The safety of vulnerable road users, including bicyclists, has become an increasing societal concern. Factors characterizing bicyclists’ crashes with motor vehicles may affect bicyclists’ injuries differently depending on the location of these crashes. The purpose of this paper is to provide a comprehensive analysis for identifying significant factors that affect bicyclists’ injury levels from crashes occurring at travel lanes and at non-travel lanes (e.g., crosswalks and bicycle lanes). For this purpose, this study applied Multinomial Logistic Regression on the Crash Report Sampling System data for three consecutive years. Bicyclists’ injuries were categorized into three levels: (1) Possible, (2) Moderate, and (3) Severe. The study found that running a separate model for each location provided better performances than running an aggregated model for both locations. Results showed common factors significantly associated with an increased likelihood of moderate and/or severe injuries at both locations. Five unique factors were associated with higher likelihoods of these moderate and/or severe injuries to bicyclists in the Travel Lane model, whereas two unique factors were found related to increased odds of these injuries to bicyclists in the Non-Travel Lane model. The results of this study contribute to a better understanding of bicyclists’ crash scenarios and the development of potential countermeasures by alternating some circumstances characterizing these crashes, when possible, to reduce potential injuries.


    Zugriff

    Download

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Identifying Factors Related to Crash Injury Levels Involving Bicyclists at Different Locations Through Crash Data Analysis


    Weitere Titelangaben:

    Transportation Research Record: Journal of the Transportation Research Board


    Beteiligte:
    Ammar, Dania (Autor:in) / Misra, Aditi (Autor:in) / Feng, Fred (Autor:in) / Bao, Shan (Autor:in)


    Erscheinungsdatum :

    2023-02-02




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Identifying crash-prone locations with quantile regression

    Qin, Xiao / Ng, Marie / Reyes, Perla E. | Elsevier | 2010