Probabilistic Design of Freeway Entrance Speed-Change Lanes Considering Acceleration and Gap Acceptance Behavior

An adequate length of entrance speed-change lanes (SCLs) is required for vehicles' acceleration and gap-searching purposes so that vehicles can merge onto the freeway comfortably. The current design guides use a deterministic approach for the design of SCL length based only on the acceleration behavior of vehicles in the SCL. This study introduces a probabilistic approach for the design of the length of SCLs that considers both acceleration and gap acceptance behavior of drivers during the merging process. A microscopic simulation, coupled with a Monte Carlo simulation technique, is used to develop a probabilistic model to evaluate the probability of a forced merge by vehicles in the SCL, termed the probability of noncompliance (PNC). Reliability measures can be developed on the basis of the distribution of PNC values for all simulated SCL vehicles at a specific site. Several such measures are estimated for seven study sites and are shown to have high potential to indicate the safety performance at entrance SCL sites. As an example application of the developed model, the mean PNC is estimated to the SCL lengths recommended in North American design guides for a specific freeway design speed but different values of controlling curve design speed and traffic volume in the freeway right lane. The results indicate that the mean PNC may change with the change of either the design speed of the ramp controlling curve or the traffic volume on the freeway right lane.