"Lean maintenance" sounds like a euphemism for cutting headcount, and that misunderstanding is why most lean-maintenance initiatives die in their first quarter. Lean here does not mean thin or cheap. It means the Toyota Production System meaning — relentless identification and elimination of waste, so that every hour of effort adds value instead of feeding friction. Applied to maintenance, lean is not about doing maintenance with fewer people. It is about stopping the maintenance team from spending most of its day on things that are not actually maintenance.
Walk into most maintenance shops and measure where the hours go, and the result is uncomfortable. Technicians spend a large fraction of their shift waiting — waiting for a permit, waiting for parts, waiting for the line to be handed over, waiting for a supervisor, hunting for a tool. The time the wrench is actually turning is often under a third of the paid shift. Lean maintenance is the discipline that reclaims the other two-thirds.
The Seven Wastes, in Maintenance Language
The classic seven wastes of lean (muda) translate directly to maintenance work.
- Overproduction — doing maintenance that adds no value. PMs performed on a calendar regardless of condition, on assets that do not need them. Tear-downs "just in case." (This is the same over-maintain trap described in our run-to-failure piece.)
- Waiting — the big one. Technicians idle for permits, parts, isolation, access, instructions, a signature. This is where most maintenance time is lost.
- Transport — travelling to fetch parts, tools, or information that should be at the point of work.
- Over-processing — extra approval steps, duplicate paperwork, sign-offs that add no control value.
- Inventory — excess spares tying up cash and shelf space, obsolete parts never purged.
- Motion — the technician walking back and forth because tools and parts are not laid out at the job. This is what 5S fixes.
- Defects — rework. Jobs redone because the diagnosis was wrong, the part was wrong, or the work was not done to standard the first time.
Naming these as waste, not as "just how maintenance is," is the first move. The second is measuring them.
Value-Stream Mapping a Work Order
The lean tool that pays off fastest in maintenance is value-stream mapping applied to a work order. Take one job class — say, a corrective repair — and trace every step from request to close, timing each: request raised, triaged, approved, planned, scheduled, parts kitted, permit issued, line handed over, executed, closed out.
At each step, ask: how long did this take, and how much of that time was the work actually being done? The answers are usually shocking. A "four-hour job" routinely takes three days elapsed, of which ninety minutes is wrench time. The rest is waiting, motion, and over-processing. The value-stream map makes the delay visible, and once it is visible it can be attacked: kit the parts ahead of time (cuts waiting), pre-approve a standard permit class (cuts waiting), stage tools at the point of work (cuts motion).
The point is not to blame anyone — the delays are systemic, not personal — but to redesign the flow so the wrench-to-clock ratio stops being embarrassing.
Pull, Not Push
A second lean principle is pull over push. Push maintenance is calendar-driven: generate a hundred PMs this week because the schedule says so, regardless of condition, and watch the team either rush through them (PM compliance theatre — ticked and not actually done) or let them stack into a backlog. Pull maintenance is condition-driven: do the work when the asset signals it needs it, informed by inspection and condition monitoring. The mature state moves as much work as possible from calendar push to condition pull, which both cuts the overproduction waste and raises the quality of the work that does get done.
Standardized Work and Continuous Improvement
Two lean foundations underpin everything above. Standardized work means every recurring job has a defined, written, improved method — the same job done the same right way every time, not re-invented by each technician. This is what checklists, standard job plans, and 5S provide; without it there is no baseline to improve from. Continuous improvement (kaizen) means the standardized work is never final — every repetition is a chance to find a faster, safer, more reliable way, and that improvement gets folded back into the standard. (We cover the cultural side of this in continuous improvement and TPM.)
Lean maintenance and Total Productive Maintenance are deeply linked — TPM is in many ways lean applied to the asset function, with autonomous maintenance and operator ownership as the engine. (See what TPM is.)
The Metrics That Tell You It's Working
- Wrench time / tool time — the percentage of the shift the technician is actually doing the work. The single most honest measure of how much waste is in the flow. Low wrench time is the signature of a non-lean shop.
- Schedule compliance — of the work planned for the week, what fraction was actually executed. Measures whether planning and scheduling are realistic.
- Planned maintenance ratio — the fraction of work that is planned in advance vs reactive. Mature shops run 80%+ planned; the rest is emergency. (Reactive firefighting is the opposite of lean.)
- Mean time to repair — falling MTTR means the support flow (parts, permits, tools) is improving, not just the wrench speed. (See our guide to reducing MTTR.)
Track these and you have a dashboard for waste elimination. Do not track them and lean stays a slogan.
What Kills Lean Maintenance
The failure modes are consistent. Using "lean" as cover for headcount cuts — the moment the team experiences lean as "do more with fewer people," they will quietly starve it of the information it needs and wait for it to pass. One-off campaigns — a lean "blitz week" that produces posters and a cleaned workshop, then reverts. No measurement — without wrench-time and schedule-compliance numbers, there is no way to see whether the effort is working, so it withers. Owned by a manager, not the floor — the technicians who do the work have to own the improvement of the work, or it does not stick.
How OpexMX Supports It
OpexMX gives lean maintenance its measurement layer: wrench-time and schedule-compliance reporting from work-order timestamps, a planned-to-reactive ratio that updates live, kitted-parts staging tied to scheduled jobs (cutting the waiting waste), standardised job plans attached to recurring PMs, and a mobile checklist flow that cuts the motion and over-processing wastes at the point of work. The tooling does not make a shop lean — the discipline does — but it makes the discipline measurable, and what gets measured gets improved.