How Far from Optimal are Current Advisory Speeds?

Abstract

The posting of advisory speed signs at sharp horizontal curves to provide the driving public with a safe speed is a practice well established in the United States. However, the operational effectiveness of these signs has long been questioned in the literature. The authors of this paper recently developed a function to model the expected safety effect of these signs. The function evolved from a statistical analysis of crash data from two-lane rural highways in the state of Oregon. In general, that initial research effort revealed that advisory speed signs tended to enhance safety. However, the results also suggested that advisory speed signs may not display the value with the greatest potential safety benefit. Because the function derived from that research proved meaningful from the engineering and human factors perspectives, the authors then extended the use of that function to compute and recommend the theoretically optimal advisory speed. A new posting procedure resulted from that effort. The authors compared the expected performance of advisory speeds from the proposed procedure with the speeds derived from current posting guidelines. The comparable performance of the proposed procedure suggests that current guidelines are close to the hypothetically optimal advisory speed. In general, speeds determined by both the current and the new computational methods performed better than those determined by the ball bank indicator method. This paper also presents a field validation analysis of the engine function of the new posting method. The results confirmed the meaningfulness of the function and therefore the potential benefit of the proposed method for determining safety-based advisory speeds.