Passive safety treatments were found to be more reliable and affordable than active safety treatments on two-lane rural roads with dangerous horizontal curves in Virginia.
A Virginia DOT study monitored nine curves along Virginia's Route 615 corridor and found that active safety treatments gave mixed results.
Made Public Date

Evaluation of Innovative Approaches to Curve Delineation for Two-Lane Rural Roads

Summary Information

Crash rates at curves along highways are three times those of straight segments, but many suggested treatments to reduce the risk of running off-road can be prohibitively expensive. The Virginia Department of Transportation (VDOT) has launched an effort to identify areas with critical-level roadway-departure crashes, and to analyze the most effective, appropriate safety treatment. Both passive systems--those relying on external light sources such as daylight or vehicle headlights for illumination--and active systems--those including internally lit lighting elements--are being considered.

This study, conducted by a team of researchers front the Virginia Transportation Research Council on behalf of VDOT, sought to assess the effectiveness of active and passive curve warning and delineation systems on two-lane rural roads. A total of six treatments, three passive and three active, were installed across nine curves along the Route 615 corridor.

Data was collected both by an observational study and by a human-factors study. In the observational study, researchers used pneumatic road tubes to measure the speed and lane position of traffic at three points along each curve. In the human factors study, researchers had participants drive experimental vehicles on a test route while having their speed and lane position tracked along each curve. In both cases, data were collected both before and after the new treatments were installed.

Existing safety treatments included curve warning signs, winding road signs, and passive chevrons. The added treatments varied by location, but included active flashing chevrons, retroreflective posts, on-pavement signage, guardrail reflectors, and dynamic curve warning signs. The dynamic sign used radar to measure the speed of oncoming vehicles and displayed a "SLOW DOWN" message when it detected vehicles exceeding the advisory speed.


In general, the human-factors study found more modest impacts from the new treatments than the observational study. The authors suggest that this is because participants were measured along the entire route, rather than at single intersections as with the observational study. Additionally, the presence of experimenters within the test vehicles may have had a notable impact on participants' driving.


  • The addition of retroreflective material to signposts was found to significantly reduce traffic speeds for the outer lanes of curves, reducing speed by 2.81 mi/h in the observational study. In the human factors study, this impact was instead statistically significant for the inner lane only.
  • The use of on-pavement signage was found in the observational study to significantly decrease the absolute offset, the distance from the center of the lane. This impact was greater than 100 mm on average at all locations the signage was installed at and applied to both inner and outer lanes.
  • The installation of continuous reflectors was found in the observational study to actually increase the average vehicle speed by over 2 mi/h. However, it simultaneously decreased absolute offset by an average of 400 mm across inner and outer lanes.
  • Active treatments were typically found to actually be less effective in general than passive treatments. The dynamic curve warning sign was found to decrease vehicle speed by both the human-factors and observational studies in one location, but to increase speeds at the other location it was installed. This implies that its effectiveness may be dependent on environmental conditions and context.
  • The active blinking curve warning sign was found by the observational study to both increase vehicle speeds, by over 3.5 mi/h, and increase average lane offset, by almost 230 mm.
  • Finally, the sequential chevrons were found by the observational study to increase average vehicle speeds for outer lanes only, raising them by 2.29 mi/h.

The authors suggested that some of the variability in the results may be explained by environmental factors. For instance, increased visibility on a wide curve was less likely to lead to higher vehicle speeds than increased visibility on a sharp curve. Additionally, the results may be limited by the relatively small range of time over which results were collected. Finally, they suggest that drivers may have noticed the new dynamic message sign, and purposefully driven faster to activate it out of a sense of novelty.


Goal Areas
Results Type
Deployment Locations