TPM — Total Productive Maintenance — is one of the best-known approaches to industrial improvement, and one of the most misunderstood. It is often reduced to “getting operators to do a bit of maintenance,” which is both true and very simplistic. TPM is in fact an organisational approach that aims to involve the whole company, from the shop floor to management, in equipment reliability. This article explains it plainly, without jargon or promises, for maintenance and production managers.
Where TPM comes from and what it aims for
TPM emerged in Japan in the 1970s, notably around the automotive industry and the work of the JIPM (Japan Institute of Plant Maintenance). It builds on a simple idea: a machine does not belong to the maintenance department alone. The person who uses it every day — the operator — is best placed to spot a leak, an unusual noise or new play in a part before a breakdown occurs.
TPM’s stated ambition holds in three “zeros”: zero breakdowns, zero defects, zero accidents. These are directions of travel, not guarantees. No plant ever reaches true absolute zero, but aiming for these goals pushes effort toward prevention rather than emergency repair. TPM seeks to maximise the overall effectiveness of the production line by tackling every kind of loss: stoppages, slowdowns, scrap, changeover time.
The main pillars of TPM
TPM is traditionally built around several pillars, resting on a foundation of method and order (often the 5S). Presentations vary, but the following almost always appear:
- Autonomous maintenance: the cultural heart of the approach. Operators take on first-level care of their machine — cleaning, inspection, lubrication, tightening, minor adjustments. Cleaning is inspecting: while cleaning, you discover problems in the making.
- Planned maintenance: the maintenance team organises preventive work methodically — schedule, procedures, parts — so interventions are anticipated rather than endured.
- Continuous improvement (sometimes “focused improvement”): small teams analyse recurring losses and eliminate their root causes, often with simple problem-solving tools.
- Training and skills development: without trained operators and technicians, neither autonomous maintenance nor preventive work survives over time.
- Quality maintenance: linking equipment condition to product quality, to prevent defects at the source rather than sorting them out afterward.
Other pillars often round out the picture: early equipment management (getting reliability right from the design stage), safety and environment, and TPM applied to support functions (offices, logistics). The point is not to tick every box, but to grasp the overall logic.
OEE / TRS, the compass of TPM
To steer the effort, TPM relies on one central metric: OEE (Overall Equipment Effectiveness), known in French as TRS. It measures the gap between what a machine actually produces and what it could produce under ideal conditions.
OEE combines three components:
- availability (is the equipment running when it should, or losing time to breakdowns and changeovers?);
- performance (is it running at its rated speed, or slowing down?);
- quality (are the parts good, or is there scrap and rework?).
OEE is the product of these three rates. Its value lies not in the single figure but in the breakdown of losses it reveals: you finally know whether the problem comes from stoppages, speed or quality. That said, an OEE calculated carelessly, or weaponised as a way to pressure teams, loses all meaning. It is a thermometer, not a goal in itself.
How to start, concretely and without dogma
The classic temptation is to roll out every pillar at once, with Japanese vocabulary and a grand five-year plan. That is the surest way to exhaust your teams. A more honest and more effective approach is to start small:
- Pick a pilot machine — often a bottleneck or a temperamental one — rather than the whole plant.
- Launch autonomous maintenance with the operators: an initial deep clean, then a few simple inspection and lubrication routines, done by the people who run the machine.
- Structure preventive maintenance on that equipment, based on its real breakdown history.
- Set up an easy way to report anomalies, so that what the operator notices does not get lost.
- Measure a few real losses before invoking a full OEE calculation.
One point deserves to be stated clearly: TPM is an organisational and cultural approach, not a piece of software. It rests first on people, hands-on management and consistency. No tool will replace the decision to involve production in machine reliability.
A CMMS therefore will not install TPM for you, but it does support it usefully once the effort is under way. It hosts scheduled preventive maintenance, gives operators a simple channel to report anomalies through requests, and keeps the history of interventions. It also feeds the maintenance KPIs that fuel loss analysis. The CMMS is a support for TPM, never a substitute for the teams’ commitment.
There is nothing magical about TPM: it is a patient way of bringing production and maintenance together around a shared goal — reliability. Start modestly, measure what matters, and let the culture take hold before the metrics do.