This study was undertaken to improve the safety of highway construction zones in new mexico. The authors examined construction-zone accidents in new mexico for a 3-year period. In contrast to the traditional technique, which relies solely on the accident record system, they utilized the locations and durations of construction activityon rural state highways. The record system was then used to identify accidents at these sites during the period of construction as wellas during the identical period in the previous year. In coparison with the prior year, accident experience increased by 26% during construction. Contingency tables were used to compare the driver, roadway, and environmental characteristics of accidents before and duringconstruction. At the 5% level of significance, the only parameter that differed between the two periods was the road surface condition, which was dry more often for accidents during construction. However, there was a moderate overrepresentation of accidents involving multiple vehicles, rear-end collisions, large trucks, and the contributing factors of following too close and improper lane changing. The authors also identified deficiencies in the accident record system that result in a substantial understatement of crash experience in construction zones. This paper appears in transportation research record no. 1230, Work-zone traffic control and tests of delineation material.
Abstract