Heavy-duty: ACT Expo opens registration
Every morning you grab your cup of coffee and scroll through your telematics dashboards, soaking up the fresh data and deciding how to put it to use today.
Here’s something to consider: Every bit of data you have on your fleet can be relevant to driver retention and recruitment. Your drivers are concerned about the minute details of the truck they drive and the company they work for. That means everything from miles driven and hours of service to fuel economy and tire maintenance is fair game.
That’s why trying to cross-reference your data to what’s going on with your drivers is so hard. With so much to process, it can be easy to jump to conclusions based on analyzing too few data points, or the wrong data altogether.
“My favorite example of misusing data for driver recruitment comes from a conversation I had with a vice president of operations,” says Chris Orban, vice president of data science for Trimble Transportation. “He told me, ‘The last three drivers who quit all came from Florida. We shouldn’t hire any drivers from Florida!’ Well, when you looked at the data, 20 drivers had quit in the last week. Yes, the last three had been based in Florida, but of the entire fleet, 50% were based in Florida. So, statistically, we would have expected 10 of the drivers who quit in the past week to be from Florida. So, in fact, Florida drivers are quitting at a lower rate.”
Orban says this is why it’s supremely important to understand the data you capture and to make sure it is clean and accurate, especially when it comes to something as complex as managing your drivers. Normalizing your data—organizing what you know in a way to produce the most successful outcome for your fleet—is paramount.
Let’s consider for a moment how your data relates to your fleet’s ability to retain great drivers.
Dr. Ashim Bose is chief data scientist and vice president of Omnitracs’ artificial intelligence, machine learning and data engineering organization. He says when it comes to driver retention, many fleets tend to focus on the negative insights, such as accidents and idling. But there are other data sets that might be more useful for a fleet’s driver retention purposes.
“If the best drivers are spending more time on the road than the rest of the fleet, there is an opportunity to change their schedule and prevent them from burning out,” he says. “Route optimization and driver navigation data are some of the most beneficial insights for drivers. Not only does this data help to limit the time drivers are spending on the road, therefore increasing driver satisfaction, it also helps fleets cut down on costs and improve customer satisfaction.”
Trimble’s Orban adds that you might already have access to the data to predict driver behavioral changes that lead to resignations; the key is to take advantage of it before you lose one of your good drivers.
“Typically, a fleet focuses on what has happened when considering retention and recruiting challenges, such as, ‘Our drivers are leaving because we aren’t paying them enough, let’s raise our base pay,’” Orban says. “You need to look at key factors, such as pay, driver sentiment in communications, the workload and the kind of work you’re asking of a driver, to get ahead of the curve. The goodwill that you generate by offering a raise, or guaranteed home time, or newer equipment before drivers begin to leave is incredibly powerful.”
What about using data to improve time behind the wheel?
Providing good pay and a great work-life balance may not be a revolutionary concept, and you don’t need an expert to tell you that safety is right up there among a driver’s chief concerns, too.
To this end, data can be one of your most useful tools for coaching your drivers and improving overall safety on the road. There’s a good chance you’re probably already helping drivers stay safe via software to produce in-vehicle driver safety alerts. Geotab’s Stephen White, heavy truck fleet enterprise business manager, recommends also implementing gamification—applying game-design elements in non-game contexts, like driving a truck—into driver scorecards as part of your fleet’s driver safety program.
“Gamification is a great strategy to motivate drivers to improve their driving behavior because they can see where they measure up against other drivers and the company’s safety standards,” White says. “Top drivers can receive bonuses or other recognition for their hard work.”
White says a driver’s ability to adhere to the predefined safety rules of the fleet such as speeding regulations, seat belt usage and braking and acceleration habits should all be included on a driver’s scorecard. And, he says drivers should have access to this data. Access allows the driver to understand how they are performing in relation to other drivers.
Trimble’s Orban adds he is also a big fan of the driver scorecard, as long as it’s a balanced one.
“Many times, drivers have a better understanding of data and what is being asked of them than we do in the office. They know that they have a lower MPG because they are constantly being asked to haul max-weight loads, versus a peer who might be hauling only lighter weight freight,” Orban says. “Therefore, making sure that all the metrics are fair and balanced across the driver population is important for driver buy-in.”
Your data is ultimately only as powerful as the one who wields it. Numbers alone might not paint the full picture when coaching driver safety behaviors. Orban says considering the time and place that an event occurred, the weather and traffic conditions at the time and the driver’s history can all be important.
“For instance,” Orban begins, “let’s say you have forward collision mitigation in your vehicles, and a driver has two events, indicating that the truck’s brakes fired to prevent a collision. You want to talk to the driver about these events, as they are serious safety concerns. But, do you know for sure the difference between a vehicle cutting your truck off, and the driver trying to avoid an accident, or if your driver was distracted and was the one who was following too closely?”
“You need to look at all the data to make some of these determinations. Even better, if you have a video solution, you can tie it directly to that event you want to analyze, and have some really powerful coaching conversations.”
So driver-facing cameras are necessary to get the most out of driver-related data?
Driver-facing cameras can be a great tool to coach drivers on how to lower their accident rate, making them aware of common distractions on the road and behavioral patterns, says Omnitracs’ Bose.
These cameras are especially helpful when it comes to using predictive analytics to help determine which drivers might be most at risk for a crash by analyzing driving behaviors, Bose says. Geotab’s White adds that driver-facing cameras can enable fleets to gain more insights that can help keep drivers safe by identifying risky driving, providing tailored driver training programs and protecting drivers against high insurance claims by exposing more context around traffic-related road events.
Yet, many drivers see them as a violation of privacy. So if the goal is driver retention, is the added data provided by the camera worth it?
It can be if your fleet has successfully communicated the positive values these driver-facing cameras can bring to the life of the driver, Trimble’s Orban says.
“We need to reassure drivers that we are not trying to watch them every minute of the day, we are trying to protect them and will only look at driver-facing video when it can be used for exoneration or for positive driver coaching. Certainly, if a driver is going to text and drive, or is fatigued, we need to use the cameras to detect that behavior as well, but I think there are few drivers who would disagree with using an inward-facing camera to save lives,” Orban says. “I think the key is to link driver-facing camera data to positive outcomes, not negative ones.”
It sounds like you need a lot of data for reliable results. Can data help smaller fleets?
While the advantages of having large buckets of data are abundantly clear, no amount of data is useless. Smaller data sets can allow you to home in on specific behaviors of your drivers, Orban says.
“In a 50-unit fleet, five or six drivers could dramatically influence the data, whereas that same population in a larger fleet would barely make an impact,” he says. “To be clear, in the smaller fleet you might actually find something critical in those five drivers because you can do more personal analysis with the drivers, but in general, a larger data set allows you to make more global decisions, and to avoid the trap of biased data.”
To help avoid the pitfalls of biased data, small fleets should ensure that the most important data is being presented to them in the simplest and most efficient way possible, Geotab’s White says.
“For instance, small businesses can customize their dashboards to display the most important information and/or have the most important reports emailed to their direct inbox,” he suggests. “Another way for small fleets to improve efficiency and cut costs is to optimize driver routes. A common philosophy among small fleets is to get the most out the least. By optimizing routes, fleets can easily decipher the most efficient and cost-friendly way to carry out their operations.”