Lean tools

TPM — total productive maintenance

In French: TPM — maintenance productive totale

TPM aims at equipment availability by involving the operators in its upkeep, rather than reserving maintenance for a dedicated department that steps in after the breakdown.

The shift it brings about

Classic maintenance repairs when things break, and prevents according to a calendar. TPM adds a third way: entrusting the simple, daily gestures to those who use the machine — cleaning, visual check, lubrication, retightening.

The reasoning is not about saving labour. It is that the operator is the first to notice an abnormal noise, an incipient leak, a new vibration. They detect several days before the stoppage what no maintenance calendar can anticipate — provided they have been taught what to look at and are listened to when they report it.

Cleaning plays a particular role, which surprises: cleaning a machine is inspecting it. Leaks, play and cracks are discovered with the rag, not with a checklist.

The big losses of a piece of equipment

TPM classifies what prevents a machine from producing value into three families, which are exactly the three components of overall equipment effectiveness.

Stoppages — breakdowns, changeovers, waiting for supplies. They can be seen and counted; they are the ones everyone talks about.

Slowdowns — minor stoppages and reduced-speed running. They are almost invisible: nobody reports a thirty-second stoppage, and two hundred of them happen a day. It is the most underestimated loss in a workshop.

Defects — scrap and rework, including the parts lost at start-up after every restart. A machine producing non-conforming parts is running without producing anything useful.

What it really requires

Time first: autonomous maintenance assumes daily minutes made available at the station. Adding them without taking them from anywhere amounts to asking people to hold the rate on top, and the ritual disappears within three weeks.

Training next: recognising the onset of wear is learnt, and an untrained operator who opens a housing does more harm than good.

A reporting circuit, finally. An operator’s report that leads to nothing is not repeated, and that is the most common failure mechanism of a TPM roll-out.

The common mistakes

Deploying autonomous maintenance without restoring the starting condition of the machines. Handing the daily upkeep of an already degraded piece of equipment to the operators makes them carry the failure of a restoration that never took place — and discredits the approach for a long time.

Tracking overall equipment effectiveness as a team objective. It is a diagnostic indicator: it serves to know where the losses are. Turned into a grade, it pushes people to produce on machines that should be stopped.

On the shop floor — Plastics processing

A record of minor stoppages over two weeks totalled ninety minutes a day on one press, in incidents of under a minute that no tracking system logged. The main cause — a badly adjusted chute guide — was dealt with in an hour.

The other tools illustrated by this trade: plastics and chemicals.

Frequently asked questions

What is the difference between TPM and preventive maintenance?

Preventive maintenance schedules interventions by a specialised department according to a calendar or an hour counter. TPM encompasses it and adds autonomous maintenance — the daily gestures entrusted to the operators — as well as work on the causes of degradation. One is an organisation of interventions, the other an organisation of work.

Is TPM reserved for heavy industry?

No. The principle applies to any equipment whose unavailability costs: medical equipment, IT, vehicles, laboratory machines. What changes is the nature of the daily gestures, not the logic of entrusting them to the user.

Where should a roll-out start?

With one machine, not with the workshop. The equipment is restored, the points to check are defined with the operators, the time is made available, and the losses are tracked for a few weeks. A first piece of equipment that holds is more convincing than ten simultaneous projects that run out of steam.

How is the result measured?

By overall equipment effectiveness, broken down into its three losses — it is the breakdown that informs, not the overall figure. A stable OEE that hides a shift from stoppages to slowdowns has improved nothing.