Understanding traffic bottlenecks of long freeway tunnels based on a novel location-dependent lighting-related car-following model

Abstract

To understand the formation of lighting-related traffic bottlenecks along the long freeway tunnels in the daytime, this study develops an intelligent driver model incorporating the location-dependent lighting-related desired speed (IDM-LLDS). This car-following model associates desired speed variations with the lighting-transition-incurred visual adaptation in the portal areas and low illumination in the interior zone of the tunnel, such that how the impact of location-heterogeneous lighting conditions on driver’s car-following behavior propagates through car platoons is described. The model is validated, and numerical examples are presented to illustrate the traffic patterns related to tunnel lighting and daylight. The results demonstrate that periodic oscillations in traffic patterns are presented under visual adaptation occurrence if traffic flow density is not significantly high. Adaptive tunnel lighting and implementation of sunscreens at tunnel portals can both improve the traffic conditions in strong daylight. The former performs better in improving traffic efficiency, while the latter excels in smoothing visual-adaptation-based traffic oscillation. With the objective of minimizing the magnitude of traffic oscillation, the tunnel lighting strategies adaptive to daylight are recommended under different traffic volumes and travel speeds where majority of cars can avoid obvious trajectory oscillation traversing the tunnel.