“Machine learning.” It’s a phrase that’s cropping up more and more in conversations about data-driven solutions. In the heavy-duty service world, one of the ways it’s being employed is through predictive maintenance algorithms, complex strings of code that crunch your service data, analyze it and then give you recommendations based on probability models.

Make no mistake: This technology is coming, soon. Both Navistar and Daimler Trucks North America have teased predictive maintenance solutions that will be backed up by warranties. With Navistar’s OnCommand Connection Live Action Plans, for example, the company has said that if the component is under warranty and the Live Action Plan makes the recommendation to replace that component, Navistar will pay for that component before it fails. Of course, the warranty means that Navistar would have paid for it after the fail anyway, but being able to catch it at a PM rather than when a truck is on the side of the road has the potential to save thousands of dollars.

So let’s put aside the question, “Does it really work?” because if OEMs are going to pony up to pay for the part before it fails and you can work that into your regularly scheduled maintenance for the truck, then you’re going to fix it whether you believe it or not. The trick will be executing the service and then rolling that info into your service management software system.

“It is imperative your solutions, whether remote or in-shop, be compatible with your entire fleet of commercial vehicles,” said Dave Covington, Noregon’s chief technology officer. Noregon has gone beyond its JPRO Professional in-shop diagnosis and repair software to offer TripVision, a vehicle health monitoring system, which provides a detailed, real-time view into every vehicle in a given fleet through any internet-connected device. Since its release two years ago, TripVision has evolved to leverage Noregon’s wealth of diagnostic data to craft its own predictive maintenance solution.

But let’s back up a moment to understand the importance of integrating your in-shop system with remote diagnostic system.

“Consider the effect it has on finding the root cause during repairs,” Covington said. “If a TripVision user detects an issue on the vehicle, he or she can view the fault history from every JPRO connection that vehicle has experienced. If the user sees a commonly recurring fault, it should be relayed to the service provider that the root cause is likely not being detected and only temporary fixes are being applied to the problem.”

The predictive maintenance capabilities of TripVision go one step further to try to stop the commonly reoccurring fault before it’s a commonly reoccurring fault.

“TripVision’s algorithm predicts failures based on learnings from fault histories and input from ASE-certified technicians,” Covington explained. “For example, we recognize certain faults that often lead to an engine de-rate and can alert the user to stop operating that vehicle as soon as the fault is present. This offers a tremendous cost-saving opportunity to the user.

“Think about it: do you want to heed TripVision’s advice to pull over and perform a forced regen, or drive until the issue festers and causes a breakdown that requires a tow and repairs?”

Keep in mind, you’re not playing a passive role in this process. The data you record within your service management systems will feed into future algorithms and probabilities. So ensure that your data is accurate and complete.

“Think about the term, ‘Garbage in, garbage out,’” Covington said. “Predictive maintenance will not be possible without vehicle owners marrying their vehicle data with their maintenance records. To improve the algorithms, we need to know the exact state of all major components when a failure occurred to detect analytical trends and provide accurate maintenance recommendations. In turn, fleets must trust the advice and be willing to perform maintenance or make repairs before a failure ever occurs, in order to realize the benefits maintenance can create.”