A Statistical Method for Assigning Priorities to Road Sections for Resurfacing

We describe a method of ranking road sections in order of priority for resurfacing. The ranking is according to estimates of the expected reduction in the number of wet-pavement skid accidents after resurfacing. The estimation is based on the assumption that resurfacing does not affect the expected number of dry-pavement accidents but reduces the expected number of wet-pavement accidents to a fixed proportion of the expected total number of accidents. Some properties of the procedure are examined. The validity of the assumptions and the performance of the procedure are investigated using historic data.