Lane Group–Based Traffic Model for Assessing On-Ramp Traffic Impact

Abstract

On-ramp merging areas are congestion-prone segments of freeways. Depending on the aggressiveness of the driving population and the congestion level, the speed variance among travel lanes due to lane changes and ramp-merging flows may be so significant as to affect the optimal settings of deployed traffic control systems, such as metering rates or advisory speed limits. Extending from METANET, this study presents a lane group–based (LGB) traffic model to reflect the temporal and spatial distributions of traffic conditions among lane groups. The proposed model would allow traffic engineers to reliably assess the impacts of lane-changing activities in both upstream and downstream segments of an on-ramp area and better design their coordinated control strategies. To assess the effectiveness of the proposed model, this study has compared its performance with METANET under various traffic scenarios. The comparison results show that the proposed model can yield up to 26.9% improvement on the accuracy of predicting the temporal and spatial evolution of a freeway’s speed at the interchange area where freeway segments often experience extensive lane-changing activities due to on-ramp merging flows.