Improving Performance of Coordinated Signal Control Systems Using Signal and Loop Data

In this project, a traffic data collection system based-on the iDEN wireless network has been developed, lab tested and preliminarily tested in the field. The objective of the system is to provide a cost-effective and easy-to-maintain system that could still reliably provide traffic data over the wireless link. The project team developed an arterial performance measurement method that is based on the signal infrastructure data collected at PT2 Lab, U.C. Berkeley. The performance of the proposed model is illustrated by using a simulation study. The six-signal simulation network covers both heavily congested and light traffic intersections. The proposed model works well at both the intersection level and the arterial level. Estimation errors of travel time, number of stops and travel time reliability are insignificant. A general approach for robust signal optimization under demand uncertainty or flow fluctuations has been developed. The approach has been demonstrated in two different settings. The first one deals with the problem of synchronization of actuated signals along arterials along arterials. The other demonstration is to optimize the signal settings including the cycle length, green splits, offset points and phase sequences in an integrated manner, taking into account the day-to-day demand variations or uncertain further demand growth. The robust timing plans resulted from both models have been demonstrated in numerical tests to perform better against high-consequence scenarios without losing optimality in the average sense. Although the robust signal timing approach is applicable more widely, this report has been focused on timing models for pre-timed arterials.