The manufacturer
Their recommendation is written for the worst case and to avoid liability. A good starting point, and almost always conservative.
Guide
A piece of equipment's service life isn't built in at the factory: it's built over time
Two identical machines installed on the same day can end up with one lasting twice as long as the other. The difference is almost never in the equipment, it's in what was done to it and when.
What is periodic asset maintenance
It's the set of interventions repeated on a piece of equipment at a defined cadence, whether based on elapsed time or accumulated use, to keep it within its operating conditions. It's not the same as checking it whenever someone remembers: what defines it is the cadence, because the cadence is what turns a set of separate interventions into a comparable series. And that series is what later lets you see something changing before it fails.
What happens to equipment that isn't maintained
It rarely breaks all at once. It degrades, and the degradation has phases.
- First it loses performance without anyone noticing: it uses a bit more, takes a bit longer, cools a bit less.
- Then it starts giving warnings: noises, vibrations, rising temperatures. Someone almost always mentions it.
- Then minor faults appear that get fixed on the spot and never get logged.
- And finally it fails outright, stopping the service, usually at the worst possible moment.
- The cost of intervening multiplies at each phase, and the information to have avoided it was there at the second one.
How the interval is decided
There are three sources, and all three are biased.
The regulation
When it exists, it's non-negotiable. It can be stricter or more lenient than what would technically make sense, and it rules either way.
Your own history
The only source that knows your installation, your usage, and your environment. And the only one that lets you adjust, which is the whole point.
Adjusting is the work, not building the plan
A preventive plan built straight from the catalog and never revisited always overshoots in one place and falls short in another, and almost nobody revisits it because it takes a piece of data almost nobody has: where failures cluster within the interval. If failures show up right before the next check, the interval is too long. If nothing has failed in three years and the check finds nothing to do, it's probably too short and you're overspending. That adjustment, done once a year on the equipment that matters most, is the intervention with the best return in all of maintenance management, and it doesn't cost money, it costs looking.
Shall we look at it with your way of working?
Leave your details and we will get in touch to see whether we fit. No commitment, no lock-in period.
By time or by use
For a lot of equipment, the calendar is a poor indicator. A treadmill with fifteen hundred gym members and one in a hotel don't wear the same even at the same age; a forklift running three shifts and one used on Fridays don't either. When wear is driven by work rather than time, anchoring the plan to a counter of hours, cycles, or kilometers avoids the two things you want to avoid: over-checking what's barely used and under-checking what never stops. Logging the reading when work is done is all it takes to make that work.
Overdoing it costs money too
The conversation about periodic maintenance almost always assumes the risk is doing too little, and doing too much has a cost that's just as real and far less documented. There's the direct cost of hours and materials spent on interventions that weren't needed. There's the downtime to check something that was fine. And there's a less obvious effect that any experienced technician recognizes: every intervention introduces a chance of error. A piece of equipment that gets opened, disassembled, and reassembled can come out worse than it went in, with a poorly seated gasket or a bolt that didn't get torqued right. That's why the right interval isn't the shortest one you can afford, it's the longest one that still avoids failure, and finding it requires exactly what history provides: knowing what was found every time it was checked.
An interval without records can't be corrected
Everything above depends on a condition that's usually assumed and almost never met: that every preventive check leaves a record of what was found, not just that it was done. The difference seems small, and it's what decides whether the plan can improve over time or stays frozen forever. If all that gets logged is that the check happened on schedule, in two years nobody will be able to decide whether the interval is right, because there won't be any information to decide from. If instead every check saves the measurements, the condition of wear parts, and what had to be fixed, the plan becomes a living document: intervals that always come back clean get longer, ones that show wear get shorter, and next year's plan comes from data instead of from last year's plan.
Frequently asked questions
How do I know if an interval is set correctly?
By looking at what's found each time. If several checks in a row find nothing to fix, the interval is probably shorter than needed and can be extended with caution. If advanced wear shows up, or a failure happened between checks, it's too long. It's a continuous adjustment, not a one-time decision, which is why it matters so much that every check logs findings and not just the date.
By time or by operating hours?
By whatever actually causes the wear. Equipment that runs very unevenly through the year is better maintained by use — hours, cycles, or kilometers — because the calendar doesn't describe what's happened to it. Equipment that runs steadily, or whose deterioration depends more on the environment than on use, works fine by time and is much simpler to plan.
How often should a piece of equipment be checked?
The manufacturer's catalog is the starting point and tends to be conservative; the regulation rules when it exists; and your history is the only thing that lets you adjust. Without the third one, the plan overshoots in one place and falls short in another.
How do I know if my interval is correct?
By looking at where failures fall within the interval. If they cluster right before the check, it's too long. If the check finds nothing for years, it's probably too short.
Better by time or by use?
By use when wear is driven by work: operating hours, cycles, or kilometers. By time when it's driven by aging or when a regulation requires it.
Does maintenance extend a piece of equipment's life?
Real service life depends far more on how something's been used and maintained than on what the catalog says. Two identical machines can last very differently, and that difference is what you're paying for when you maintain them.
Review an interval
Bring a piece of equipment that's giving you trouble and we'll look at whether its schedule is the right one.