Highway bottlenecks caused by traffic incidents, lane drop, ramp merging or slow vehicles negatively impact traffic mobility and safety. Traffic regulation techniques, such as adjusting speed limits, providing lane-change recommendations, and restricting on-ramp vehicle inputs with traffic signals in response to an incident, have been found to mitigate congestion from these types of bottlenecks. However, most existing research assumes that traffic models and measured traffic data are accurate and that vehicle drivers always comply with recommendations from the infrastructure. These assumptions are rarely true in the real world and can lead to inconsistencies between the theoretical benefits and the actual benefits obtained in field tests.
Researchers at the University of Southern California developed, analyzed, and evaluated an innovative approach to alleviate highway bottleneck congestion. The approach includes issuing variable-speed advisories and lane-change recommendations when needed to the upstream vehicles, as well as ramp control to manage incoming traffic, while accounting for inaccuracies in traffic data and road information and the complex behavior of human driving. This policy brief summarizes the key findings from that research.