Education

Snow day decisions? Could be an app for that

Researchers envision role for artificial intelligence

Kris Hartgrave, director of transportation for College Community School District, early last Monday scouts rural roads south of Cedar Rapids. Hartgrave is responsible for helping determine whether roads are safe enough for cars and school buses to travel as the district weighs whether to call for a late start or cancellation. Researchers at Iowa State University are working an artificial intelligence technology they hope will lead to better predictions about road conditions and lessen the need for school officials to risk driving on rural roads to see for themselves. (Rebecca F. Miller/The Gazette)
Kris Hartgrave, director of transportation for College Community School District, early last Monday scouts rural roads south of Cedar Rapids. Hartgrave is responsible for helping determine whether roads are safe enough for cars and school buses to travel as the district weighs whether to call for a late start or cancellation. Researchers at Iowa State University are working an artificial intelligence technology they hope will lead to better predictions about road conditions and lessen the need for school officials to risk driving on rural roads to see for themselves. (Rebecca F. Miller/The Gazette)
/

CEDAR RAPIDS — At 4 a.m. last Monday, hours before most commuters had rolled out of bed to face a blustery day, Kris Hartgrave was on the road.

With a silver coffee thermos by her side, Hartgrave, director of transportation for College Community School District, maneuvered her minivan through the night’s snowfall. Wind had caused drifting, leaving white dunes creeping across the district’s rural roads. Plow trucks were out, but many roads had yet to be cleared of snow.

“I see a lot of drifting snow, which for us will be problematic,” Hartgrave said as the van pushed through another mound of snow.

She was one of many school officials — from districts across the state — out on the roads that early morning to get a firsthand look and feel of road conditions to help decide whether to hold classes that day.

“It’s all about safety,” she said. “I don’t ever want to have a bus in the ditch. I can’t imagine what that would be like and I don’t want to know.”

Research taking place at Iowa State University eventually could lessen the need for school officials like Hartgrave to be out so early on treacherous rural roads — and provide another tool for school superintendents to use when deciding whether to cancel classes on a winter day.

Researchers are working to combine state weather data with artificial intelligence technology to allow officials to predict — with a level of accuracy — what transportation conditions across a school district will be like when the buses need to roll.

Making the call

Doug Wheeler, superintendent for the College Community School District, also was out at 4 a.m. that morning checking road conditions. He said delays or cancellations are called for student safety, but many other factors are weighed, too, including the interruption it can have on education and hardships it can cause parents who will need to find child care.

ARTICLE CONTINUES BELOW ADVERTISEMENT

In addition, for children on free- or reduced-price lunches, missing school also could mean missing a meal.

“Looking at student safety first and looking at all those other factors is what makes it a difficult decision. It’s pretty clear a lot of times what the right decision is, (but) it doesn’t make it any less difficult,” Wheeler said.

That Monday, College Community classes started two hours late, with buses traveling only on hard-surface roads. That meant parents on gravel roads were responsible for taking their children to a paved surface to get on a school bus.

Classes at most districts were canceled for days last week because of snow, ice and bitter cold.

“This has been a really interesting stretch, in terms of the weather,” said Brad Buck, superintendent for the Cedar Rapids Community School District. “I don’t recall a run like this in years.”

In addition to thinking of students on buses and staff members driving in to work, Buck said it’s important to keep in mind the number of high school students driving themselves to class.

“We have a lot of inexperienced young drivers out on the road, so that’s part of what goes into it as well,” he said. “That’s why it’s difficult to set hard and fast rules ... because there’s all of these competing factors that go into play.”

ROLE FOR TECHNOLOGY?

As transportation directors and superintendents across the state wake up long before the sun rises to grapple with the difficult choice of canceling school, ongoing research at ISU could provide another tool for making those decisions.

Neal Hawkins, associate director of ISU’s Institute for Transportation, said the hope is to use Google artificial intelligence programming to take state data collected by the Iowa Department of Transportation — paired with weather data and other resources — to provide quicker alerts or the ability to detect roadway issues as they occur.

“Really, every second does matter and anything we as researchers can do to help the DOT or schools in being ready for these kinds of events, the better off we all are,” Hawkins said.

ARTICLE CONTINUES BELOW ADVERTISEMENT

Thank you for signing up for our e-newsletter!

You should start receiving the e-newsletters within a couple days.

Potentially, camera images of the state’s highways could be used with data streams to provide pavement conditions — without having to go drive on the road.

What’s more, Anuj Sharma, associate professor with the ISU institute, said data potentially could be used to better predict road conditions.

“I believe no matter which agency you look at, they sit on the source of data and then they have to make these decisions. And our hope is that we can give the decision-maker better opportunities to look for estimates of those risks and what some of the repercussions of making a decision may be,” Sharma said.

Tracey Bramble, information specialist with the Iowa DOT, said the ultimate goal is that predictive modeling would allow motorists or school superintendents the ability to better foresee roadway conditions without having to be on the road.

“If we can predict ahead of time what those road conditions are going to look like, say at 8 a.m. when the buses are delivering those kids, without the superintendents and transportation directors to have to get out and drive those roads at 4:30 — 5 a.m., it’s safer for them and it’s safer for kids. So that’s kind of our pie in the sky hope,” Bramble said.

Hawkins said much of the data available from the Iowa DOT focuses on state-owned highways or heavily-traveled roadways, while rural roads have much fewer data points to reference.

“It gets a little harder to do, and it won’t ever totally eliminate some people having to go out and check some of these roads,” he said.

Wheeler and Buck said another resource when considering a school cancellation would be appreciated, but at the same time it’s difficult to envision a future when firsthand experience won’t be necessary.

ARTICLE CONTINUES BELOW ADVERTISEMENT

“Any piece of information that can help us make a sound decision for safety is always going to be welcome,” Wheeler said. “But one of the things you just can’t replace is the actual conditions. I’m not going to say that it’s safe for students or buses to be on the road if I haven’t been on the road myself.”

• Comments: (319) 398-8309; mitchell.schmidt@thegazette.com

Give us feedback

We value your trust and work hard to provide fair, accurate coverage. If you have found an error or omission in our reporting, tell us here.

Or if you have a story idea we should look into? Tell us here.

CONTINUE READING

Give us feedback

We value your trust and work hard to provide fair, accurate coverage. If you have found an error or omission in our reporting, tell us here.

Or if you have a story idea we should look into? Tell us here.