The effects of paved shoulders on accidents on rural highways.

This paper presents the results of a project that sought to determine the safety effect of paving shoulders in rural roads Victoria, Australia. Data were obtained on the location, condition and cost of recent shoulder-paving projects on two-lane two-way roads (i.e. one lane in each direction). Accident data were obtained for these sites, and a before-and-after comparison, using control sites, was undertaken. Most shoulder-paving programs examined involved a low-cost paving of an existing shoulder, typically involving an interim bituminous sealing treatment, followed a year or so later with a reseal in conjunction with a pavement reseal. The shoulder width is typically between 600 and 1200 mm, with 600 or 800 mm being the most common. The results for this type of treatment indicate that shoulder paving was associated with a statistically-significant reduction in casualty accident frequencies at sites where it was installed on two-lane two-way rural highways in Victoria. Overall, casualty accidents were reduced by 41 per cent on a per vehicle kilometre basis at such sites, which is equivalent to a reduction of 0.071 casualty accidents per million vehicle kilometres. The break-even point (the point at which it is economically worthwhile to pave shoulders) is at a traffic flow of about 360 vehicles per day. The main accident reductions are for accidents involving rear end, overtaking--out of control, off carriageway to left, and off carriageway to right into fixed object. A relationship between measures of economic worth (net present value and benefit/cost ratio) and traffic flow is developed.