Arterial traffic management in congested urban areas is challenging today due to increasing travel demand, multiple travel modes, and changing driving behaviors. The causes of arterial congestions are compound. While traditional traffic analysis, such as control delay estimation or green band analysis, is still widely used. It is receiving criticism for not identifying the real reasons for congestions.  In this talk, we present a new arterial traffic monitoring system driven by state-of-the-art traffic big data. The new system can discover the impact of multimodal signal operations (e.g., preemption) on the background traffic progression, spatio-temporal pattern of traffic congestions under actuated traffic signal coordination, vehicles’ ground-truth control delay and ground-truth queue length estimation at intersections. Such new information can provide more information to understand the reasons for congestions. The used traffic data include connected vehicle data (Wejo vehicle trajectories (10%~17% of all vehicles in the US), point-to-point travel time (Powered by Google), high-resolution traffic signal events, weather (precipitation), and probe vehicle trajectories (via a customized smartphone app). The objective of this effort is to develop novel measures of effectiveness for arterial traffic operations. Traffic managers can form new insights on arterial management toward better-informed arterial management using this system.

暂无