Plowing Toward Safety

When it comes to operating heavy snow removal equipment, practice makes perfect. Not surprisingly, the regions with the least snowfall do not necessarily have the lowest accident rates.

It’s a dark and snowy night. You’re a snowplow driver in Arizona’s high country, chugging up icy mountain passes and careening down steep, hairpin turns slick with slushy snow.

You’ve got your defrosters turned on full blast. The hot air dries out your eyes and drowns out the voice of the radio dispatcher. The cab is so stifling that you’ve stripped down to your shirtsleeves. Even still, the ice cakes up your wipers. Eighteen-wheelers thunder past, spraying more dirt and snow on your windshield until it looks as if the glass is coated with a layer of Vaseline.

While you try to keep your truck steady, impatient motorists illegally pass on the right shoulder, nearly clipping the tip of your plow. As you climb a long rise, cars begin to spin out and stall all over the icy highway. Without taking your eyes off the road, you reach for the handle that controls the truck’s 13-speed gearbox and begin downshifting.

You slow down, knowing that a collision likely would result in serious injury, even a fatality. You then fumble for the lever that lowers a 3,000-pound plow blade and another that spreads de-icing materials onto the roadbed. All the while, you keep your eyes focused ahead; aware that one skid into a ditch could wreck the $200,000 piece of machinery in your care.

Even one truck out of commission would worsen the highway havoc. Miles of unplowed snow would begin to pile up, as would the costs of delays to the commercial trucking industry and other businesses.

You’re only eight hours into a 12-hour workday and already you’re hungry and exhausted from the pressure. Your neck muscles are so tensed and sore that they are painful to the touch.

“The first time the drivers described this to us, I could feel their white-knuckle experience,” says Mary Kihl, a professor in the School of Planning at Arizona State University. Kihl directs a research team that is helping the Arizona Department of Transportation (ADOT) find ways to improve snowplow-driver performance on Arizona’s highways.

Kihl enlisted the help of Don Herring and Peter Wolf, members of ASU’s industrial design program. At the heart of their two-year study is the question of whether or not to purchase a series of driving simulators. The tool would add a new component to the state’s existing driver-training program. That program already includes a regimen of classroom instruction and hours behind the wheel. The team’s goal is to determine whether logging time on the simulators could better prepare snowplow operators to handle the hazards of real-world storms.

Purchasing simulators for Arizona, the land of sunshine and swimming pools, seems counterintuitive, says Steve Owen of ADOT’s Arizona Transportation Research Center (ATRC). But according to a report issued by the center, of ADOT’s nine statewide districts, five routinely experience severe weather. Most of the others have high-elevation areas that are hit with freezing rain and snow intermittently. To help keep these roadways safe and passable, the State of Arizona maintains a stable of 250 operational snowplows and employs between 400 and 500 drivers.

When it comes to operating this heavy equipment, practice makes perfect. Not surprisingly, the regions with the least snowfall do not necessarily have the lowest accident rates.

But even drivers in the snowiest part of the state can get out of practice.

According to the ATRC, before the wet winter of 2004- 2005, only 20 percent of drivers in the Flagstaff district—among the snowiest in the state—had experience in heavy snowstorms. Nonetheless, these drivers bear a huge responsibility for clearing Interstate 40, a major cross-country commercial trucking route that is regularly strafed by winter storms.

“When it comes to driving simulators for snowplow operators, you wouldn’t think that Arizona would be at the top of the list,” Wolf says. “People in the Minnesota Department of Transportation told us that their drivers plow snow so much of the year, they never get rusty. We talked to drivers in Arizona who only plowed snow twice in five years on the job. Seat time in the simulator isn’t the same thing as driving on a real highway, but it helps to brush up skills.”

Herring and Wolf are charged with the job of determining whether or not the simulators make an appreciable difference in the skills—and ultimately the safety record—of snowplow operators.

The ASU researchers are specialists in a branch of product design known as human factors: the study of the interaction between humans and machines. To familiarize themselves with the challenges that drivers face, the pair traveled to Flagstaff in winter 2004.

In Flagstaff, they accompanied ADOT snowplow operators on a test drive along Interstate 40. Herring, Wolf, and Kihl also conducted a series of focus group interviews with snowplow drivers. They asked participants to list the simulator exercises that best prepared them for challenges on the road. They were also asked to suggest ways in which the simulator exercises could be programmed to complement their behind-the-wheel training.

In December 2004, Herring and Wolf began a third research phase. They observed drivers mastering the on-screen challenges presented by the simulator.

The simulator being used is manufactured by the Utah company L3 Communications. The device mimics many of the features of a real truck cab including a full seat with air suspension, dashboard, steering wheel, and gearbox. Three flat screens offer trainees a 180-degree view of the road along with a rear-view mirror perspective.

The simulator offers different plowing scenarios. It can be programmed to introduce surprises that drivers are likely to encounter on the road. Hitting an unexpected patch of ice, for example, forces drivers to learn to counter steer their way out of a dangerous skid.

The virtual rig can be buffeted by sudden crosswinds or experience a tire blowout. Drivers also gain proficiency in such fundamental operations as properly shifting on 13-speed truck transmissions, a skill that can realize savings in fuel consumption and engine wear and tear.

The real goal of the simulator training is not to polish motor skills but to improve decision making under stressful conditions. To analyze these dynamics, Herring and Wolf found themselves using knowledge that ordinary people normally don’t associate with design. The study of human factors, Wolf says, often is viewed as the process of optimizing the physical relationship between people and the machines they operate.

“If you have to pull a lever, for example, it is useful if the lever is located not too far away from your hand and if the handle fits your hand,” Wolf explains. “But there’s also the cognitive aspect.” Snowplow drivers, for example, rely on finely honed motor skills. At the same time, they are called upon to make judgments—often split-second decisions behind the wheel. Those decisions are based on ADOT protocol for helping stranded motorists, say, or the proper timing for dispensing de-icing compounds.”

Wolf says that by using a simulator in a safe environment, researchers can keep increasing the load of distractions on the driver. They then watch for the point at which the driver is essentially unable to multitask anymore.

“The point is to make sure that the drivers are so on their toes when they get into the real truck that there won’t be a moment’s hesitation when they are faced with real-world crises,” Wolf continues. “One of our challenges is to figure out what the simulator can and can’t be used for. We want to make sure that these things are very clearly a part of the driver-training program.”

In their final research phase, Herring and Wolf plan to head out at the first sign of heavy snow to the Globe district some 90 miles east of Phoenix. Not much of the heavy stuff fell in 2006.

ADOT purchased a simulator and began training district drivers in 2005. Outfitted with night-vision video cameras and audio recorders, the ASU researchers plan to accompany drivers on their 12- hour shifts to get a first-hand look at how they put their simulator training to the test.

In the meantime, Kihl is reviewing five years of data to determine whether there is a measurable difference in performance due to driver training on the simulator. Her research juggles a mind-boggling array of variables. Those items range from potential savings in fuel economy to a reduction in the number of insurance claims for human injuries, cracked windshields caused by rocks that have been churned up by the plows, and damaged roadside mailboxes and highway signage.

These costs will be weighed against the price tag of each simulator. The machines cost a whopping $150,000 for a package that includes simulator hardware, software, and training instructors.

Keep the price in perspective, Wolf adds. “If the simulators are well used as a training tool, that cost could be a relative bargain compared to replacing a $200,000 truck or paying out damages in a multimillion dollar lawsuit.”—Adelheid Fischer