Classification analysis of driver's stop/go decision and red-light running violation.

When the driver encounters a signal change from green to yellow, he is required to make a stop or go decision based on his speed and the distance to the stop bar. Making the wrong decision will lead to a red-light running violation or an abrupt stop at the intersection. In this study, a field data collection was conducted at a high-speed signalized intersection, where a video-based system with three cameras was used to record the drivers' behavior related to the onset of yellow. Observed data include drivers' stop/go decisions, red-light running violation, lane position in the highway, positions (leading/following) in the traffic flow, vehicle type, and vehicles' yellow-onset speeds and distances from the intersection. Further, classification tree models were applied to analyze how the probabilities of a stop or go decision and of red-light running are associated with the traffic parameters. The data analysis indicated that vehicle's distance from the intersection at the onset of yellow, operating speed, and position in the traffic flow are the most important predictors for both the stop/go decision and red-light running violation. This study illustrates that the tree models are helpful to recognize and predict how drivers make stop/go decisions and partake in red-light running violations corresponding to the traffic parameters.