Temporal Variation of Road Accident Data Caused by Road Infrastructure

Road accident data are collected annually in the United Kingdom (UK) using the STATS 19 accident report forms. The data is then stored on databases for subsequent analyses. The purpose of this paper is to present a methodology which may be used to extract knowledge from such databases in an automated manner using data mining techniques. The analysis examined accident causation due to road infrastructure features over a period of 5 years. The data mining system used was WEKA. Accident frequency histograms for each year were computed together with the probabilities of occurrence of a fatal/serious injury accident for different circumstances. Using Bayesian classifiers and an appropriate search algorithm in WEKA it was possible to analyze the data considered by producing rules that describe the interrelationship between different data. The success of this methodology was demonstrated by the agreement between the results and the matching road policies adopted in this period by the road authorities concerned. For example, results show that generally it is more probable that an accident will occur in daylight than in darkness. Another finding of the study is that the possibility of fatal accidents in darkness, when lights are unavailable, is twice as high as the possibility of serious injury accidents. It was also found that the probability of fatal and serious accidents occurring on wet surfaces had declined over the five-year period under consideration.

• Record URL:
  • http://onlinepubs.trb.org/onlinepubs/conferences/2011/RSS/2/Ehsaei,A.pdf
• 
• Supplemental Notes:
  • Distribution, posting, or copying of this PDF is strictly prohibited without written permission of the Transportation Research Board of the National Academy of Sciences. Unless otherwise indicated, all materials in this PDF are copyrighted by the National Academy of Sciences. Copyright © National Academy of Sciences. All rights reserved
• Corporate Authors:

  Transportation Research Board

  500 Fifth Street, NW
  Washington, DC  United States  20001
• Authors:
  • Ehsaei, Amirhossein
  • Evdorides, Harry
• Conference:
  • 3rd International Conference on Road Safety and Simulation
  • Location: Indianapolis Indiana, United States
  • Date: 2011-9-14 to 2011-9-16
• Publication Date: 2011

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; References; Tables;
• Pagination: 16p
• Monograph Title: 3rd International Conference on Road Safety and Simulation

Subject/Index Terms

• TRT Terms: Crash analysis; Crash causes; Crash data; Crash rates; Data mining; Databases; Fatalities; Infrastructure
• Uncontrolled Terms: Bayesian analysis
• Geographic Terms: United Kingdom
• Subject Areas: Highways; Safety and Human Factors; I82: Accidents and Transport Infrastructure;

Filing Info

• Accession Number: 01506187
• Record Type: Publication
• Files: TRIS, TRB, ATRI
• Created Date: Jan 30 2014 1:14PM