|M.Sc Student||Zuabi Shua'a|
|Subject||Evaluation Based on Detectors of Traffic Quality in Urban|
|Department||Department of Civil and Environmental Engineering||Supervisor||Professor David Mahalel|
Congestion in major urban areas is an increasing problem. Real-time monitoring of traffic helps identify congestions and helps to remove them as soon as possible. This could reduce delays, increase safety, improve air quality and improve quality of flow. There is a strong interest in developing an easy way to monitor arterial streets.
The purpose of this research is to develop models that can evaluate in real time the quality of flow and detect congestion. This research deals with urban streets linked to signalized intersections.
The data collection was based on inductive-loop detectors that were installed along an arterial in Jerusalem, In addition, a video camera was installed and the visual data were compared to the detectors data.
This research focuses on the congested condition, because it is the most problematic condition that requires serious interest and deep investigation.
This research presents 3 major flow conditions:
· Short queue (Low flow) - maximum 11 vehicles in the queue.
· Medium queue (Under saturated flow) - number of vehicles in the queue is above 11 but less than 18.
· Long queue (Over saturated flow) - minimum 18 vehicles in the queue.
Flow classification is based on the queue length that was recorded by the video. This research presents general algorithms for flow classification. The algorithms require two kinds of input: (1) intersection controller input, which includes volume and occupancy gathered from detectors and cumulative green-time, and (2) user input, which includes the number of lanes, turning percentages, detector location and historic capacity values (if it is available).