Validation of probabilistic model for design of freeway entrance speed change lanes
Traffic operation on entrance speed change lanes (SCLs) is more complicated than on exit SCLs as vehicles interact constantly while merging from the SCL to the freeway. An adequate SCL length gives road users sufficient time to accelerate and to search for a gap so that they can comfortably merge onto the freeway. This study focused on analyzing entrance SCLs from a safety point of view in a probabilistic framework. Sixteen entrance-limited-type SCLs along Highway 417 were considered to investigate the link between the probabilistic measure, probability of noncompliance (P), and collision frequency. Regression analysis was conducted to quantify the effect of the geometry of the SCL, driver merging behavior along the SCL, road users' exposure, and P on collision occurrence. The model containing P outperformed other models, and the inclusion of P in the regression models resulted in significant improvement in the model fit. The study provided positive evidence on the validity of P as a surrogate measure of safety.