We compile citations and summaries of about 400 new articles every week.
Email Signup | RSS Feed

HELP: Tutorials | FAQ
CONTACT US: Contact info

Search Results

Journal Article


Rajamaki R. Res. Rep. Finn. Transp. Agency 2010; (41): online.


(Copyright © 2010, Finnish Transport Agency)






The objective of the research project was to study the effect of shoulder and centreline rumble strips on accident rate. The types under study were a centreline rumble strip on two-lane roads, shoulder rumble strips on two-lane roads, and shoulder rumble strips on motorways. The study material consisted of approximately 500 km of motorways and 3000 km of two-lane roads with rumble strips. The rumble strips were installed during 2004-2008, and accident data accident data were collected for the period 2003-2009. Accident data included all police-reported accidents. The method was a before-after analysis of head-on collisions and single vehicle accidents. This was done as a simple accident rate before and after computation, using main roads without rumble strips as a reference group. This was supplemented by an empirical Bayesian study of all police-reported accidents on two-lane roads. Speed development was also checked at thirteen automatic measurement points on two-lane roads. Centreline rumble strips showed a reduction of 10% in off-road left and head-on personal injury accidents on two-lane roads, when combining results obtained by different methods. All police-reported off-road right accidents increased by 4%, but it seems that only property damage accidents increased, not personal injury accidents. Based on the results, centreline rumble strips reduce all motor vehicle accidents (accidents without animals or unprotected road users) by approximately 3%. Research material on shoulder rumble strips on two-lane roads proved to be too sparse for reliable conclusions. Off-road right accidents seemed to have declined by 10-20%, depending on the research method, on roads with milled rumble strips, and increased on roads with rolled rumble strips. Off-road right accidents declined especially in winter, and on the basis of speed data, also average speeds declined in winter. Thus, the winter season off-road right accident decrease may be due to a reduction in speed, which is probably not due to shoulder rumble strips. The number of off-road left and head-on accidents seemed to have remained unchanged on roads with shoulder rumble strips. On motorways with shoulder rumble strips, the accident rate of police-reported off-road accidents decreased by 10%. Personal injury off-road right accidents decreased by 18% and off-road left accidents by 17%. Off-road accidents decreased more than other accident types on motorways. Due to the lack of a reference group, it is uncertain, however, whether the decline in motorway off-road accidents was due to shoulder rumble strips. There are about 3,000 km of main roads on which a centreline rumble strip would be appropriate for the traffic volume, but where it does not yet exist. If half of these roads could be equipped with a centreline rumble strip, it would prevent around a dozen personal injury accidents a year. This report may be found at


All SafetyLit records are available for automatic download to Zotero & Mendeley