Traffic congestion often occurs at recurrent bottleneck areas on freeways, which would result in propagation and dissipation of kinematic waves and is more likely to cause rear-end traffic collisions afterwards. Thus, exploring surrogate safety measures for traffic collisions in the vicinity of recurrent bottlenecks would be beneficial for developing dynamic traffic control methods for preventing collisions. The primary objective of this study is to investigate the relationships between real-time traffic flow rate parameters and the traffic collisions risk at recurrent bottleneck areas on freeways. A three-year crash data and relevant real-time traffic flow rate information which was obtained from loop detectors implemented on a 10-km corridor of I-880 freeways in California were used to fit these relationships by using a Logistic regression modeling technique. After testing a series of real-time traffic flow rate parameters that aggregated into 5-min prior to the occurrence of each collision and comparing the different combinations of candidate traffic flow rate parameters, the combination of the average of upstream flow rate and the sum of standard deviation of upstream and downstream speed were selected as the optimal model. This model can be used as a surrogate safety measure for evaluating the real-time collision risk of rear-ends crashes at recurrent bottleneck areas on freeways.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Development of Crash Prediction Model for Rear-End Collisions at Recurrent Bottlenecks on Freeways


    Contributors:
    Chen, Ruoyun (author) / Zhang, Meihui (author) / Li, Zhibin (author) / Wang, Wei (author)

    Conference:

    The Twelfth COTA International Conference of Transportation Professionals ; 2012 ; Beijing, China


    Published in:

    CICTP 2012 ; 2591-2602


    Publication date :

    2012-07-23




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English