A coordinated signal priority strategy for modern trams on arterial streets by predicting the tram dwell time

Abstract

The modern tram has been recognized as a popular and efficient mode of public transit, and the signal priority for trams, particularly coordination on arterial streets, is a crucial and feasible strategy for improving the operation efficiency and level of service of trams. Different from the traditional “green wave” method, the fluctuations in dwell time at stops should be considered, and sacrifices to the automobile right-of-way must be controlled. Accordingly, a Support Vector Machine (SVM) model was proposed to predict the dwell time of trams at stops, which determines tram arrival times at the stop line. A two- sequential programming model to determine signal timings, including offsets and reallocations of green time, was then established. The upper level programming aims to minimize the stop frequency of a tram at the coordination region, and then, the optimal solution becomes the new restriction of the lower programming, whose objective is to minimize the overall delay of automobiles. This strategy was evaluated using the microscopic simulation software VISSIM using Tongjiang Road in Changzhou as the study region. The findings demonstrate that this strategy can significantly reduce more tram travel delays and stop rates on arteries than the conventional Transit Signal Priority (CTSP), and it definitively outperforms and the Static Two-direction Green Wave (STGW) method in decreasing these indicators of automobiles driving along arteries. Moreover, the satisfactory decreases in delay and average queue length at each intersection reflect a remarkable reduction or even elimination of negative effects from tram signal priority (tram SP). Finally, the stability of the proposed strategy in promoting both tram and automobile travel efficiency was proven by traffic volume sensitivity tests. The results and findings are meaningful for traffic managers to enhance the operation efficiency and attractiveness of trams by flexible signal control strategies. In addition, the conflicts between automobiles and tram SP can be successfully solved.