Exploring Traffic Speed Patterns for the Implementation of Variable Speed Limit (VSL) Signs

Traffic congestion, bottlenecks, queuing of vehicles, and resulting shockwaves are commonly observed phenomena during peak hours of travel on many roads across the United States. Researchers and practitioners are exploring sustainable and economic approaches to mitigate congestion and associated effects by examining historical traffic patterns and exploring data-driven strategies.

Intelligent transportation system (ITS)-based solutions have the potential and are being widely explored to address congestion-associated transportation problems. The application of advanced traveler information systems (ATIS) such as VSL signs is one such ITS solution. They are being strategically implemented to improve mobility during extreme weather conditions, in workzones, and at locations with special events.

VSL signs promote speed harmonization of the corresponding road link or section. Existing VSL systems mostly use simulated algorithms to generate the speeds needed for the corresponding time of the day and day of the week. Selecting an algorithm or technique to improve the flow is one of the most common challenges due to the dynamic nature of traffic conditions. It requires a prior understanding and mimicking of the existing traffic speed patterns for various temporal epochs across the road. Some of the simplest algorithms used include the display of speeds in increments of 5 mph based on the 85th percentile speeds.

The traffic volumes and incidents during peak and off-peak hours are spatially dependent and result in speed variability. The level of variability depends on the time of the day, day of the week, and road functional class. While harmonizing speeds for the roads with speed variability promotes better mobility, safer traffic, and improved operational performance, it is not an effective solution for all road links or sections. Hence, it is important to research suitable traffic speed measures to identify the road links which are susceptible to higher variability of traffic speeds to mitigate congestion and improve mobility. As simulation-based analysis has limitations in accounting for such variations, using real-world data and data-driven approach is more suitable.