METHOD FOR RELATING TYPE OF CRASH TO TRAFFIC FLOW CHARACTERISTICS ON URBAN FREEWAYS

In this paper, a method is developed for determining how crash characteristics are related to traffic flow conditions at the time of occurrence. The method involves nonlinear canonical correlation applied together with cluster analyses to identify traffic flow regimes that have distinctly different crash taxonomies. A case study using data for more than 1,000 crashes in Southern California identified 21 traffic flow regimes for 3 different ambient conditions: dry roads during daylight, dry roads at night, and wet conations. Each of the regimes has a unique profile regarding the types of crashes most likely to occur, and a matching of traffic flow parameters and crash characteristics reveals ways in which congestion affects highway safety.

  • Record URL:
  • Availability:
  • Supplemental Notes:
    • Publication Date: August 2003. California PATH Program, Institute of Transportation Studies University of California, Berkeley CA. Remarks: Also available on the World Wide Web at http://www.path.berkeley.edu/PATH/Publications/PATH/index.htm. Format: website
  • Corporate Authors:

    University of California, Berkeley

    Berkeley, CA  United States  94720-1720

    University of California, Berkeley

    California PATH Program, Institute of Transportation Studies
    Richmond Field Station, 1357 South 46th Street
    Richmond, CA  United States  94804-4648

    University of California, Irvine

    Institute of Transportation Studies
    4000 Anteater Instruction and Research Building
    Irvine, CA  United States  92697

    California Department of Transportation

    1120 N Street
    Sacramento, CA  United States  95814
  • Authors:
    • Golob, Thomas F
    • Recker, Wilfred W
  • Publication Date: 2003

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01001611
  • Record Type: Publication
  • Source Agency: UC Berkeley Transportation Library
  • Report/Paper Numbers: UCB-ITS-PWP-2003-12
  • Files: CALTRANS, STATEDOT
  • Created Date: Jul 7 2005 12:00AM