Ask yourself a simple question: on your most critical line, how many people can really diagnose a serious fault? Often two. Sometimes one. The exposure is already there, today, long before anyone retires. If that person is off sick on a busy Monday, the line waits.
Mimorian is an industrial intelligence platform that models equipment, structures fault diagnosis and captures the know-how of maintenance teams through a multi-agent AI architecture. Beyond speeding up diagnosis, it makes visible a blind spot nobody measures: how far knowledge is concentrated in a handful of people.
Key-person dependency plays out long before retirement
There is a lot of talk about senior staff leaving and the know-how that goes with them. That is real, but it hides a more immediate risk: knowledge concentrated here and now, in people who are still on site.
On an old or bespoke machine, most of the diagnosis sits in the heads of two or three people. They know its faults, its quirks, the history of its breakdowns. None of that appears in any document. The day one of them is missing, for leave, a transfer or sick leave, part of the plant's capability goes missing too.
Critical know-how left unwritten is an unprovisioned risk
For a management team, this dependency is a risk that appears on no dashboard. Spare parts are provisioned, machines are insured, but the fragility tied to concentrated knowledge goes unmeasured.
The context makes it worse. 89% of manufacturing executives acknowledge a talent shortage, with 2.4 million roles potentially unfilled by 2028 [Deloitte/Manufacturing Institute, 2018]. When an expert leaves, replacing them gets harder, and undocumented knowledge leaves with them. Overall, large organisations lose an average of 47 million dollars a year to ineffective knowledge sharing [Panopto, 2018].
In a sale or an acquisition, this risk surfaces at the worst possible moment: an audit revealing that continuity of operations rests on two people weighs on the valuation. Critical know-how left unwritten stays a risk, measured or not.
Mapping the risk before it turns into a breakdown
You cannot fix what you cannot see. The first step is to make the dependency visible: for each critical machine, who actually carries the experience, and how far that knowledge is concentrated.
By cross-reading the intervention history (who did what on which machine) with equipment criticality, Mimorian draws that map. On a given asset, it becomes possible to see that a single technician carries most of the accumulated experience. The module also shows a skills profile by domain (mechanical, electrical, automation, pneumatic, hydraulic) and flags the areas where nobody reaches the expected level. The risk of a knowledge break stops being a corridor hunch and becomes data that management can steer.
Taking back control, without dispossessing your experts
Seeing the risk is not enough, you need a path to reduce it. And that path does not consist of asking experts to write everything down.
At each diagnosis, the reasoning that was followed is captured with no extra data entry. The know-how of the two or three specialists gradually flows into a base the whole team can reach. Junior staff build their skills by reasoning with support, rather than clicking on answers. Over the months, knowledge of a machine stops resting on two people and becomes the workshop's own.
The expert loses nothing, quite the opposite. Their know-how is finally recognised and kept for the real questions, instead of being called on for the same fault every month.
Conclusion
Key-person dependency is one of the most common and least measured risks in maintenance. Making it visible is already half the work: you then know where to concentrate the effort of passing knowledge on. The rest is built intervention after intervention, as the specialists' knowledge becomes the whole team's. You know what your best technician does. The real question is what happens on the day he or she is not there.
For the full framework of capturing field know-how, read our guide to capturing maintenance know-how.