Intelligent scheduling of contraflow control operation using hierarchical pattern recognition and constrained optimization

Contraflow operation is frequently used for reducing traffic congestion near tunnels and bridges where traffic demands from the opposite directions vary periodically. In this work, a hierarchical traffic flow pattern recognition mechanism that is capable of accurately predicating the coming traffic demands using the sensed traffic flow data was developed to improve the real-time control of contraflow operation. This mechanism works in tandem with a constrained optimization scheme that minimizes traffic delays at the bottleneck passage point. Application of the proposed method to the dynamic contraflow operation control at the George Massey tunnel in Vancouver, BC Canada is discussed as a case study. The approach leads to significant reduction of traffic congestion, and has a great potential to be applied to similar contraflow control problems.