Trucking nears predictive maintenance inflection point

Predictive maintenance is one of those intriguing fleet technologies that trucking Nostradamuses routinely insist will break through next year. It’s an annual tradition.

Heck, Fleet Maintenance was writing about predictive maintenance as far back as 2007. We wrote:

The biggest drawback to predictive maintenance is initial cost—both in terms of the diagnostic equipment/software and technician training. Fleet managers may also have concerns about tying up resources too often on predictive tasks instead of the more routine—but essential—preventive maintenance and service tasks.

Nearly 20 years later, fleets still have concerns. Even a few years ago, a major one was the lack of data to train models. A predictive model needs historical data at scale, contextualized with real-time data, to make accurate forecasts on when a pump or engine may fail.

But predictive maintenance solutions are rapidly becoming far more abundant and effective. Every truck that sends health data to the cloud and shop that digitizes its work orders feeds the predictive beast, making it bigger and more effective in hunting down potential failures. A predictive pilot project can quickly become a grounded maintenance strategy thanks to the deeper wells of historical data and smarter algorithms (which only get better as AI evolves).

Does all this mean the technology will finally have its breakthrough moment in commercial vehicle maintenance and repair next year? No, because we think it already happened last month with the acquisition of Uptake by Bosch and Pitstop by Fullbay. That’s two leading predictive solutions gaining unfettered access to the data of a global Tier 1 supplier and a massive heavy-duty shop management platform, respectively.

You may not immediately notice the effects in your fleet operation, but these moves signal that predictive is entering its maturation phase in the CV sector, and there’s no more waiting for next year; it’s ready for the big leagues.

Fleet Maintenance spoke with leaders involved in both deals to get their take on how these moves may push predictive maintenance into an indispensable tool to fight downtime and make the most out of a fleet’s data.

Bosch and Uptake

Bosch, one of the preeminent auto parts suppliers in the world, makes just about everything from brakes and steering components to wiper blades and diesel injectors. The German company also helped pioneer Electronic Control Units and auto diagnostics, so yeah, it’s kind of a big deal.

Uptake was founded in 2014, and its data modeling and predictive analytics have proved a solid solution to help fleets avoid major downtime events. United Road, a fleet of 1,000+ vehicles, was estimated to save more than $3,300 per vehicle annually using the technology.

On March 19, Bosch announced its intention to acquire Uptake. This came after months of partnership discussions, according to Uptake CEO Adam McElhinney. He said Uptake was looking for a way to get more truck data to better serve its customers, and Bosch had that with its Automotive Connectivity Hub (C-Hub), a telematics device that provides vehicle health and diagnostics data. 

“The discussions regarding collaboration with Bosch naturally evolved toward the benefits of being fully together so we could combine our AI expertise with Bosch’s vehicle system know-how, creating a unique market capability,” McElhinney said.

Existing services and integration will remain, he said, though “customers should expect to see some potential new exciting features in the near future.”

He could not talk specifics at this early stage; however, feeding Uptake’s predictive model a steady diet of Bosch’s data is like pouring a dump truck of spinach into Popeye’s mouth.

“When you combine [Bosch’s data] with Uptake’s predictive analytics platform, you are connecting intelligence at the source of the data with advanced AI on top, McElhinney said. “For customers, that translates into earlier and more accurate detection of issues because we are working with higher fidelity data coming directly from the vehicle and its components. It also enables tighter integration into existing workflows, from diagnostics to maintenance planning, so insights do not just sit in a dashboard; they drive action.”

Ultimately, the benefit is very tangible. Less unplanned downtime, better maintenance decisions, and lower total cost of ownership across the fleet.

This scale will have major, though possibly nuanced, effects on customer uptime. McElhinney explained:

“With predictive maintenance, you’re fundamentally trying to detect events that are extremely rare, but very costly. A total coolant system failure on a truck, for example, may almost never happen, but when it does, it can potentially mean a $60,000 to $70,000 engine replacement—not to mention safety risks, operational disruption, and downtime.

“Because these events are so rare, scale becomes critical. You need to see as many assets, operating conditions, and failure examples as possible to train models that can reliably detect early warning signs. That’s where the Bosch partnership is transformative; by connecting to Bosch’s global network of connected vehicles and equipment, we dramatically expand both the volume and diversity of data. That helps our models to detect issues earlier, with greater accuracy, and apply those learnings across a much broader set of assets.”

Fullbay and Pitstop

A week after the Uptake acquisition, heavy-duty shop management provider Fullbay acquired Pitstop, a Canadian startup that estimates its AI-enabled platform saves fleets an average of nearly $2,000 per vehicle by using AI to provide maintenance alerts and scheduling assistance.

Shiva Bhardwaj, Pitstop’s founder and CEO, grew the company out of his family’s Toronto auto and truck repair shop. He noted his company’s AI has identified potential failures with more than 94% accuracy, improving fleet uptime by 30%.

“That's genuinely disruptive on its own,” Bhardwaj explained to Fleet Maintenance. “But here's what makes this combination something else entirely: Pitstop optimizes how efficiently the repair happens. Fullbay optimizes how well the shop runs. When those come together, it's the ultimate optimization for the vertical.”

He said current customers will continue to have the same platform, but it’s now “supercharged with a direct line from vehicle intelligence into actual repair workflows.”

“The reality of predictive maintenance is that it's only as good as the data underneath it, and most platforms are working with a fraction of what we have access to,” Fullbay CEO Trent Broberg said. “So this isn't just about bringing a new tool to our customers. It's about the fact that we can power predictive insights at a scale and accuracy level the industry hasn't seen before.”

Fullbay has processed more than 20 million service orders from independent shops, dealerships, and fleets. Annually, the shop management platform processes $6.5 billion in repair orders from its users.

Customers will now also be able to access fleet and shop management data from one solution.

“Those customers have always had to stitch together separate systems for repair management and fleet maintenance, and a lot of that work was manual. Pitstop closes that loop,” Broberg said.

And the longer the two platforms work in tandem, the better the solution becomes.

“The more repairs flow through Fullbay, the smarter Pitstop gets,” Bhardwaj said. “The smarter Pitstop gets, the fewer emergency repairs those shops have to handle. It's a flywheel.”

Broberg said Fullbay vetted “every solution in the market” and that Pitstop met their criteria the best, learning from fleet users that Pitstop “overdelivers,” he said. and allows the shop management platform to take a giant leap forward.

“This wasn't just a technology acquisition,” Broberg asserted. “We were buying capability and knowledge that would have taken us years to build organically.”

For Pitstop, the acquisition breathes new life into the platform.

“Two things kill AI companies: bad data and no distribution,” asserted Pitstop CEO and founder Shiva Bhardwaj. “This deal solves both overnight.”

Details are still being worked out, but Broberg noted, “predictive maintenance only creates value when it's connected to the repair workflow. Insights sitting in a separate system don't help anyone. So, however this ends up being packaged, the goal is that it works with how our customers are already operating.”

He said they are not going to rush out a model that doesn’t best serve customers, and that more details will be revealed at Fullbay’s annual Diesel Connect Conference in May.

Overall, Broberg also believes these acquisitions are just the thing to make predictive a fleetwide tool to fight downtime.

“The technology has been promising for years, but seeing the future and getting there are different things,” he said. “The biggest challenge continues to be that AI hasn’t had enough data to work with.  Our thesis with Pitstop is that the missing ingredient has always been data. Pitstop brings the AI and the predictive engine. Fullbay brings more than a decade of service history from thousands of repair organizations across the country. That combination is what we think finally makes this real for mainstream shops and fleets, not just a proof of concept for enterprise customers with unlimited IT budgets.”

For Bhardwaj, the acquisition is also about future-proofing a predictive solution.

“Predicting a failure and triaging maintenance is only half the problem, he said. “The other half is getting it fixed, fast, at the right shop, with the right parts in stock. Fullbay and Pitstop are building the first vertical solution that covers the entire arc, from asset failure detection all the way into the last mile of the repair. Owning that process end-to-end is the only way to actually keep commercial fleets running as efficiently as possible."

It's still way too early to tell on the true impact of these deals, and thus, the accuracy of our predictions on their signifigance, but Bhardwaj is confident that this maturign technology will have a huge impact for decades to come.

"In an autonomous future, this maintenance flow will be table stakes," he concluded. "We're excited to be pioneering it now.”