Citation

Wright PH, Zador P. Transp. Res. Rec. 1981; 819: 8-17.

Copyright

(Copyright © 1981, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Engineering surveys were performed at 214 locations in Georgia [USA] where single-vehicle fatal rollover crashes occurred over a one-year study period. Similar surveys were made at comparison locations 1.6 km (1 mile) upstream from the crash locations. The most prominent roadway feature associated with fatal rollover crashes in Georgia was horizontal curvature, particularly along left curves. It was found that fatal rollover crash locations can be discriminated from comparison locations by curvature greater than 6 deg, the same value suggested in the fixed-object studies. Steep gradients were also found to be strongly and significantly associated with rollover crash location. The pattern of distribution of longitudinal slopes observed in earlier studies of fixed-object crashes, in which negative slopes tended to occur upstream and positive slopes downstream, was also apparent at rollover crash locations. Rollover sites were characterized by significantly larger changes in lateral slope at the shoulder edge than were found at comparison sites. The rollover sites were also more likely than the comparison sites to have embankments along the roadside but less likely to have trees and certain other spot fixed objects. Similarly, the rollover crash sites had longer embankments, banks, and ditches than were found at fixed-object crash sites than at the rollover crash sites.