01504277

Component

http://onlinepubs.trb.org...ences/2011/RSS/program.pdf

Crashes are getting more attention in many countries. Crash prevention is vital in traffic safety especially on motorways where crashes are usually severe due to high speed. This paper proposes an approach for identifying traffic crash risk on the motorways in real-time using meteorological data and individual vehicle traffic data. The study site is on Swiss motorway A1 between Bern and Olten. Crashes interested in are rear-end and sideswipe. Traffic conditions when there is no crash, called non-crash cases, are matched with traffic conditions leading to crashes, called pre-crash cases. After matching process, relevant non-crash cases are selected to compare with pre-crash cases whereas; irrelevant non-crash cases are not further considered. In reality, if a new traffic condition is not relevant to any pre-crash case it can be declared as non-crash. Thereafter, models are developed to distinguish pre-crash and relevant non-crash cases. Variables called risk indicators are generated and single-variable models are developed. Each risk indicator has certain performance in differentiating pre-crash and non-crash cases. To improve that performance, risk indicators are fused into a unique high performance model. The authors results present high accuracy of more than 75% of non-crash and pre-crash cases correctly identified. Finally, to predict near future traffic crash risk in a real-time framework, traffic crash risk identified during the last intervals are used. The test with historical data shows that if two or three last intervals are identified as pre-crash, the chance for the coming interval to be pre-crash is more than 95%.

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

01501394

English

10p

2011

3rd International Conference on Road Safety and Simulation

Digital/other

Figures; Maps; Photos; References

Crashes; Prevention; Real time information; Rear end crashes; Risk; Side crashes; Traffic data; Traffic safety

Highways; Safety and Human Factors; I80: Accident Studies

TRIS, TRB, ATRI

Jan 9 2014 12:54PM