A Link Transmission Model with Variable Speed Limits and Turn-Level Queue Transmission at Signalized Intersections

The link transmission model (LTM) is an efficient and widely used macro-level approach for simulating traffic flow. However, the state-of-the-art LTMs usually focused on segment-level modelling, in which the traffic operation differences among multiple turning directions are neglected. Such models are incapable of differentiating the turn-level queue transmission, resulting in underrepresented queue spillbacks and misidentification of bottlenecks. Moreover, a constant free-flow speed is usually assumed to formulate LTMs, restricting their applications to model dynamic traffic management strategies involving variable speed limits (VSL) and connected and automated vehicles. This study proposed an extended LTM with VSL and turn-level queue transmission to capture the traffic flow propagation at signalized intersections. First, each road segment (RS) with multiple turning directions is divided into many free-flow and queueing parts characterized by the triangular fundamental diagrams. Then, the vehicle propagation within the link is described by the turn-level link inflow, queue inflow, and outflow, which depends on the free-flow speed changes. A node model involving an iterative procedure is further defined to capture the potential queue spillback, which estimates the actual flow propagation between the adjacent RSs. Simulated and field data were used to verify the proposed model performance. Results reveal that the proposed LTM predict traffic operations of complex intersections with multiple turning movements, VSL and signal control schemes, and enables an accurate yet computationally tractable representation of flow propagation.