Abstract
It is essential to give priority to transit vehicles at signalized intersections in the megacities of China such as Nanjing. This study developed a transit signal priority (TSP) strategy for fixed-time signals with the use of techniques for advanced and continuous vehicle detection and prediction of arrival time. In the proposed model, transit vehicles were traced once they were positioned at one intersection and were advancing to the subject intersection. For instance, the location and the speed of a transit vehicle were measured every few seconds, and the prediction was made simultaneously by a model integrating historic and real-time data. With the predicted arrival time, the length of the transit signal phase would be adjusted to give the green light to the transit vehicle. The proposed strategy was evaluated with VISSIM simulation and compared with conventional TSP strategies (green extension and early green) in a typical roadway and traffic setting of Nanjing. Results indicated that the conventional TSP strategy could hardly result in significant improvement in transit delay and that the proposed strategy significantly reduced the transit travel time. Despite these findings, the adverse impacts on the delay of other road traffic by TSP could not be ignored. Results of the simulation model indicated that the proposed predictive TSP strategy was less disruptive to other traffic and that its performance in nonpeak periods was superior to that in peak periods. The results also suggest that a well-designed algorithm for the prediction of arrival time is the key to success in implementing predictive TSP control.
Get full access to this article
View all access options for this article.