Can a high-tech wrist band measure driver fatigue and predict crashes?

by Julia Kuzeljevich

VANCOUVER, B.C. — Managing fatigue has always been an issue in the trucking industry, all the more so as the professional driver population ages.

According to data from the 2011 National Household Survey report, the average truck driver age is actually 46 years, four years older than that of the average worker at 41.5 years.

Hours-of-service regulations, sleep apnea tests and tools such as the recently-released Web site for the North American Fatigue Management Program aim to manage the issue of fatigue in the industry, but are they enough to determine where the problem areas can and do occur?

In terms of getting aggregate data on how tired drivers are, some companies whose employees perform shift work are testing technology such as that developed by Vancouver, B.C.’s Fatigue Science, a technology start-up that makes a wristband to measure the sleep patterns of the user and to predict levels of fatigue and alertness during their waking hours.

Sean Kerklaan, CEO of Fatigue Science, said that just as pedometers count steps as a way of calculating movement, “We’ve quantified that movement into sleep or awake periods. We bought the algorithm from the US military to translate the amount of sleep or lack thereof into a fatigue measurement.”

Kerklaan said the technology, manifested in a watch-like device called the ReadiBand, has been scientifically validated by both the Department of Transportation (DoT) and Federal Aviation Administration (FAA) in the US. It is also FDA- and Health Canada-approved, he added.

“We take a couple of approaches. Polysonography is used in hospitals for sleep apnea, to check for respiratory rates. Our simple watch worn on the wrist is 93% effective in measuring similar data,” he said.

Using the Fatigue Avoidance Scheduling Tool (FAST), developed by the US Air Force in 2000, ReadiBand calculates a series of algorithms to measure sleep quality from continuous wrist movements or actigraphy. Fatigue Science then downloads data from those movements for work schedule assessment. The data is then entered into the Sleep, Activity, Fatigue, and Task Effectiveness (SAFTE) model to show a minute-by-minute curve from the beginning to the end of the worker’s schedule to show the worker’s level of effectiveness at any point in time.

“When the user is barely moving or not moving at all, we can determine what stage of sleep they are in. Using that data, tracked over a period of several days, we can predict with a high degree of accuracy whether drivers are going to be safe behind the wheel. We can lay out sleep and work schedules that ensure they are getting enough rest to be safe on the road,” said Kerklaan.

In a June 2009 report, the US Federal Motor Carrier Safety Administration evaluated several emerging driver fatigue detection measures and technologies, among them mathematical models/algorithm technologies and the SAFTE model on which the ReadiBand is based.

At the time, the report concluded that the SAFTE Model “does not include the effects of

physical work, workload, or level of interest in the task. Two additional limitations are that the model does not provide an estimate of group variance about the average performance prediction and it does not incorporate any individual difference parameters, such as age, chronotype (morningness/eveningness), and individual sleep requirements.”

When asked about the results of this report, Fatigue Science founder Pat Byrne said the issue is that the 2009 review was based on technology available at that time, but not the current SAFTE model, which has undergone further development and validation.

In 2012 the FAA (part of the US DoT) put together a report based on a field validation of the SAFTE model using data from a broad sample of 178 aviation cabin crews from the 2009-2010 FAA Civil Aerospace Medical Institute.

“The earlier concerns of the FMCSA did not show up in actual field tests conducted with 178 individuals and over 10,000 individual comparisons with actual PVT (fatigue tests),” said Byrne.

According to the 2012 report: “Despite inherent technical limitations and issues of inter-individual variability, these results clearly support the validity of the SAFTE model for population-level prediction of fatigue-induced impairments in objective neurobehavioral performance capacity in extremely dynamic 24-hour field operations such as commercial aviation.”

Arrow Transportation Systems in B.C. has just finished a trial of the ReadiBand. Dan DePalma, general manager, northern operations, told Truck News that the fleet considered testing the ReadiBand technology after hearing about it from a sports connection.

Kerklaan confirmed that Fatigue Science is increasingly working with professional sports teams and has had its technology screened by the Vancouver Canucks.

“The ReadiBand group is involved in different industries, and this dovetailed into something that would be very useful for our drivers. I know in a lot of industries fatigue is an issue, so testing the ReadiBand was an attempt to manage fatigue, which is the cause of many other health issues,” DePalma said.

He said it was not hard to convince drivers to test the technology “once we explained what we were doing and gave them the reasons why, and that we were coming out of a place of care and support for drivers. I was pleasantly surprised at the buy-in. At the end of the day I just want them to learn about themselves and help them improve toward safety, health and overall well-being. It’s an accessible tool. I don’t think there’s any danger in losing productivity. I think that hiding from challenges is the danger – if we use the tool and learn from it, there’s no question that we’ll get better numbers, and better overall performance from our drivers,” explained DePalma.

While drivers can view their individual results, fleet managers can only view the data in aggregate.

“It gives me something to reference on schedules and we do everything we can to adjust rosters and schedules. We’re just trying to get people to understand the role of sleep in managing family and work. Sleep should be higher up on the priority list. The most important thing for us at Arrow is we wanted to address any issues. We want to improve and respond quickly,” he said.

Going forward, Arrow Transportation intends to continue using the tool with director of safety Rick Viventi as project manager.

“We just did a sample to start and now we’re going to do a 12-month project kick-off in September. We did an initial sample of ‘X’ number of drivers but we’ll now try to get as many of our people in Alberta involved as we can and look for trends on what we’re learning or doing better,” said DePalma.

Though Fatigue Science is a Vancouver-based tech company, Kerklaan said the company has been primarily focused on the Australian transportation industry because legislation there has forced the industry to pay attention to fatigue.

“Australia made managers responsible for the work shifts they make their workers do,” he said. “We can take a shift and run it through the algorithm to see if that shift exposes workers to extreme fatigue. We know if you’ve been up for 19 consecutive hours, you are 40% slower to react. They become personally liable for that,” said Kerklaan. “Much of our business up until now has been reactionary. When there is an accident and it’s investigated, our algorithm can be used retrospectively, to see where there was danger. Correlating the data against accident data shows the fatigue exposure. Forecasting forward means manipulating work and break periods.”

In North America this kind of metric is more of a forward thinking mechanism for companies looking at their insurance costs, for example.

“Our tool is not meant to solve all the problems in the world. It’s meant to qualify in a meaningful way the roster of hours your employees have to work and what this exposure is. Our goal is to make sure that companies don’t lose any productivity,” said Kerklaan.

“We push real-time data; when they push a button on the ReadiBand it shows them their current data. The US DoT has correlated a 70% fatigue threshold to a blood alcohol level of .08 (an inebriation level). The level changes based on how you’ve been sleeping the last few days.”

Kerklaan acknowledges it is a challenge to change the way business is done in certain industries.

“We know we can’t just change the business, but could a 30-minute nap increase the safety of the shift?” he said. “We’re looking at some systemic issues here and how government can address these.”

Have your say

This is a moderated forum. Comments will no longer be published unless they are accompanied by a first and last name and a verifiable email address. (Today's Trucking will not publish or share the email address.) Profane language and content deemed to be libelous, racist, or threatening in nature will not be published under any circumstances.