Many facilities have their critical equipment covered, and a blind spot everywhere else. This gap is likely where your next unplanned shutdown is coming from. But the technological leaps and bounds are already out there to help address this problem, writes Jared Pfeiffer, a manufacturing reliability specialist at Augury. And it’s only becoming more affordable.
Walk into a modern manufacturing facility today, and you’ll hear the same story. The costly equipment, the assets production planners lose sleep over, is wired and watched around the clock. Vibration sensors, temperature monitors, oil analysis: the works. Maintenance teams can tell you the health status of those machines down to the minute.
Then there’s everything else, like the pumps supplying coolant to critical assets, the conveyors that transport material between them, and the fans, valves, and auxiliary drives that keep the entire system breathing. In most plants, these so-called “secondary” assets run dark: no sensors, no monitoring, and nothing but a preventive maintenance schedule probably designed in a different era for different conditions.
The logic seems defensible at first. Resources are scarce; prioritize where the exposure is highest. The issue is that this reasoning has a crucial flaw.
The cascade effect
I’ve seen this happen more times than I can count. Organizations assume the assets they prioritize are the most important. But without conducting a full review and involving the right personnel, secondary and tertiary assets, and other support equipment across the entire plant are overlooked. They might have a backup plan, but they haven’t tested it. They don’t know if it can actually work.
I remember once at a major wire manufacturer, the production process depended on chilled water to cool the product after it had moved through a long, continuous vulcanization tube. One day, we had a catastrophic failure on one of the chillers and lost an entire stage, and that chiller took the entire front half of the operation down. No predictive monitoring was in place. Nothing flagged the developing fault.
This is the central paradox of the way most facilities approach machine health: by protecting only your critical assets, you’ve left the door wide open for everything around them to take them out.
Overloaded, not oblivious
Why does the gap persist? It’s less about ignorance and more about overload. Typically, teams are too busy firefighting to invest the necessary time. They’re focused on fixing what’s directly in front of them. They can’t step back and see the big picture. Reliability managers, when they are present, are often double-dipped into other functions. The strategic perspective is overshadowed by urgent issues every day, until something breaks badly enough.
For a long time, the economics seemed to justify the gap. Deploying sensors across an entire facility was expensive, and the data those sensors produced was only valuable if you had the expertise and bandwidth to act on it. Most maintenance teams didn’t. Many still don’t.
The improved economics of fuller coverage
Fortunately, the economics of broader coverage are getting better. The rise of AI-driven analysis is the main factor changing the game, along with increased competition in this rapidly expanding market, both of which are leading to lower prices. As technology and software advance, we will start seeing leaps and bounds in how condition monitoring is being done.
Imagine being able to ask questions that normally require a dedicated role to answer, and receiving a plan or contingency based on what the system observes. As we progress and our product offerings expand, we close the gap that’s always been about not enough time and not enough money.
The question facing maintenance and reliability teams now isn’t really whether to monitor secondary assets. The failure data has settled that argument. The question is how to do it in a way that’s practical, scalable, and actually useful, rather than just adding more noise to an already noisy environment.
The big picture
The key is to adopt a more integrated view of the facility, one that considers not just individual assets but the overall health of the entire production system. Imagine an operator with a screen showing how the product is moving along, right next to a screen displaying the health of their operations and processes, all at the same time.
When something begins to change, you can trigger an alert to maintenance or production: here’s what’s happening, and based on previous events, here’s what you should do. This is not a distant vision. It’s where the industry is actively heading.
What’s emerging is more a shift in philosophy than a technology story. The companies that are staying ahead of this aren’t just buying more sensors. They’re rethinking what “critical” really means, and understanding that in an interconnected system, the answer could be: almost everything.
