Highlights Crash risk models were developed to link safety to various phases and phase transitions. The Bayesian inference approach was used to estimate crash risk models. A safety level was assigned to phases and transitional states. Contributing factors to crash risks can be explained by phase transitions. Free flow can be divided into a true free flow and a platooned traffic state.

    Abstract Crash risk prediction models were developed to link safety to various phases and phase transitions defined by the three phase traffic theory. Results of the Bayesian conditional logit analysis showed that different traffic states differed distinctly with respect to safety performance. The random-parameter logit approach was utilized to account for the heterogeneity caused by unobserved factors. The Bayesian inference approach based on the Markov Chain Monte Carlo (MCMC) method was used for the estimation of the random-parameter logit model. The proposed approach increased the prediction performance of the crash risk models as compared with the conventional logit model. The three phase traffic theory can help us better understand the mechanism of crash occurrences in various traffic states. The contributing factors to crash likelihood can be well explained by the mechanism of phase transitions. We further discovered that the free flow state can be divided into two sub-phases on the basis of safety performance, including a true free flow state in which the interactions between vehicles are minor, and a platooned traffic state in which bunched vehicles travel in successions. The results of this study suggest that a safety perspective can be added to the three phase traffic theory. The results also suggest that the heterogeneity between different traffic states should be considered when estimating the risks of crash occurrences on freeways.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Safety performance of traffic phases and phase transitions in three phase traffic theory


    Beteiligte:
    Xu, Chengcheng (Autor:in) / Liu, Pan (Autor:in) / Wang, Wei (Autor:in) / Li, Zhibin (Autor:in)

    Erschienen in:

    Erscheinungsdatum :

    2015-08-24


    Format / Umfang :

    13 pages




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch





    Phase Transitions in Nonequilibrium Traffic Theory

    Zhang, H. M. | British Library Online Contents | 2000


    The comfortable driving model revisited: traffic phases and phase transitions

    Knorr, Florian / Schreckenberg, Michael | Tema Archiv | 2013



    Phase Transitions in Traffic Flow

    Kerner, B. S. | British Library Conference Proceedings | 2000