Citation

Zegeer CV, Perkins DD. Transp. Res. Rec. 1980; 757: 25-34.

Copyright

(Copyright © 1980, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

A critical review was conducted of available studies on the effect of shoulder width and condition of safety. A set of criteria was established for use in evaluating the reliability of the conclusions reported in past studies on this subject. Most studies based conclusions on the analysis results of pre-1955 accident data and only two of them considered the effect of shoulder width on related accident types (run-off-the-road and head-on accidents). Several studies did not control for the effect of intersections and differing roadway alignment (tangent or curved sections) on rural highway accident rates. Wider shoulders were found to be associated with safer conditions in the studies that were judged most reliable. Shoulder stabilization was effective in reducing accident rates on two-lane roads, particularly on identified high-accident sections. Shoulder widening was found to be cost effective on high-accident sections that had shoulder widths less than 1.2 m (4 ft). In particular, sections of rural two-lane roads that had six or more run-off-the-road or head-on accidents per 1.6 kilometer per year were likely to result in benefit/cost ratios greater than one. Shoulder widening was not cost effective, however, for low-volume roads (less than 1000 vehicles/day) that had a low frequency of accidents. Shoulder paving or stabilization is generally desirable from a safety standpoint, although its cost-effectiveness is not all established. Rural winding highway sections and sharp horizontal curves were recommended as the best candidates for shoulder improvements, particularly those that have a high incidence of run-off-the-road and head-on accidents. Should widths of 1.8-2.7 m (6-9 ft) are recommended for rural, two-lane roads.