Virginia I-95 Study Found Variable Speed Limit System with 48 Signs Produced $2.57 Million Annual Benefit by Reducing Severe Crashes.
Preliminary Results Showed that Drivers Responded to Variable Speed Limits on I-95 Between Washington, DC, and Richmond, Virginia.
Fredericksburg, Virginia
Operational and Safety Effects of the I–95 Variable Speed Limit System in Fredericksburg, Virginia.
Summary Information
Variable speed limit (VSL) systems, which set speed limits dynamically as a function of traffic or roadway conditions, can improve traffic flow and safety. VSL systems are typically employed to provide better guidance on safe speeds during adverse weather or in work zones or mitigate congestion by encouraging more uniform speeds. In June 2022, the Virginia Department of Transportation (VDOT) activated a VSL system on I–95 northbound from milepost 115 to 130 just south of Fredericksburg. I–95 has three lanes in the northbound direction that carried a directional annual average daily traffic of between 50,000 and 57,000 vehicles in 2022. This corridor experiences significant delays, unreliable travel times, and numerous congestion-related crashes and incidents, particularly on summer weekends when recreational traffic volumes increase. The goal of the VSL system was to generate smoother traffic flow and consistent travel speeds along the corridor.
METHODOLOGY
The I–95 VSL system consisted of 48 LED VSL signs (24 pairs) with an average spacing of approximately 0.6 miles. VDOT activated the system on June 22, 2022. It used an algorithm, operating in both predictive and reactive modes, to set speed limits for the corridor. The algorithm parameters were updated in February 2023 and again in November 2023. The VSL evaluation was designed as a before-after analysis of system performance. “Before” data were from January through March 2022, and “after” data were from June 2022 through November 2023. The analysis included the following metrics:
- Benefit-cost analysis. Benefits were calculated using savings from reductions in cost of user delays and crashes.
- Average speed and speed limit compliance. Descriptive statistics of the 30-second average vehicle speeds were calculated, as was the cumulative percentage of vehicles traveling above or below the posted limit.
- Travel time and reliability. Probe travel time data were used to examine average travel time and a planning time index.
- Throughput. A 5-minute aggregation interval was used to smooth fluctuations in flow and converted to an equivalent hourly flow rate.
- Safety. A preliminary safety assessment was performed using number of crashes, crash severity, and crash type.
FINDINGS
- One year of crash data before and after deployment showed total crashes to be relatively flat and about a 13 percent reduction in fatal and injury crashes.
- The system produced an estimated annual benefit of approximately $2.57 million (planning level estimate). This benefit was attained primarily from reductions in severe crashes and did not consider long-term annual operational and maintenance costs which could be substantial.
- When posted speed limits were 45 or 55 mph, the system reduced average speeds by 4 or 5 mph and speed differentials between adjacent lanes by about 0.5 mph.
- Average speeds were relatively similar between the before and after periods during uncongested conditions (posted limits over 60 mph) and during congested conditions (posted limits of 35 mph).
- Travel times generally improved on weekends (a 4.7 percent decrease), but these improvements were largely offset by degradations during the week (a 4.3 percent increase).
- No significant changes in maximum throughput were observed.
