Abstract:
Expected left-turn gap acceptance (EGA) refers to the average time headway between two successive vehicles in the opposite direction traffic stream that is accepted by drivers who must make unprotected left turns at unsignalized intersections. The EGA value decreases as the waiting time becomes longer. As a result, drivers tend to take a risk to accept a much short gap in such situations. The factors of drivers individual differences (e.g., age and sex), environment (e.g., urban or rural location, darkness and weather), perceptual and behavioral contributors (e.g., misperception of vehicle speed), and intersection design become critical when the gap is shorter than they used to accept, and thus left-turn related crashes are more likely to occur. It is very important to better understand the relationship between EGA and waiting time in the field of traffic safety. The objective of this research is to examine the relationship between EGA and waiting time and provide the decision support in selecting gap acceptance thresholds in warning criteria or converting unsignalized intersections into signalized intersections. A logistic probability regression model was developed to investigate drivers’ gap acceptance behaviors. A theoretical model was established to represent the relationship between EGA and driving behavior under ideal conditions. With the field data and the logistic probability regression model, the paper estimated the EGA values for ideal waiting time, normal waiting time, long waiting time and over-long waiting time. The results suggest that the longer they wait the shorter gap they most likely accept.
