Pattern recognition and classification of fatal traffic accidents in Israel: a neural network approach

Prato, Carlo Giacomo, Gitelman, Victoria and Bekhor, Shlomo (2011) Pattern recognition and classification of fatal traffic accidents in Israel: a neural network approach. Journal of Transportation Safety and Security, 3 4: 304-323. doi:10.1080/19439962.2011.624291


Author Prato, Carlo Giacomo
Gitelman, Victoria
Bekhor, Shlomo
Title Pattern recognition and classification of fatal traffic accidents in Israel: a neural network approach
Journal name Journal of Transportation Safety and Security   Check publisher's open access policy
ISSN 1943-9962
1943-9970
Publication date 2011
Year available 2011
Sub-type Article (original research)
DOI 10.1080/19439962.2011.624291
Open Access Status Not Open Access
Volume 3
Issue 4
Start page 304
End page 323
Total pages 20
Place of publication New York, NY United States
Publisher Taylor & Francis
Language eng
Abstract This article provides a broad picture of fatal traffic accidents in Israel to answer an increasing need of addressing compelling problems, designing preventive measures, and targeting specific population groups with the objective of reducing the number of traffic fatalities. The analysis focuses on 1,793 fatal traffic accidents occurred during the period between 2003 and 2006 and applies Kohonen and feed-forward back-propagation neural networks with the objective of extracting from the data typical patterns and relevant factors. Kohonen neural networks reveal five compelling accident patterns: (1) single-vehicle accidents of young drivers, (2) multiple-vehicle accidents between young drivers, (3) accidents involving motorcyclists or cyclists, (4) accidents where elderly pedestrians crossed in urban areas, and (5) accidents where children and teenagers cross major roads in small urban areas. Feed-forward back-propagation neural networks indicate that sociodemographic characteristics of drivers and victims, accident location, and period of the day are extremely relevant factors. Accident patterns suggest that countermeasures are necessary for identified problems concerning mainly vulnerable road users such as pedestrians, cyclists, motorcyclists and young drivers. A "safe-system" integrating a system approach for the design of countermeasures and a monitoring process of performance indicators might address the priorities highlighted by the neural networks.
Keyword Accident factors
Accident patterns
Cluster analysis
Kohonen networks
Feed-forward back-propagation neural networks
Q-Index Code C1
Q-Index Status Provisional Code
Institutional Status Non-UQ

Document type: Journal Article
Sub-type: Article (original research)
Collection: School of Civil Engineering Publications
 
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 2 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Thu, 14 Apr 2016, 09:13:04 EST by Anthony Yeates on behalf of School of Civil Engineering