Influential Evaluation of Data Sampling Techniques on Accuracy of Motorway Crash Risk Assessment Models

Recently, many studies have been focusing on real-time detection of rear-end and sideswipe crash risks on motorways thanks to the availability of traffic data provided by traffic detectors and crash record databases. In these studies, traffic evolution leading to individual crashes called pre-crash cases is considered and differentiated with traffic conditions where there is no crash recorded, called non-crash cases. This trend of studies reflects the need of identifying traffic crash risk in real-time in order that appropriate countermeasures could be implemented to prevent the risk from further developing and ending up with a crash. These studies are called disaggregate studies as the units of analysis are crashes themselves, according to (Golob et al., 2004). However, one of issues for these studies is the imbalance between pre-crash and non-crash cases because crashes are rare events on motorways and there should be certain traffic conditions for rear-end and sideswipe crashes to occur. Therefore, it is important to choose appropriate noncrash cases to compare with pre-crash cases, which is usually neglected in previous disaggregate studies. In the present paper, four different techniques for sampling non-crash cases is reviewed and compared using individual vehicle traffic data and crash databases altogether available from 2003 to 2007 on Swiss motorways A1.

• Record URL:
  • http://onlinepubs.trb.org/onlinepubs/conferences/2011/RSS/3/Pham,M-H.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:
  • Pham, Minh-Hai
  • Dumont, Andre-Gilles
• 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; Maps; Photos; References; Tables;
• Pagination: 9p
• Monograph Title: 3rd International Conference on Road Safety and Simulation

Subject/Index Terms

• TRT Terms: Countermeasures; Crash data; Databases; Real time control; Rear end crashes; Side crashes; Traffic data; Traffic safety; Vehicle detectors
• Geographic Terms: Switzerland
• Subject Areas: Highways; Safety and Human Factors; I80: Accident Studies;

Filing Info

• Accession Number: 01506384
• Record Type: Publication
• Files: TRIS, TRB, ATRI
• Created Date: Feb 3 2014 9:17AM