Optimizing the integrated off-ramp signal control to prevent queue spillback to the freeway mainline

This study presents an integrated off-ramp control model that incorporates the impact of ramp queue spillback on the freeway mainline in design of the off-ramp intersection's signal plan. The proposed model consists of a mesoscopic traffic module to capture the complex interrelations between the freeway mainline's traffic state and the lane-changing maneuvers triggered by offramp queue spillback, and a local signal optimization module to maximize the total benefit for both the freeway and intersection in the vicinity of an interchange. Using the simulation-optimization solution logic, the proposed model will first estimate the off-ramp queue length, produced from the signal module, to assess the resulting impacts on the freeway mainline, and then feed such impacts back to the control objective to iteratively search for the optimal cycle length and phase durations with the genetic algorithm. To assess the effectiveness of the proposed model, this study has conducted a two-stage evaluation, where the first stage uses the extensive field data to confirm the reliability of its freeway queue impact module and the second stage focuses on evaluating the benefits of accounting for the queue impacts on the freeway in design of the off-ramp signal plan under various traffic scenarios. The results from extensive tests have confirmed that the proposed model can effectively minimize the likelihood of causing off-ramp queues to spill back to the freeway mainline, and the failing to incorporate such impacts in design of the off-ramp signal will contribute significantly to the formation of freeway bottlenecks in the interchange area.