Safety fences on motorway central reserves have been accepted as necessary safety features since 1973. However on all-purpose dual carriageway roads where design standards vary it was not until 1981 that such barriers were generally accepted and then only on heavily trafficked new dual carriageways. The paper describes the initial results from a study designed to develop installation criteria for such fences. The overall objectives of the study were: (i) to identify road features and traffic flow parameters which contribute to the causation of crossover accidents and to quantify their effects; (ii) to establish on the basis of likely accident cost savings the conditions under which safety fence installations would be cost effective. The feasibility phase of the study was concerned with reviewing the literature and available accident statistics and developing suitable methodologies for a follow-up study. In this phase accident data for some 700 km of dual carriageway were analysed and geometric data for 64 km of such roads collected. Accident data collected before and after barrier installation gives an indication of the decrease in number of casualties from median related accidents. Simple regression analysis shows that some casualty rates are significantly related to flow parameters. Further data will be analysed by more sophisticated multiple regression techniques in the main study, and before and after studies of all recently fenced sections will be made. (A) This is paper 4 of the TRRL papers presented to the 1986 TRB annual meeting.
Samenvatting