Lean teams use TPM because unreliable equipment breaks flow and forces extra inventory. Stable machines make takt time, standard work and just-in-time possible. TPM progress is often tracked with OEE and the six big losses.
Where it comes from
TPM grew in Japan from preventive maintenance practices that originated in the United States. The Japan Institute of Plant Maintenance and Seiichi Nakajima did much to define and spread it.
At a glance
The eight pillars of TPM
Swipe sideways to see the whole diagram.
How to apply Total Productive Maintenance
- 1
Pick a pilot machine
Start with one important, troublesome asset. A focused pilot shows the method before you spread it across the plant.
- 2
Measure the baseline
Track OEE and log downtime by cause. Group losses into the six big losses so you know where to focus.
- 3
Clean and inspect to restore basic condition
Operators and maintenance clean the machine together and tag every defect they find. Fix leaks, loose parts and contamination sources.
- 4
Set up autonomous maintenance
Create short daily checklists for operators covering cleaning, inspection and lubrication. Use visual marks so normal and abnormal conditions are obvious.
- 5
Build planned maintenance
Schedule preventive tasks based on run time or condition. Track work orders, spare parts and failure history in one place.
- 6
Attack chronic losses
Use focused improvement teams and root cause analysis on the biggest recurring losses. Feed what you learn into equipment design and purchasing.
Worked example (illustrative)
A plastics plant chooses its least reliable molding press for a TPM pilot, with an OEE of about 58% and around 14 minor stops a shift. Operators and technicians clean it together, tag 12 defects and close 10 of them within two weeks. They add a five-minute start-of-shift check. After two months, minor stops are down to about 5 a shift, OEE is near 68%, and the checklist becomes the template for other presses.
Common mistakes
- Treating TPM as a maintenance department program without operator involvement.
- Skipping the initial cleaning and restoration and jumping straight to schedules.
- Rolling out to every machine at once instead of proving the method on a pilot.
When Total Productive Maintenance is not the right tool
TPM is a long-term program, not a quick fix. A plant that needs a fast win on one machine may start with a focused repair and OEE tracking.
Autonomous maintenance also needs time, training and safe procedures for operators before they take on routine care.
What does TPM stand for?
TPM stands for total productive maintenance. Total means everyone takes part, from operators to managers. Productive means maintenance is done while keeping production running where possible. Maintenance means keeping equipment in good condition.
The Japan Institute of Plant Maintenance (JIPM) promoted the approach, and its TPM award gave plants a structured path to follow.
The 8 pillars of TPM
The common model has eight pillars. Autonomous maintenance gives operators daily care of their machines. Planned maintenance schedules work based on failure and wear data. Focused improvement attacks specific losses. Quality maintenance prevents defects caused by equipment conditions.
Early equipment management designs new equipment to be easy to maintain. Training and education builds the skills the other pillars need. Safety, health and environment aims for zero accidents. TPM in office applies the same ideas to support functions. Progress is often tracked with OEE.
The six major losses
TPM targets six major losses: failures, adjustments, minor stoppages, reduced speed, scrap and rework. OEE measures their combined effect.
In the illustrative pilot above, most of the gain came from minor stoppages, one of the six losses. That is common early in TPM, because cleaning and inspection find the loose parts and leaks behind many short stops.
The eight pillars of TPM
TPM is often organized around eight pillars. The Japan Institute of Plant Maintenance notes that names vary between companies, so treat them as one connected system rather than a list to memorize:
Autonomous maintenance
Trained operators do routine care such as cleaning, inspection and lubrication.
Planned maintenance
Maintenance work is planned before failures stop production.
Focused improvement
Small teams find and remove repeat losses.
Quality maintenance
Equipment conditions that affect product quality are defined and kept.
Education and training
People build the skills their assigned tasks need.
Early equipment management
Maintenance experience feeds the design and launch of new equipment.
Office TPM
Administrative work that supports production, such as parts ordering, is improved too.
Safety, health and environment
Hazards and environmental risks are dealt with as part of the system.
The pillars depend on each other: operator inspections need training, findings need a planned response and repeat issues need focused improvement. Judge each one by the work it produces, such as an inspection standard or a fixed defect, not by a presentation.
What TPM can improve
Results depend on how well TPM is run and on local conditions; it does not guarantee a set saving. The usual areas of gain are linked:
Fewer unexpected breakdowns
Daily checks and planned work catch problems earlier. Define what happens after someone reports an abnormality: who reviews it, sets priority, arranges the work and confirms closure.
More from existing capacity
Short stops, slow running, setups, scrap and rework all hide capacity. Look at the bottleneck's losses before asking for a new machine.
Steadier quality
Quality maintenance ties equipment condition to defects. Check the product result, not just that an adjustment was done.
Maintenance spending in context
Judge cost against production losses, not as an isolated budget. NIST research has linked heavy reliance on reactive maintenance with more downtime and defects.
Shared equipment knowledge
Operators, technicians and engineers review one loss together and capture what they learn in short standards.
Separate recovered capacity from cash savings: extra production time is only worth money if it is used to meet demand, avoid overtime or defer an investment.
Safety boundaries for operator care
Giving routine care to operators does not remove maintenance hazards. In the United States, OSHA's hazardous energy standard (lockout/tagout) covers servicing and maintenance where an unexpected startup or release of stored energy could injure someone.
Before any task moves to operators, write a task-specific responsibility matrix:
- Who may do it: Operator, technician or specialist, and the training required.
- Equipment state: Running, stopped or locked out, as decided by qualified safety staff.
- Method: The approved steps, tools and visual standards.
- Stop and escalate: The conditions where the operator must not proceed and who they call.
Never use a productivity target to justify bypassing guards or energy control. Cleaning, inspection, lubrication and tightening each need their own review.
Planning and scheduling maintenance
Keep two jobs apart. Planning decides what work is needed, how to do it, which skills and which parts. Scheduling sets when it happens and matches it with people, parts and a production window.
For every planned task, ask what failure it prevents or detects and what happens after a bad finding. Change intervals only through an approved review, not because a task is inconvenient. After a restore, separate getting the line running again from preventing the failure coming back, and record the cause in a form the next shift can use.
A TPM scorecard
OEE multiplies availability, performance and quality, so it separates stopped time, slow running and bad output. With illustrative factors of 90%, 95% and 98%, OEE is 0.90 × 0.95 × 0.98 = 83.79%. Pair it with measures that show whether the practices actually happen:
| Measure | Review question | What to count |
|---|---|---|
| Unplanned downtime | Is the target interruption less frequent or shorter? | Stops and minutes by cause |
| Repeat failures | Does the same failure return after repair? | Failures with the same cause within an agreed period |
| Equipment-related defects | Has the target quality problem changed? | Defects traced to equipment condition |
| Planned work completion | Was approved maintenance done on time? | Tasks done on schedule ÷ tasks due |
| Finding closure | Were tagged defects fixed, checked or formally decided? | Open findings by age |
| Maintenance cost per good unit | How did spending move against good output? | Agreed with finance |
| Task skill | Can assigned people do the work correctly? | Observed skill against task criteria |
Unplanned downtime
Is the target interruption less frequent or shorter?
What to count: Stops and minutes by cause
Repeat failures
Does the same failure return after repair?
What to count: Failures with the same cause within an agreed period
Equipment-related defects
Has the target quality problem changed?
What to count: Defects traced to equipment condition
Planned work completion
Was approved maintenance done on time?
What to count: Tasks done on schedule ÷ tasks due
Finding closure
Were tagged defects fixed, checked or formally decided?
What to count: Open findings by age
Maintenance cost per good unit
How did spending move against good output?
What to count: Agreed with finance
Task skill
Can assigned people do the work correctly?
What to count: Observed skill against task criteria
Give each measure an owner, a source and a review frequency. Note changes in product mix, hours or staffing, and do not count the same recovered hours as both extra sales and avoided overtime.
TPM examples by type of loss
These illustrative cases show how to frame a TPM project without promising a result:
Packaging short stops
A line has repeated short stops at a transfer point. The team logs time, product format, location and visible condition at each stop, then separates equipment causes from material, setup or upstream supply.
Machining defect
An intermittent dimensional defect is reviewed by maintenance, production and quality together, testing tooling, material and measurement as well as the machine.
Repairs waiting for parts
Repairs keep waiting for a part the stock record says is there. Office TPM looks at part descriptions, storage locations and the request process, not the machine.
TPM training by role
Training should connect ideas with hands-on practice on the actual equipment. Map what each role needs:
- Operators: Spot abnormalities, follow approved care standards and escalate concerns.
- Technicians: Inspect, diagnose, document and improve maintenance on the pilot equipment.
- Supervisors: Protect time for care tasks, review findings and coordinate across shifts.
- Engineers and improvement staff: Investigate losses, check changes and update technical standards.
- Managers: Provide resources, settle competing priorities and review business value.
When a course offers certification, check what is assessed and by whom. Individual training certificates and plant awards, such as JIPM's TPM awards, are different things.
Choosing TPM software
A CMMS can hold asset data, work orders, history and spare parts, but it does not create discipline by itself. Test it with a real pilot workflow: an operator reports an issue, a planner prepares the job, a technician closes it and a supervisor finds the history.
- Ease of reporting: Can users find the asset and describe the problem without extra steps?
- Ownership: Can everyone see who is responsible and what happens next?
- Usable history: Can reviewers link symptoms, actions, parts and repeat work?
- Standards: Can approved instructions be reached and revised under control?
- Administration: Who keeps asset records, permissions and data quality up to date?
TPM mistakes that weaken results
- 1
Moving work to operators without capacity. Check time, training and support first.
- 2
Rewarding completed checklists. Review the quality of findings, responses and verification too.
- 3
Blaming the reporter. Ask for evidence and problem descriptions, not a defense of the number.
- 4
Documentation overload. Keep what supports safe work, diagnosis, learning or a required record.
- 5
Unsupported savings. Require traceable assumptions and separate real cuts from estimates.
- 6
Findings left open forever. Give every tagged issue a decision and an owner.
| Approach | What triggers the work | Who does it |
|---|---|---|
| Preventive maintenance | A calendar or usage interval | Maintenance staff |
| Predictive maintenance | Condition data such as vibration or heat | Maintenance and reliability staff |
| TPM | A plant-wide program across all losses | Operators, maintenance and management |
Guides on this topic
- Muri: how to spot and reduce overburden in leanMuri explained: what overburden means, how it differs from muda and mura, the common causes, an observation checklist, six ways to reduce it, examples from factories to software, and how to measure progress without hiding overload.Read the guide
- How to measure OEE: start tracking without a big IT projectHow to measure OEE with a free paper tracking sheet before buying software: what to record, how to calculate it and when to automate OEE tracking.Read the guide
- One-piece flow: how single-unit flow cuts waiting and WIPOne-piece flow explained: a worked comparison with batches, takt and cycle time, the benefits to check, production and transfer batches, flow and pull, where it fits, the conditions it needs, a pilot plan and how to measure it.Read the guide
- Kaizen 5S: how 5S and kaizen work together in leanKaizen 5S explained: what 5S and kaizen each do, how they support each other in lean management and how to run them as one daily habit.Read the guide
Sources
Books and papers show their publication year. Web pages show the month we last checked them.
- 1.Seiichi Nakajima, Introduction to TPM: Total Productive Maintenance (Productivity Press, 1988)
- 2.Douglas S. Thomas and Brian A. Weiss, NIST, Economics of Manufacturing Machinery Maintenance: A Survey and Analysis of U.S. Costs and Benefits (NIST AMS 100-34) (nist.gov, checked October 2026)