Michigan potholes abound after a particularly brutal winter. Lonny Garris / Shutterstock.com

A new statistical model might make the annual spring road repair cheaper for the Mitten State—and for the trucking industry.

In the great American north, spring is a time of particularly ecstatic renewal. Shed the coats! Toss the sweaters! Get outside more than the moment it takes you to sprint to your car without slipping on ice! And bring on the trucking load regulations, which arrive as surely as the snow thaw.

Northern climes institute trucking load regulations for the simple reason that trucks wreck roads as pavement defrosts from the winter. This is Potholes 101: Water seeps into and softens pavement, then freezes once it gets cold enough. The constant thawing and re-freezing of the water during the chilly season sometimes stretches pavement past its breaking point, leading to cracks and holes. And the problem only gets worse in the spring: melting ice leaves more gaps inside the pavement, which is easily smashed to smithereens by, say, gigantic truck wheels and their heavy loads.

Which is why there are seasonal trucking load regulations—often called “frost laws”—in the first place. These are instituted on a rolling basis, depending on the weather, and restrict weighty trucks from roads when they’re susceptible to cracking. Institute the frost laws too early, and the trucking industry will lose out on the money it could have made hauling large loads without risk of road damage. Institute them too late, and you’ve got a screwed-up road. It’s hard to get the timing right, and expensive to get it wrong: In 2014, the Michigan Department of Transportation spent $10 million filling potholes.

So how does a state know when to institute, and then lift, frost laws? Most have a few approaches. They look at weather reports. They straight-up peer at the roads. And then send workers out to measure the roads’ frost depth. In Michigan, this involves a nifty technology called “frost tubes,” which are burrowed six feet into ground and filled with a solution that changes color when it’s frozen. By looking at these, MDOT can get an accurate picture of how deep the frost goes into the soil.

The problem with this approach is that it’s labor-intensive—and expensive. Workers have to scramble all over the state to check the condition of its highways. Can data-assisted math do it better? That’s the principle behind an MDOT-sponsored report by Gilbert Baladi and Pegah Rajaei, engineers who specialize in pavement at Michigan State University.

Baladi and Rajaei conducted a multi-year study that devised a number of statistical models to handle seasonal change on the roads. The data they used comes from Road Weather Information Systems (RWIS)—there are 25 of these stations dotted throughout Michigan, each armed with subsurface temperature sensors. From these sensors, researchers can consistently pull information on area maximum and minimum daily temperatures. They can also use what they know about thermodynamic properties of the local soil to determine how it freezes and thaws. This isn’t quite real-time data, Baladi tells CityLab, but that doesn’t matter—the model just needs to get a sense of the historical weather and soil patterns to function.

The Road Weather Information Systems in Michigan used in the frost study. (Baladi and Rajaei)

And it works. Baladi says his team has tested the statistical model in other states, and it accurately predicts when roads are thawed enough for heavy vehicles to drive on them, destruction-free. “The model we developed works for the Upper Peninsula, the Lower Peninsula, Minnesota, and Wisconsin,” he says. “It doesn’t require very expensive parameters, and doesn’t require expensive tests to get those parameters.”

In Minnesota, which has used similar data-driven research to change the way it institutes its frost laws, the Department of Transportation estimates it’s increased the service life of some asphalt roads by 10 percent, for a total estimated savings of $14 million. Its Mitten State neighbor is hoping to post similar numbers.

But the state has to get funding first. Baladi estimates it might be a few years before a model-driven program could be instituted for all of Michigan. Meanwhile, the region has been having some wacky weather. “Winter started on the first of April,” he says: it’s been well below freezing during these so-called “spring” nights, while the actual “winter” was pretty temperate. The weather—and the road—answers to no man. Or no truck, for that matter.

About the Author

Most Popular

  1. A photo-illustration of several big-box retail stores.
    Equity

    After the Retail Apocalypse, Prepare for the Property Tax Meltdown

    Big-box retailers nationwide are slashing their property taxes through a legal loophole known as "dark store theory." For the towns that rely on that revenue, this could be a disaster.

  2. Life

    Inside the Movement to Derail Amazon HQ2 Incentives

    New York and Virginia politicians and activists could still make changes to Amazon HQ2 packages—or at least stop the next bidding war from mirroring this one.

  3. A photo of British Prime Minister Theresa May announcing her government's Brexit deal outside No. 10 Downing Street
    Equity

    Britain Finally Has a Brexit Deal. Everyone Hates It.

    Amid resignations, it's clear the U.K. government massively misjudged how leaving the European Union would play out.

  4. A photo of a small small house in San Francisco's Noe Valley that sold for $1.8 million in 2014.
    Equity

    Why Cities Must Tackle Single-Family Zoning

    As cities wake up to their housing crises, the problems with single-family-home residential zoning will become too egregious to ignore.

  5. A man holding a toddler walks past open-house signs in front of condominiums for sale.
    Life

    Millennials Are More Likely to Buy Their First Homes in Cities

    New research finds that Millennials are 21 percent more likely to buy their first homes near city centers than Generation X.