Roadway departure crashes make up a significant number of the crashes on Texas roadways each year. In 2016 alone, Texas roads experienced 61,973 roadway departure crashes, most of which occurred on two-way two-lane (TWTL) highways (92 percent). These numbers — especially in a state that just saw its second-deadliest year on record for road fatalities (2021) — prompt the need to effectively identify and deploy countermeasures that can prevent future crashes.
The Texas A&M Transportation Institute (TTI) conducted the research project Evaluation of Roadside Treatments to Mitigate Roadway Departure Crashes. Sponsored by the Texas Department of Transportation (TxDOT), Project 0-6991 was a multi-year project led by researchers at TTI.

Researchers focused on rural TWTL highways and examined roadside features to recommend appropriate countermeasures for systemic implementation in Texas. These recommendations are intended to serve as a framework for TxDOT districts to use when prioritizing sites that are at risk for a roadway departure crash. By prioritizing sites that are most at risk, districts can make safety improvements proactively (e.g., improving guardrails and barriers) rather than reactively (e.g., based on crash history).
“When a roadway departure crash happens, it’s a chain of events that typically starts with things that we as engineers can’t control, like a distraction or a mechanical malfunction,” says Raul Avelar, TTI research scientist and lead researcher on the project. “When a crash happens, everything in the past that you did on the road to make it safe is not useful anymore. This means you must go to plan B and have the roadside protected.”
Crashes in rural areas are significantly affected by the random nature of crash occurrences. This is more prevalent in crash types such as rollovers, guardrail hit crashes and other fixed-object collisions. Using a systemic approach that focuses on high-risk roadway features rather than the crash record at specific locations, it is possible to anticipate which locations are likely to experience crashes based on their roadway characteristics known to be associated with higher crash risk.
After identifying 420 roadway segments across the state using a balanced stratified sample, researchers examined key safety variables including average daily traffic, average daily truck traffic, shoulder width, lane width and speed limit at each segment. These factors were then weighted based on site and traffic characteristics to help identify at-risk areas.
According to Avelar, roadway departure crashes can be prevented by:
- designing roadways with geometric parameters that make it less likely a driver will depart from the lane (e.g., wider lanes and softer horizontal curves);
- giving feedback to a driver who is departing from the lane (e.g., rumble strips and profile edge markings);
- widening the length of the shoulder to increase the likelihood of a driver returning to the roadway; and
- using more forgiving roadside designs, like providing flatter foreslopes, wider clear zones, and guardrails protecting against hazardous conditions such as walls, trees and poles.
“A good research project always leads to more research,” says Rebecca Wells, director of transportation operations for the TxDOT Atlanta District. “For the positive results that came out of this project, we double down on what works. If we know texture works on shoulders, then we continue to find ways to do that — whether it’s through construction or safety funds — to ensure that we’re implementing the things that we know are working well.”