I always get super hyped when I get to write about complex technology, like this month’s feature on predictive maintenance. What’s great about predictive maintenance is that it can impact so many facets of the transportation industry, as it accumulates millions to billions of these data points. On their own, they may, like the flap of one butterfly’s wings, barely cause a ripple on a pond. But combined, this ocean of data has the force to manifest a tsunami of positive change.
Here’s a very possible scenario: A maintenance department receives one diagnostic trouble code (DTC) from a truck that lets them know something’s up with the aftertreatment system. Normally, they would still have to go through a possibly lengthy problem-solving process. But their predictive maintenance platform detects other DTCs coming from that engine, and then an algorithm searches the platform’s immense (anonymized) historical data spanning several fleets and finds thousands of similar cases with this combination of DTCs, with this make and model, powertrain, and mileage. With all that calculated, it can spit out the most likely cause and current level of severity. If a failure is imminent for a truck on the road, the driver now knows if it’s time to head to the nearest service center or complete the route and get checked out at the fleet’s terminal.
See also: Top critical and minor fault codes, according to PitstopSome solution providers are already doing this at a basic level for fleets, as we detail in the main feature, but this is just the beginning. Once more data is available, like for tires and wheel ends, which will be coming as more trailers become connected, the solutions will have a larger sample size to work with and, thus, more accuracy.
Because no two trucks wear down the same way, due to topography and loads, trailer data will further enable predictive accuracy, according to Great Dane. The OEM’s FleetPulse telematics solution provides loaded weight data, which impacts component wear.
The company said: “Two similar trailers could have each gone 30,000 miles, but if one of them is averaging 15,000 lb. more weight carried per load, its components will have more wear, and that information will let you more accurately plan its maintenance schedule.”
Add work order and repair information to that, then you’re really getting a clear vision of the future. That, in turn, helps a fleet avoid costs down the road.
Where breakdowns are concerned, we’re talking “Minority Report” levels of prognostic clarity, though you’ll be stopping breakdowns, not murders. And as we detail in our comprehensive roadside maintenance piece, breakdowns can be a real killer to profits, and more importantly, endanger the brave mobile service techs on the side of the road and generally decrease highway safety.
“Everybody would like to have that magic solution” to know when a breakdown will happen, noted Melanie Simard, director of compliance, client service & technical support at Isaac Instruments. “After drivers’ compensation, maintenance and fuel are the two major costs, so it would really help. It’s already hard as it is for a fleet to make a margin.”
Prior to Isaac, Simard worked at fleets for 20 years, so she knows those costs all too well, and Isaac’s future telematics solutions are becoming more predictive, she said.
To make that hypothetical future vision a reality, solutions will need to be more prescriptive. A maintenance team that leverages this form of data analytics-empowered maintenance takes real-time sensor data into account and will narrow issues down even more. This also gives a better sense of which assets have a high risk of downtime and prescribe what actions should be taken.
Maintenance solution provider Pitstop is already doing this with their predictive platform.
“We take voltage data every few seconds and come up with a health score on the electrical system of the vehicle,” noted Pitstop CEO Shiva Bhardwaj. In the case of electrical DTCs, this allows the platform, to identify if the battery, alternator, or starter is the issue.
At some point, once these platforms can confidently detect the true problem on a consistent basis, they can more deeply integrate with a fleet’s shop management system and ecommerce solutions to schedule work and ensure the needed repair parts are in stock.
There are some serious challenges to figure out before that happens, such as who has the right to repair. This was brought up by one maintenance executive I recently chatted with on the subject.
“Predictive maintenance is a great concept and I am looking forward to it adapting,” he said. “My challenge is there are issues that come from premature parts replacements due to warranty considerations. We need the industry to begin working with predictive maintenance providers to allow fleets to change an alternator, for instance, once it is determined to be charging at a reduced rate due to the predictive maintenance reports.”
After talking with several experts on the subject, my prediction is these issues will get solved in the next five years. In terms of the Gartner Hype Cycle, we are probably at the “Peak of Inflated Expectations” in the transportation industry and headed soon for the “Trough of Disillusionment.” But with patience and collaboration, unexpected maintenance can be a thing of the past.
