Proactive congestion management via data-driven methods and connected vehicle-based microsimulation

Traffic congestion is a phenomenon that has been extensively explored by researchers due to its impact on reliability and safety. This research focused proactively detecting mitigating freeways fuzing conventional traffic data obtained from radar loop detectors with newer sources, such as Bluetooth connected vehicles (CV). Data-driven signal-processing techniques are develop algorithms use near- or real-time measurements predict the onset intensity level of congestion. The developed algorithm can be applied both low penetration CV-based datasets identify four types congestion, is, normal, recurring, other non-recurring, incident. also demonstrates advantage using travel time estimates calibrate microsimulation models over fixed point-based derivations spot speeds. Finally, set mitigation strategies consisting speed harmonization dynamic rerouting implemented in calibrated simulation network demonstrate their effectiveness reducing recurring non-recurring final derived effective predicting level, an overall mean prediction error 30.2%. A limitation algorithm’s methodology it cannot disentangle type when two more occurring simultaneously only predicts/classifies anticipated highest level. However, this does not impair user’s ability readily deploy appropriate alleviate predicted