Total Productive Maintenance: The 8 Pillars, KPIs, and Implementation Plan
Total productive maintenance (TPM) is a company-wide operating system that engages production, maintenance, quality, engineering, safety, and support teams in preventing equipment losses.

What is total productive maintenance?
Total productive maintenance (TPM) is a management system for improving production-system effectiveness by preventing equipment-related loss across the asset life cycle. The Japan Institute of Plant Maintenance introduced TPM in 1971 and describes it as participation across functions and levels, not a maintenance-department project. The practical aim is to make abnormalities visible, restore basic equipment conditions, remove recurring losses, improve skills, and design future equipment and processes for reliable operation.
The eight pillars of TPM
The pillars work as one system. Autonomous maintenance gives operators clear daily ownership without turning them into maintenance technicians. Planned maintenance controls technical work. Focused improvement removes chronic losses, while quality, early equipment management, training, safety, and administrative TPM prevent the same losses from being designed or managed back into the operation.
| Pillar | Operating example | Evidence in the maintenance system |
|---|---|---|
| Focused improvement | Cross-functional team removes a recurring minor stop | Loss code, root cause, action, before/after result |
| Autonomous maintenance | Operators clean, inspect, lubricate, and tag abnormalities | Operator checklist, finding, escalation work order |
| Planned maintenance | Maintenance schedules work from failure modes and condition | PM plan, labor, parts, meter readings, history |
| Quality maintenance | Control equipment conditions that create defects | Quality limit, inspection result, corrective action |
| Early equipment management | Reliability and maintainability are reviewed before acceptance | Commissioning punch list, standard job plan, spare-parts plan |
| Education and training | Skill gaps are mapped to equipment tasks | Qualification, training record, procedure revision |
| Safety, health, and environment | Hazards and environmental losses are designed out | Risk control, inspection, incident follow-up |
| Office TPM | Planning, purchasing, and stores remove delays and errors | Approval time, stockout, vendor, and schedule data |
A phased 30/60/90-day TPM implementation plan
A 90-day plan should prove the operating model on one pilot area, not claim company-wide TPM maturity. Choose a constrained line or asset family with visible loss, stable leadership support, and enough data to establish a baseline. Expand only after roles, standards, escalation, and review routines work in practice.
| Phase | Primary work | Exit evidence |
|---|---|---|
| Days 1-30: Baseline | Name the sponsor and pilot; map losses; restore basic conditions; define safety boundaries; baseline OEE, downtime, MTBF, MTTR, and planned work | Signed charter, asset list, loss Pareto, red-tag backlog, baseline scorecard |
| Days 31-60: Standardize | Launch operator checks; revise PM tasks; train operators and technicians; create escalation rules; run one focused-improvement event | Used checklists, completed work orders, skill matrix, first verified countermeasure |
| Days 61-90: Control | Audit standards; close recurring-loss actions; review KPI movement; stabilize parts and job plans; decide whether to expand | Audit result, sustained action ownership, updated standards, expansion decision |
TPM KPIs and calculations
Use a small, balanced scorecard. OEE summarizes productive use of planned production time, but it can hide the source of loss, so retain the availability, performance, and quality components. Maintenance metrics explain whether equipment reliability and restoration are changing. Review hours and event counts behind the percentages so a small denominator does not create a misleading trend.
| Metric | Formula | Interpretation |
|---|---|---|
| Availability | Run time / planned production time | Loss from breakdowns and setup or adjustment stops |
| Performance | Ideal cycle time × total count / run time | Loss from speed reduction and small stops |
| Quality | Good count / total count | Loss from defects and startup scrap |
| OEE | Availability × performance × quality | Share of planned production time that is fully productive |
| Planned maintenance % | Planned maintenance hours / total maintenance hours × 100 | How much maintenance labor is controlled in advance |
| MTBF | Operating time / functional failures | Average operating exposure between failures |
| MTTR | Total restoration time / completed repairs | Average time required to restore service |
| Downtime rate | Downtime / planned production time × 100 | Share of planned time unavailable |
TPM audit worksheet
Score each statement from 0 to 3: 0 means not defined, 1 means documented but inconsistent, 2 means routinely followed, and 3 means verified and improving. Require evidence for every score. A high self-rating without current records is a discussion prompt, not an audit result.
| Audit question | Evidence to inspect |
|---|---|
| Are pilot losses defined and ranked? | Loss tree, Pareto, agreed definitions |
| Are equipment basic conditions restored? | Red tags, leaks, guards, cleaning and lubrication standards |
| Do operators complete safe autonomous tasks? | Used checks, abnormalities, escalations |
| Do PM tasks address known failure modes? | Failure-mode link, procedure, interval rationale |
| Are defects linked to equipment conditions? | Quality limits, defect code, equipment finding |
| Are skills matched to assigned work? | Skill matrix, qualification, observed performance |
| Are safety boundaries explicit? | Lockout rules, permitted operator tasks, risk review |
| Are recurring losses assigned to an owner? | Action log, due date, verification |
| Are parts and job plans ready before planned work? | Kitted parts, job plan, schedule attainment |
| Does the weekly review change standards or priorities? | Minutes, revised task, closed-loop action |
How work orders and asset history support TPM
TPM depends on evidence moving across teams. Operator findings should create traceable work rather than disappear on a board. Work orders preserve the symptom, loss category, asset, priority, labor, parts, cause, repair, downtime, and verification. Asset history then shows recurring failures, interval performance, restoration time, and the effect of countermeasures. A CMMS does not create participation, but it gives the pilot one shared operational record and prevents each pillar from maintaining a separate spreadsheet.
TPM vs preventive, RCM, and lean maintenance
These approaches overlap but answer different questions. TPM is the broad participation and loss-elimination system. Preventive maintenance is a work strategy inside it. Reliability-centered maintenance selects appropriate policies by functions, failures, consequences, and technically effective tasks. Lean maintenance applies lean thinking to maintenance flow and waste. A plant can use RCM to redesign the maintenance program, execute selected PM and condition tasks through a CMMS, and sustain the routines through TPM.
| Approach | Primary question | Typical scope | Distinctive feature |
|---|---|---|---|
| TPM | How do all functions prevent and remove production losses? | Organization and operating system | Operator ownership and cross-functional pillars |
| Preventive maintenance | What planned work should occur before failure? | Tasks and schedules | Time, usage, condition, or risk trigger |
| Reliability-centered maintenance | What policy preserves required functions at acceptable risk? | Asset systems and failure modes | Consequence-based task selection |
| Lean maintenance | How can maintenance flow create more value with less waste? | Planning, execution, materials, information | Queue, handoff, motion, waiting, and rework reduction |
Common TPM implementation failures
TPM stalls when it becomes a cleaning campaign, a maintenance-only project, or an OEE reporting exercise. Other warning signs include unsafe transfer of technical tasks to operators, too many pilot assets, scores without evidence, unresolved red tags, PMs copied from manuals without failure-mode logic, and improvement events whose gains never become standard work. Leadership must protect time for the routines and remove cross-functional barriers that the maintenance team cannot solve alone.
TPM audit and implementation downloads
Use the audit to establish a pilot baseline and the 90-day checklist to keep owners, evidence, and review dates visible.
Frequently asked questions
What does TPM stand for?
TPM stands for total productive maintenance. It is a cross-functional management system for preventing equipment and production losses across the asset life cycle.
What are the eight pillars of total productive maintenance?
Focused improvement, autonomous maintenance, planned maintenance, quality maintenance, early equipment management, education and training, safety-health-environment, and office TPM are the classic eight pillars.
Is autonomous maintenance the same as operators repairing machines?
No. Autonomous maintenance gives trained operators defined, safe tasks such as cleaning, inspection, lubrication, basic adjustment, and abnormality tagging. Technical repairs and safety-critical work remain with qualified personnel.
How is OEE calculated?
OEE equals availability multiplied by performance multiplied by quality. Use decimals or percentages consistently, and review the three components separately to locate the loss.
Can a company implement TPM in 90 days?
A company can establish and evaluate one TPM pilot in 90 days. Company-wide behavior, standards, and improvement routines take longer and should expand only after the pilot is stable.
Does TPM require CMMS software?
No, but a CMMS helps preserve work orders, asset history, operator findings, PM schedules, failure codes, labor, parts, and KPI inputs in one shared record.
Sources and further reading
Turn TPM findings into traceable maintenance work
Keep operator findings, planned work, causes, downtime, labor, and countermeasures connected to each asset while your TPM routines mature.
Standardizing your reporting? Use the maintenance KPI reference to define MTTR, PM compliance, and cost per work order before comparing periods.
CMMS software and comparisons
Related terms
Preventive Maintenance: Complete Guide, Schedule, Examples, and Templates
Preventive maintenance is planned work performed before functional failure, triggered by time, usage, condition findings, or asset risk to reduce avoidable downtime and preserve equipment performance.
What Is a CMMS? Meaning, Systems & Buyer's Guide
Maintenance management systems centralize work orders, assets, inspections, and preventive schedules so teams can plan, execute, and measure maintenance in one system of record.
MTBF (Mean Time Between Failures)
The average operating time between one failure and the next for a repairable asset — a core reliability metric maintenance teams track to spot bad actors.
MTTR (Mean Time To Repair)
The average time it takes to restore a failed asset to working order — a key maintainability metric that captures how fast your team responds.
Root Cause Analysis (RCA)
A structured method for finding why a failure happened — not just what broke — so the same problem can be eliminated instead of repeatedly repaired.
Condition-Based Maintenance
Servicing equipment when a monitored condition (temperature, vibration, pressure) crosses a defined threshold — not on a fixed schedule.