Exploring Clusters of Contributing Factors for Single-Vehicle Fatal Crashes Through Multiple Correspondence Analysis

Approximately 60% of roadway fatalities in Louisiana involve single vehicle crashes. In 2012, 384 out of a total of 652 fatal crashes in Louisiana were single vehicle crashes. In order to reduce the number of these crashes through effective crash countermeasures, safety policies, regulations and technological advancements, it is critical that causation factors of single vehicle Run-Off-Road (ROR) crashes are identified. The persistently high rate of ROR crashes in Louisiana, as well as the overall United States, calls for innovative research that can further benefit the on-going research in assessing the performance of roads, vehicles and humans. A single vehicle crash can be caused by various factors such as those related to roadway design, vehicle mechanical problems and, most importantly, the driver performance or behavior. More often than not, it is the combination of these factors that leads to a single vehicle crash being fatal. The current commonly used crash analysis methods lack the ability to identify the cluster of factors simultaneously. The Multiple Correspondence Analysis (MCA) method is an exploratory data analysis method that can visualize the patterns of the cluster consisting of crash contributing factors. This paper uses the MCA method to analyze eight years (2004-2011) of single vehicle fatal crashes in Louisiana in order to identify the important contributing factors and their degree of association with the crashes. The results reveal that the combination of impaired driving, particularly on undivided roadways with no streetlights at night, is most likely responsible for run-off fatal crashes. Other combinations resulting in ROR fatal crashes include: young males driving on a wet surface during the weekend, older female drivers (64 plus in age) on hilly terrain, distracted motorcycle drivers, and female drivers age 35-44 using cell phones while driving. The results of the MCA can guide the selection of crash countermeasures. The future work on the degree of association of the identified crash contributing factors can help safety management systems select the most effective and efficient crash reduction strategies.

• Supplemental Notes:
  • This paper was sponsored by TRB committee ANB20 Safety Data, Analysis and Evaluation.
• Corporate Authors:

  Transportation Research Board

  500 Fifth Street, NW
  Washington, DC  United States  20001
• Authors:
  • Das, Subasish
  • Sun, Xiaoduan
• Conference:
  • Transportation Research Board 93rd Annual Meeting
  • Location: Washington DC
  • Date: 2014-1-12 to 2014-1-16
• Date: 2014

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; References; Tables;
• Pagination: 17p
• Monograph Title: TRB 93rd Annual Meeting Compendium of Papers

Subject/Index Terms

• TRT Terms: Crash causes; Data analysis; Methodology; Nonparametric analysis; Ran off road crashes; Visualization
• Geographic Terms: Louisiana
• Subject Areas: Data and Information Technology; Highways; Safety and Human Factors; I72: Traffic and Transport Planning; I83: Accidents and the Human Factor;

Filing Info

• Accession Number: 01518152
• Record Type: Publication
• Report/Paper Numbers: 14-2411
• Files: TRIS, TRB, ATRI
• Created Date: Mar 12 2014 9:33AM