For that barrier, they can thank big data.
“Even though in many parts of [I-80] we have a very wide median, we found there was a need for additional shoulder protection, because of the types of crashes we were still having,” explained Robert Miles, director of traffic and safety at the Utah Department of Transportation.
Utah DOT engineers looked to crash data from a number of sources to gain better insight into the different dynamics affecting traffic along this 200-mile stretch of highway. They used data analysis tools like Numetric, which organize and sort vast amounts of information to give engineers and safety officials a better understanding of traffic conditions, highway design features or other variables that could influence driver behaviors.
That level of analysis happens all across the state, as UDOT prioritizes safety projects.
“We help DOTs be more efficient and make more data-driven decisions and improve their state’s roadways, from a safety perspective,” said Numetric CEO Nate Bowler.
Some of the data collected and organized by Numetric include mile-markers and locations, as well as the location of objects such as signage or striping.
“And when you take all of that information, and combine it, you can start to do some very interesting analysis, and figure out, ‘where are my hot spots in my roadway network?’” said Bowler.
On I-80, UDOT officials were able to identify areas where the median wasn’t adequate, said Miles.
“The space was masking problems that were happening, that weren’t evident unless you really got into the crash records,” he added.
Miles is careful to note that the department uses other data-driven analysis tools besides Numetric — which itself is a cloud-based service, accessed and searched via a Web browser — and data analysis has been part of the agency’s decision-making for some time. The department also relies on the United States Road Assessment Program, otherwise known as usRAP, a program offered by the Roadway Safety Foundation.
“Numetric is one of the tools that we have in our toolbox,” he explained. “And it’s a good tool for allowing us to visualize what has happened in the past.
“We’ve used data for a long time. We’ve captured the information on the crash records, even back when we were doing it on paper copies, and put them into different data bases. But the change lately has been the ability to marry those data bases, and the popularity of the data science field, the emergence of the data science field, as a tangible thing for a practicing engineer. I would say that’s really taken off in the last six to eight years.”
Too often, in the past, highway safety projects were identified largely through “anecdotal” feedback, where a department director merely steered projects in a way that he or she believed appropriate, said Miles.
“It was what they were hearing most about, or what was foremost in their consciousness,” Miles said. “Each director would come in and say, ‘Hey, we’ve got to pick our safety projects. What do we have? What are our biggest problems?’
“So it was really anecdotal. And these new projects allow us to move out of that anecdotal into more of a fact-based decision-making process, because you’ve got something to measure things on equal ground."
It’s not just offering more detailed and searchable data that technology firms like Numetric bring to transportation departments. The job often includes cleaning up the data so that it’s easier to access and decipher, said Bowler.
“You may have multiple names for roads. You may have incorrect or inconsistent data around latitude-longitude … and so we kind of normalize those things,” he explained.
“Every state we work with has an issue with data, that either requires some cleaning, or that’s siloed and kind of separated from other data, managed by different groups or departments, but need to have a third party like us to bring together."
Numetric has worked with transportation departments in New Jersey, South Carolina and New York, among others.
Getting traffic and crash related data in front of more eyes in an effort to make better data-driven solutions is a guiding philosophy at the Utah DOT and many transit and transportation departments.
“The more people we have looking at those crash records, those safety records, the more likely you are to find different solutions that we haven’t thought of before," said Miles. "And so, one of our mantras here is, we want to get safety data as close to the hands of the designers as possible, as close to the problem-solvers as possible. We want lots of opinions, and we want those experts to provide depth and meaning. … We want people to feel like they own the safety on their projects.”
