Abstract
Over the years the capacity drop phenomenon at freeway bottlenecks has remained a topic of interest and intrigue. Capacity drop has an undeniable impact on freeway performances, directly affecting the throughput. Various studies have tried to measure, predict, and model capacity drop. In this paper, the authors integrate a simplistic lane changing model and an acceleration model together into the Cell Transmission Model framework. The capacity drop is modeled as a combined effect of a lane changing area upstream of the bottleneck location and an acceleration region downstream. Each lane changing vehicle is considered to contribute towards density on two lanes during the lane changing maneuver as it effects the following vehicles on both the original and target lanes. This effect is modeled through the introduction of a â??perceived densityâ?? variable. This perceived density is obtained by scaling the actual density up by the lane-changing intensity, and is used to determine the demand at the bottleneck. A demand function linearly decreasing in density under over-saturated conditions is used to model the acceleration process of vehicles as they discharge from the bottleneck. It is shown that the capacity drop can then be predicted from calibrated demand-supply functions and lane changing intensity.