The majority of night- time accidents on rural roads involve a single vehicle. Since horizontal curves are over- represented among the sites of such accidents the potential of roadway delineation as an accident countermeasure has received considerable research attention. Most of this research has used driver performance data, rather than accident data, as a criterion for evaluation. Moreover, only 'normal' drivers have been studied. Since over 75 per cent of single- vehicle, night- time accidents on rural road curves involve alcohol- affected drivers this study included the effects of alcohol on performance. Nine roadway delineation treatments were evaluated in a field experiment. Drivers were tested at BAC levels of zero and 0.05 per cent. Alcohol had a negligible effect on mean speed but increased the incidence of extreme lane positioning at the curve mid- point and led to more frequent departures from the lane. The most beneficial form of long- range delineation was chevron alignment signs which facilitated the execution of the drivers' preferred corner- cutting curve negotiation strategy. Short- range delineation (edgelines) generally added little to performance except that the use of a wide edgeline tended to reduce the extreme lane positioning evident for alcohol- affected drivers. Consequently, the 'optimum' delineation treatment seems to be one which combines chevron alignment signs with a wide edgeline (a). The ISBN of the microfiche version is 0 86910 117 X.
Samenvatting