The report reviews research results relating accident frequency to cross-sectional elements of urban arterial roads. Within the range of practical lane widths, lane width itself has only a marginal effect on accident rates. The density of access to the arterial from unsignalised side streets and driveways can have a greater effect on accident rates than the cross-section treatment. Cross-section treatments which provide space for turning vehicles to stand clear of through traffic result in accident rate reductions of 20 to 50 percent. The USA literature favours continuous two-way turn lanes to achieve this, whereas Australian practice favours raised medians and protected turn lanes. The comparative safety performance of divided urban arterial roads in Australia is considerably better than that pertaining to the divided arterials considered in the USA research. This implies that the accident prediction relations developed in the USA, and the related findings regarding two-way turn lanes, are not directly applicable to Australian conditions. (A)
Abstract