TAMPA, Fla. – Maintenance programs often seem to involve compromises. A focus on corrective maintenance may fix out-of-service components, but it also leads to the added costs of roadside repairs, excessive downtime and fines. And Preventive Maintenance (PM) programs can involve replacing parts that have yet to reach the end of their useful service life.

An approach known as “condition-based maintenance” may offer the perfect balance – measuring the health of equipment at a specific point in time, and scheduling maintenance only when it is absolutely necessary.

“Wouldn’t it be great if a truck could send you (data to announce) there was fuel in the oil, or the alternator was putting out 12 volts?” Glen McDonald, director of maintenance for Ozark Motor Lines in Memphis, Tenn., mused during this year’s Technology and Maintenance Council meetings. “Wouldn’t it be nice to get a notification when the starter circuit is failing, before it’s dead in the parking lot?” Real-time data from a sensor could even be used to schedule different maintenance resources, directing the truck to a shop that is equipped with the right tools, parts and experts to conduct a specific task.

“Frankly, this sounds too good to be true,” admitted Guy Rini of GTR Development. But everything from additional vehicle sensors to communications technology and databases may be on the verge of transforming the way the trucking industry could approach maintenance. McDonald’s wife even owns an SUV that will already generate an e-mail if the oil begins to degrade. Couldn’t trucks offer the same feedback?

“It’s maintenance evolution,” Rini said. “Condition-based maintenance is the next major advancement in maintenance, and we have the technology to do that. These vehicles are very sophisticated.”

That may be an understatement. A 2010 Cummins engine includes no fewer than 28 sensors and 15 actuators, watching over everything from pressure to temperature and positions, and the number of sensors is bound to rise. Yet most of the data is largely ignored by a fleet until it generates a fault code.

“Maybe you don’t have to wait until it breaks. Maybe it’s possible to use condition-based maintenance technology to reduce maintenance costs,” Rini said, noting how it is possible to combine an engine ECU’s data with the readings on everything from anti-lock brakes to transmissions, instrument clusters and other computers that have emerged on a truck. “You’ve already bought everything. It’s there,” he told a crowd of maintenance supervisors.

While it would never replace corrective or Preventive Maintenance efforts altogether (there is no substitute for visual inspections when looking for a leak, crack or frayed wire), condition-based maintenance could add a new dimension to every decision in the shop.

The concept has certainly interested NASA and various branches of the US military, who are looking to the approach as a way to ensure that “mission critical” components do not fail. The Society of Automotive Engineers has also established a subcommittee that is looking at tools such as statistics, histograms, probability and distribution functions, and trend analysis, measuring the time to failures and the time between them.

“Condition-based maintenance is basically a superset of what we already have. It pulls it all together,” says Ken DeGrant of DG Technologies. The challenge is that most of today’s technologies focus on components rather than the overall vehicle. “It’s a shotgun approach…we don’t have the data to pinpoint that bulls-eye.”

Transforming the streams of data into information that predicts maintenance needs is no small task. A condition-based maintenance model would likely involve added sensors and real-time data, crunching all of that information with the computer algorithms that can translate the details into the condition of a component.

One of the first steps in preparing for the shift will be to clean up the data that is being loaded into a shop’s database. Rini suggests that this begins with accurate Vehicle Maintenance Reporting Standards, which offer codes to classify equipment, labour tasks, parts and brands.

Fleets will also need to decide where the additional sensors and information can make the biggest impact.

“Identify and prioritize the right components or systems for CBM adaptation,” Rini says, referring to work in the early days of the transition. The leading items on his list includes more data about the condition of oil, batteries, air filter systems, charge air cooling systems, crankcase pressure, fuel filters, wheel bearings and corrosion.

In each case, the benefits will need to outweigh the costs. There will also need to be the right combination of data that will lead to a specific component, rather than generating the multiple fault codes that leave mechanics “chasing ghosts.”

“The future of vehicle support requires more and more data. In fact, it requires the data more often,” suggests Mark Zachos of the Dearborn Group. That can mean reporting the data back to a central location, expanding onboard computing systems, and offering the continuous connection that can be used for inspections and maintenance. “Somehow we transfer it to a server someplace, and we mine it and process it.”

This will also require faster connections and networks, which could make it possible for vehicles to communicate in real-time. “If you have lots of data and can’t move it around, it’s not that useful,” he adds.

McDonald stresses another challenge: Individual solutions are emerging, but they are coming from too many different vendors, making it difficult to tie everything together. “I need a total truck solution,” he says. “I need a smart truck.”
It may not be there yet, but it’s coming. The emerging maintenance trend could even lead to options that have yet to be conceived.

Besides maintenance, says Zachos, “some smart people, someplace, will also use that data for other purposes.”