- BYU professors created a statistical model that can predict crashes expected on roadways
- UDOT is using the model to prioritize safety projects with the highest impact for saving lives
- The model is one of the most advanced traffic safety models in the country
Three BYU researchers are behind the wheel of a new effort to help the Utah Department of Transportation come closer to the goal of zero fatalities on Utah highways.
Led by civil engineering professor Grant Schultz, the team has developed one of the country’s most advanced traffic safety models. Using Bayesian statistics, the model can predict the number of crashes expected on Utah’s roadways to help identify areas of concern.
The tool gives UDOT a new set of keys to prioritize funding for improvements to the highest-risk roadway segments. Those improvements are aimed at cutting down on the frequency and severity of crashes and, ultimately, keeping people safe on the road.
“Our goal here is to save lives,” Schultz said. “We’re just trying to do all that we can to truly reach the goal of zero fatalities on the roadways.”
Combined with a geographic information system, the model pinpoints hotspots where the number of crashes exceeds the number of predicted crashes. With that information, UDOT can advise policy makers on which safety projects have the greatest potential to reduce serious injury and fatal crashes.
Coincidentally, one of the first hotspots identified by the model is next to BYU’s campus: the intersection of University Avenue (U.S. Route 189) and Bulldog Boulevard. The model revealed a greater-than-usual number of accidents from left-hand turns, so UDOT decided to install a new signal system there this summer.
“Whenever you need to do taxpayer-funded improvements, you have budget limitations,” said co-researcher Mitsuru Saito, a BYU professor of civil engineering. “This system will help decision makers spend the funds where they get the highest benefit to reduce the number of crashes."
Schultz, Saito and statistics professor Shane Reese have researched the issue for the past five years, producing a number of publications along the way. Most recently the group published findings in the Transportation Research Record:Journal of the Transportation Research Board.
UDOT and the Utah Department of Public Safety have been so pleased with the early results from the Utah Crash Prediction Model that they recently awarded the trio of professors the Executive Director’s Excellence in Transportation Safety Award.
“We asked BYU to build a statistical model for Utah knowing it was quite an ambitious challenge,” said Scott Jones, UDOT safety programs engineer. “It took a couple of years of refining, but now it’s at the point where the model is pinpointing where to do safety projects and we can pull the trigger on those projects.”
Since 2000, fatalities across the state have dropped 41 percent. With the development of the Utah Crash Prediction Model to aid UDOT, Schultz, Saito and Reese anticipate this number will continue to improve toward the goal of zero fatalities.
But the researchers, along with the hundreds of folks in UDOT and the Department of Public Safety, can’t reduce fatalities by themselves.
“We need the public’s help to make it a priority in how they drive and what they do when they drive,” Schultz said. “We can model and look for hotspots all over the state, but we need to have drivers do their part to make this successful.”