Crash Severity Formulation and Analysis Under Extreme Weather Conditions

Weather conditions have tremendous impact on traffic crash frequency and severity. Weather-related traffic crashes result in 6,250 fatalities and over 480,000 injuries each year in the U.S. In the southwest region, unique weather extremes, including enormously high temperature, strong wind, flash flood, fog, dust, snow, etc., have resulted in more frequent and serious traffic crashes with significant loss in human life and property. Substantial research efforts are needed to better understand significant causal factors and their impacts on crash severities under extreme, adverse weather conditions for developing effective countermeasures and proper policies to minimize weather-related risks to traffic safety. This research project will develop a new approach for discovering the underlying patterns behind crash data and will identify a series of significant contributing attributes on crash severities impacted by weather extremes in the southwest region. Project objectives include: 1) designing and building a relational database that stores all the crash data, roadway geometric data, weather condition data, and traffic data; 2) developing a new hybrid approach to integrate multinomial Logit model with Bayesian network to discover the underlying patterns behind crash data; 3) identifying high crash risk locations on the selected region-wide highways and better understand crash causes under extreme weather conditions; and 4) recommending cost-effective countermeasures for reducing crash severities impacted by weather extremes. The hybrid approach is superior to the existing methods due to its flexibility to capture cause-effect relationships between contributing attributes and crash severity outcomes and better interpret their heterogeneous influence on crash severity outcomes from the attribute changes in terms of region-wide weather extremes, driver behavior, demographic features, and environmental characteristics. Through this research effort, a better understanding of significant contributing attributes and their impacts on crash severities can be achieved. Implementation will help transportation agencies to develop cost-effective countermeasures to reduce crash severities under extreme weather conditions and minimize the weather-related risks to traffic safety in the southwest region.


    • English


    • Status: Active
    • Contract Numbers:


      SPTC 14.1-86

    • Sponsor Organizations:

      New Mexico Department of Transportation

      State Highway Department Building
      1120 Cerrillos Road
      Sante Fe, NM  United States  87504-1149

      Research and Innovative Technology Administration

      University Transportation Centers Program
      1200 New Jersey Avenue
      Washington, DC  United States  20590
    • Performing Organizations:

      University of New Mexico, Albuquerque

      MSC01 1070
      Albuquerque, New Mexico  United States  87131-0001
    • Principal Investigators:

      Tarefder, Rafiquel

      Zhang, Guohui

    • Start Date: 20140801
    • Expected Completion Date: 0
    • Actual Completion Date: 20160731
    • Source Data: RiP Project 37454

    Subject/Index Terms

    Filing Info

    • Accession Number: 01543093
    • Record Type: Research project
    • Source Agency: Southern Plains Transportation Center
    • Contract Numbers: DTRT13-G-UTC36, SPTC 14.1-86
    • Files: UTC, RiP
    • Created Date: Nov 5 2014 1:01AM