As a novel intersection control strategy, contraflow left-turn lane (CLL) has been applied at signalized intersections to increase the capacity by appropriately using opposing receiving lanes as left-turn approaching lanes to achieve a continuous traffic flow. Through the control procedure, a special signal called pre-signal is set on the median to control the access of CLL. To further improve the efficiency of the CLL system when applying CLL on a set of intersections along the arterial, this paper models urban road network with CLL as a finite capacity queueing network with virtual queues. In this macroscopic traffic model, the location of the pre-signal is considered by the upper bound of the CLL queue as a flexible endogenous parameter. Furthermore, the proposed model brings pre-signal installation into the systematic consideration from spatiotemporal perspective rather than being fixed empirically in traditional approaches. In addition, to avoid possible direct conflict brought by arterial CLL system, rigorous restriction is considered based on the interactions between upstream and downstream queueing conditions. Meanwhile, an optimization framework is established to minimize the total travel delays, in which the upper bound of CLL queue, the green split of both main signal and pre-signal and the cycle time are explicitly considered. Simulation experiments via AIMSUN is implemented based on the field data of West Yan’an Road in Shanghai to verify the proposed methodology. Compared to the original design, the optimization of pre-signal installation improves the utilization efficiency of CLL control system and reduce travel delay of a set of intersections along the arterial. Keywords: Pre-signal installation, Arterial Management and control system, delay minimization

CICTP