This document summarizes the research conducted and the findings from a two-year investigation of driver behavior on horizontal curves on rural two-lane highways. This research included a review of existing procedures for setting curve advisory speed, the collection of speed data on horizontal curves, and the calibration of speed prediction models. The findings from the research were used to develop criteria for setting the curve advisory speed, a method for determining this speed, and guidelines for identifying appropriate curve-related traffic control devices. A conclusion of this research is that there are some challenges associated with the use of the ball-bank indicator that make difficult the task of achieving curve advisory speeds that are uniform among curves and consistent with driver expectation. A method for establishing advisory speeds is described that overcomes these challenges. The method is based on the measurement of curve geometry and the use of these measurements to estimate the average curve speed of truck traffic. The average truck speed is then used as the basis for establishing the advisory speed.
Abstract