Variable speed limit systems reduced the number and severity of crashes at three pilot sites in Texas. Benefit-to-Cost ratios ranged from 7:1 to 14:1.
An evaluation of variable speed limit (VSL) systems under a variety of operational conditions.
Made Public Date


WB 1604
San Antonio
United States


NB IH-35
United States


Ranger Hill
United States

Evaluation of TxDOT Variable Speed Limit Pilot Projects: Final Report

Summary Information

In 2014, acting on a ruling by the Texas Transportation Commission, a variable speed limit (VSL) pilot program was implemented by TxDOT and the Texas A&M Transportation Institute (TTI) was asked to study the effectiveness of temporarily lowering speed limits to address inclement weather, congestion, road construction, and other factors that can affect the safe and orderly movement of freeway traffic.


VSL systems implemented in the summer 2014 were evaluated at the following locations:

  • Urban Congestion Site: San Antonio, WB SL604 from US 281 to IH-10. Data collected from June 30, 2014 – December 31, 2014
  • Construction Work Zone Site: Temple, NB IH-35 NB from MM 297 to MM 301. Data collected from June 23, 2014 – November 30, 2014
  • Weather-Related Site: Ranger Hill, Eastland County, EB and WB on IH-20 (proximal locations). Data collected from July 21, 2014 – January 30, 2015.

For the TxDOT pilot deployments, a 5-mi section of freeway upstream of a known bottleneck was used to demonstrate the concept. The section of freeway was divided into segments, generally between 0.5 and 1 mi in length. Each freeway segment represented a zone over which speed limits were varied approaching the bottleneck location. A variable speed limit sign was used to establish the recommended speed limit for travelers along the segment of freeway. Traffic sensors that measured speeds, volumes, and occupancies across all lanes were used to establish the speed limit to be displayed within the segment. The speeds were then displayed dynamically with VSL signs also associated with each segment.

In all cases, the VSL system issued an alert to the operator when a VSL needed to be initiated. The operator then had to confirm the alert to begin operating the signs in a VSL mode. Note that not all requests for activations were approved by operators. Once the signs were activated, the displayed speeds were updated every minute automatically without operator intervention.

The project team examined the impact of the VSL deployment on average speeds at each of the deployment locations. Speed data from 4:00 PM to 6:15 PM were used in the analysis, as this time period represented the period when the VSL systems were most active. The analysis also only examined data from Tuesdays and Thursdays, as these days were deemed by the TTI team to best represent typical weekday traffic.

TxDOT’s Crash Records Information System (CRIS) was used to analyze the safety of the study sites. Approximately six months within the three years of Before data (2011, 2012, and 2013) and one year of After data (2014) were used in the analysis. Data included in the analysis contained the length of the deployment site, cross-section of the facility, free-flow speed on the facility, and historical crash data. In all instances, the analysis assumed the speed harmonization installation was a full, permanent installation on the selected facility.


The table below excerpted from the source report shows the potential B/C ratios. The benefits predicted in these deployments were based on a seven percent reduction in crashes estimated from the safety analysis.

Deployment Site
Annual Benefits
Annual Costs
Net Benefit
B/C Ratio
San Antonio (WB SL1604)
Temple (NB IH-35)
Ranger Hill (EB and WB IH-20)

The project team conducted the analysis using a tool developed by Federal Highway Administration (FHWA) for planning related analyses of operational strategies. Each individual deployment was analyzed separately assuming the VSL installation would be a full permanent installation. Included in the analysis was the estimated and projected value of capital equipment (basic infrastructure equipment and incremental deployment equipment) and operational and maintenance costs annualized over a 20-25 year useful life of the equipment.

The authors noted that significance testing was not done due to the short duration of each project. The results were high-level estimates.
Goal Areas
Deployment Locations