate the acceleration and the acceleration’s noise.
In the research we performed a comparison between traffic data collected by detectors and traffic data collected by a probe vehicle equipped with GPS in an inter-urban road segment. We have shown, that there is no relationship between an average speed of vehicles in a point (measured by a detector) and the speed of one vehicle (measured by GPS).
In the research, we presented a tool to examine the degree of matching between the green wave program of the traffic lights and the actual traffic condition. This tool enable to update the offsets of the green wave program in order to achieve a more efficient operation of the traffic lights system along the street.