This study concerns the relative safety of the various types of median design, including the positive barrier median, on divided highways carrying traffic volumes in excess of 15,000 vehicles per day, and the development of tentative criteria for the installation of positive median barriers. A report covering a previous median study of divided highways which carried volumes up to 25,000 vehicles per day was presented at the hrb thirty-second annual meeting. An analysis was made of the approximately 8,000 accidents which occurred in 1956 and 1957 on some 265 mi of divided highway with deterring and non-traversable median designs. Operating conditions, as measured by the average daily traffic (adt) volume, apparently influenced the relative safety of the deterring and non-traversable medians. In the volume range of up to 130,000 vehicles per day, the deterring-type median had the lower accident and injury rate. In the volume range of 130,000 or more vehicles per day, the advantage shifted to the non-traversable medians which had the lower accident and injury rate. To emphasize the cross-median fatal head-on-type accident, the 407 fatal accidents which occurred on freeways in 1956, 1957 and 1958 were then analyzed. During this period, the cross- median collisions accounted for 19 percent of the fatalities on freeways. Freeways carrying more than 60,000 vehicles per day accounted for one-fifth the mileage and two-thirds of the fatal cross-median collisions. Therefore, in order to make a significant attack on the cross-median fatal accident problem, it would be necessary to reach down to the 60,000 adt level with the installation of median barriers. Past experience indicates that barriers may convert cross-median accidents to other types. However, newly-developed barrier designs may reduce the severity of collisions with the barriers and results in fewer casualties even though the accident rate may rise. /author/.
Abstract