TPM, Lean Maintenance and Human Factors

Introduction

Maintenance has traditionally been viewed as a technical function concerned with repairing equipment after failure. Modern manufacturing requires a broader view. Equipment performance depends not only on engineering design and maintenance technology, but also on operating practices, workplace organisation, skills, communication, leadership and the quality of decisions made under pressure.

Total Productive Maintenance (TPM), lean maintenance and human-factors engineering provide complementary approaches to this challenge. TPM develops shared ownership of equipment performance. Lean maintenance aligns maintenance work with customer value and removes avoidable waste. Human-factors engineering designs work systems that help people perform safely and reliably.

The strategic purpose is not simply to increase machine utilisation. It is to achieve reliable, safe and economical production while protecting people, product quality and the environment.

Learning Objectives

By the end of this chapter, a student should be able to:

1. Explain the principles, objectives and pillars of TPM.
2. Design a practical autonomous-maintenance programme.
3. Apply 5S, Kaizen and lean-waste concepts to maintenance processes.
4. Evaluate maintenance performance using appropriate indicators.
5. Explain how competence, ergonomics, communication and error-prevention methods influence reliability.
6. Develop controls for contractor safety and maintenance quality.
7. Recommend actions for building an organisational reliability culture.

1. Total Productive Maintenance

1.1 Meaning and development

TPM is an organisation-wide system for maximising the effectiveness of equipment throughout its life cycle. It integrates production, maintenance, engineering, quality, management and support functions. Operators are responsible for basic equipment care, while maintenance specialists concentrate on technical tasks, improvement and restoration of equipment capability.

The word “total” has three related meanings:

• total participation by all departments and levels;
• total equipment effectiveness, including availability, performance and quality; and
• total equipment life-cycle involvement, from design through disposal.

TPM is not merely a preventive-maintenance schedule, a housekeeping programme or a requirement that operators perform maintenance trades. It is a management system that develops ownership, standardisation, defect prevention and continuous improvement.

1.2 TPM objectives

Typical TPM objectives include:

• eliminating breakdowns and recurring minor stoppages;
• reducing equipment-related quality defects;
• preventing accidents and environmental releases;
• improving maintainability and operability;
• reducing the time required for setup, inspection and repair;
• developing operator and maintenance competence; and
• increasing the useful life and economic value of assets.

A useful measure is Overall Equipment Effectiveness (OEE). For a defined production period:

where A is availability, P is performance efficiency and Q is quality rate. Each factor is expressed as a decimal between 0 and 1, or as a percentage when multiplied by 100. The factors must be defined over the same period and for the same equipment boundary.

A common formulation is:

Here, operating time excludes downtime during the planned production period; ideal cycle time is the theoretically shortest cycle for one unit; and all time and production quantities must use consistent definitions. OEE is useful for identifying losses, but it is not a complete measure of business performance. A machine can have high OEE while producing the wrong product, excessive inventory or unsafe work.

Worked example: OEE

A machine is scheduled for 480 minutes. It experiences 60 minutes of downtime. Its ideal cycle time is 0.5 minutes per unit, and it produces 700 units, of which 672 are good.

Operating time = 480 − 60 = 420 minutes.

Therefore,

The OEE is approximately 70.0%. The largest opportunity is not identified by OEE alone; the loss tree shows that planned-time losses, performance losses and quality losses should be investigated separately.

2. The TPM framework

TPM is commonly represented by a set of interdependent pillars. Organisations may name or combine them differently, but the following structure is widely applicable.

2.1 Focused improvement

Focused improvement uses cross-functional teams to remove chronic losses. Teams define the problem, establish a baseline, analyse causes, test countermeasures and verify sustained results. Tools may include Pareto analysis, five-whys analysis, cause-and-effect diagrams, fault-tree analysis and small-group problem solving.

A “five-whys” investigation should not stop at a human action such as “the operator did not tighten the bolt.” It should ask why the error was possible: Was the specification unavailable? Was the tool unsuitable? Was training inadequate? Was the design difficult to assemble? The goal is to improve the system rather than assign blame.

2.2 Autonomous maintenance

Autonomous maintenance transfers selected routine-care activities to trained operators. It normally develops through stages:

1. initial cleaning and inspection;
2. elimination of contamination sources and difficult-to-access points;
3. establishment of lubrication, tightening and cleaning standards;
4. general inspection and skill development;
5. autonomous inspection and condition recognition;
6. workplace standardisation; and
7. continuous self-management and improvement.

Operators do not replace qualified maintainers. They perform defined, safe tasks such as cleaning, visual inspection, lubrication where authorised, detection of abnormalities and restoration of basic conditions. Isolation of hazardous energy, electrical diagnosis, pressure-system work and other specialised tasks remain controlled activities for competent and authorised personnel.

2.3 Planned maintenance

Planned maintenance combines preventive, predictive, corrective and improvement work. The plan should be based on failure consequences, failure behaviour, equipment criticality, manufacturer information, operating conditions and evidence from inspection.

Time-based replacement is appropriate when failure probability increases predictably with age or when consequences justify replacement. Condition-based maintenance is appropriate when a measurable condition gives sufficient warning of functional failure. Run-to-failure may be reasonable for non-critical, inexpensive and easily replaceable items, provided safety and secondary damage are not significant.

2.4 Quality maintenance

Quality maintenance aims to prevent conditions that create defects. It links equipment parameters to product characteristics and establishes controls for settings, calibration, contamination, tool wear and process drift. The emphasis is prevention at the source rather than sorting defective output after production.

2.5 Early equipment management

Maintenance and operations personnel should provide design feedback for new or modified equipment. Important criteria include accessibility, cleanability, inspection visibility, maintainability, spare-parts standardisation, safe isolation, diagnostic capability and energy efficiency. Designing out maintenance difficulty is generally more effective than compensating for it later.

2.6 Training and education

Competence development must cover technical knowledge, procedural discipline, hazard recognition, communication and decision-making. A competence matrix identifies the skills required for each role, the current level of each person and the evidence needed to authorise work. Attendance at training is not proof of competence; observation, demonstration, assessment and supervised experience provide stronger evidence.

2.7 Safety, health and environment

TPM activities must never encourage production pressure to override legal or organisational controls. Risk assessment, lockout/tagout or equivalent energy-isolation procedures, machine guarding, confined-space controls, chemical management, permit-to-work systems and emergency arrangements must be integrated into the programme.

2.8 Office TPM

Planning, purchasing, stores, work-order administration and engineering documentation can create maintenance losses. Office TPM reduces delays caused by inaccurate bills of material, unavailable permits, late purchase orders, poor technical records and ineffective scheduling.

3. Lean maintenance

Lean maintenance applies lean thinking to the maintenance value stream. Value is defined by the customer and by the safe, reliable function required from the asset. Work that consumes resources without improving asset function, safety or necessary control is examined as potential waste.

The principal wastes in maintenance include:

Lean does not mean minimising maintenance at all costs. Deferring essential work can increase risk, total cost and downtime. Lean decisions must consider the whole life-cycle consequence of an action.

3.1 Maintenance value-stream mapping

A value-stream map follows a work request from identification through planning, approval, scheduling, execution, testing and closure. It distinguishes processing time from waiting time and exposes information delays. A map may reveal that a two-hour repair remains open for three weeks because of permit queues, planning backlogs or missing materials.

Useful questions include:

• How is the defect detected and communicated?
• Who decides priority?
• What information is required before planning?
• Are parts, tools and drawings available at the job site?
• Is the asset returned to service using an agreed test?
• Is the failure history accurately recorded?

3.2 5S in maintenance

5S is a workplace-management method:

1. Sort—remove unnecessary items.
2. Set in order—locate tools and materials so their use is obvious.
3. Shine—clean while inspecting for abnormal conditions.
4. Standardise—establish visual and procedural standards.
5. Sustain—audit, learn and improve the system.

5S is not cosmetic housekeeping. A clean and ordered workplace makes leaks, loose fasteners, abnormal vibration, missing tools and unsafe conditions more visible. Standards should specify ownership, frequency, acceptance criteria and response to abnormalities.

3.3 Kaizen

Kaizen means continuous, incremental improvement. A sound Kaizen cycle is:

1. select a problem based on evidence;
2. define the current condition;
3. set a measurable target;
4. identify causes;
5. develop and test countermeasures;
6. check results;
7. standardise successful changes; and
8. review for further improvement.

Kaizen should be distinguished from major capital projects. Small improvements may involve relocating a lubrication point, adding a visual indicator, changing a fastener, improving access or revising a checklist. However, modifications affecting safety, regulatory compliance or machine integrity require engineering review and management of change.

4. Human factors and reliability

Human factors concerns the interaction among people, technology, tasks, organisation and environment. Maintenance errors are often symptoms of system conditions rather than isolated personal failings. Examples include ambiguous drawings, similar connectors, poor lighting, interruptions, fatigue, inadequate supervision, weak shift handover and conflicting performance targets.

4.1 Competence and human performance

Competence includes knowledge, practical skill, judgement, communication and the ability to recognise limits. Reliable organisations define critical tasks, specify required competence and verify it periodically. They also provide conditions in which competent people can succeed.

A practical competence system includes:

• role and task analysis;
• training and supervised practice;
• formal authorisation for hazardous work;
• periodic reassessment;
• lessons from incidents and repeat failures; and
• records that are accurate, accessible and privacy-conscious.

4.2 Ergonomics

Ergonomics designs tasks and equipment around human capabilities and limitations. Maintenance design should consider reach, force, posture, visibility, vibration, noise, temperature, manual handling, access and cognitive workload. Good ergonomics reduces musculoskeletal injury and also improves task quality by reducing fatigue and awkward compensating behaviours.

Examples include providing lifting aids, reducing the force needed to remove covers, improving illumination, positioning displays within a comfortable viewing range and designing connectors that cannot be incorrectly interchanged.

4.3 Communication and shift handover

A maintenance handover should communicate asset status, hazards, isolations, incomplete work, temporary repairs, abnormal observations, permits, tests completed and actions required. Structured formats such as SBAR—situation, background, assessment and recommendation—can reduce omission, although the format must be adapted to the industrial context.

Handover is a two-way verification process, not merely a written message. The receiving person should confirm understanding, particularly for safety-critical conditions.

4.4 Error prevention

The hierarchy of error controls favours design solutions over reliance on memory. Examples include:

• elimination: remove the need for a risky step;
• substitution: use a less hazardous material or method;
• engineering control: interlocks, guards, keyed connectors or automatic checks;
• administrative control: procedures, permits, checklists and training; and
• personal protective equipment: the final protective layer, not the sole control.

Poka-yoke, or mistake-proofing, prevents an error or makes it immediately evident. A connector that can fit only one socket is stronger than a warning label instructing the user to choose correctly.

Checklists are effective when they are short, task-specific and used at meaningful control points. They should not become a substitute for competence or a ritual detached from risk.

4.5 Just culture and reporting

A reliability culture encourages reporting of hazards, near misses, weak signals and equipment abnormalities. A just culture distinguishes among human error, at-risk behaviour and reckless behaviour. Honest error should normally lead to learning and system improvement; deliberate disregard of known controls may require accountability. Excessive blame suppresses information and allows latent conditions to persist.

5. Contractor management

Contractors may possess specialised skills, but outsourcing work does not outsource the organisation’s duty to control risk or assure work quality. Contractor management should cover:

1. prequalification based on competence, safety performance and resources;
2. clear scope, technical standards and responsibilities;
3. site induction and task-specific risk assessment;
4. verification of licences, qualifications and fitness for assigned work;
5. permit, isolation and access control;
6. supervision proportionate to risk and contractor familiarity;
7. coordination among simultaneous work groups;
8. inspection, testing and acceptance of completed work; and
9. performance review, learning and corrective action.

Commercial incentives must not encourage unsafe shortcuts or premature closure of work orders. Contracts should define quality evidence, reporting expectations, stop-work authority and requirements for preserving technical knowledge and maintenance records.

6. Reliability culture and governance

Organisational reliability culture is expressed through decisions, resource allocation and everyday behaviour, not slogans. Its characteristics include visible leadership, accurate information, disciplined standards, learning from weak signals, willingness to stop unsafe work and alignment between production and asset objectives.

Leadership should ensure that reliability indicators are balanced. Useful measures include planned-maintenance compliance, schedule compliance, emergency-work proportion, repeat failures, mean time to repair, backlog age, defects after maintenance, near-miss reporting and energy or environmental performance. Metrics can be distorted: for example, a high schedule-compliance percentage may result from excluding difficult work, while a low number of reported incidents may indicate under-reporting rather than safety.

Mean Time Between Failures (MTBF) is often calculated as:

For repairable equipment, MTBF is an observed average over a defined population and period; it is not automatically the same as a probability distribution parameter or a guarantee of future performance. Mean Time To Repair is commonly:

Definitions must specify whether waiting for parts, permits and personnel is included. Without consistent boundaries, comparisons are misleading.

Governance also requires management of change. Changes to software, control logic, equipment settings, materials, procedures or staffing can alter risk. The change should be assessed, approved, documented, communicated and reviewed after implementation.

7. Digital technologies and AI

Computerised maintenance-management systems support work requests, planning, parts control, history and performance analysis. Sensors and condition-monitoring systems can provide vibration, temperature, pressure, current or oil information. Analytics may help detect patterns and prioritise inspection.

Artificial intelligence can assist with anomaly detection, image-based inspection, work-order classification and knowledge retrieval. Its output should be treated as decision support rather than unquestionable diagnosis. Important controls include data quality checks, human review, cybersecurity, model-performance monitoring, protection of personal information and clear responsibility for the final decision. A predictive model trained on biased or incomplete failure data may produce false alarms or miss rare high-consequence failures.

Technology should be introduced only when it improves a defined decision. A sensor that generates data without a response process creates information waste.

8. Illustrative case: TPM in a packaging line

Background

A fictional food-packaging plant operates a filling and sealing line. The line experiences frequent short stops, seal defects and emergency maintenance. Operators clean the machine at the end of each shift, but practices differ between crews. Technicians report that many work orders lack failure descriptions and that changeover tools are often missing.

Problem analysis

A four-week loss study identifies three dominant losses: product accumulation at a sensor, incorrect sealing temperature after changeover and delays waiting for tools. Investigation finds contamination around the sensor, no standard for sensor cleaning, inconsistent temperature-setting records and poor organisation of the changeover trolley. The failures are not caused by a single careless individual; the work system makes variation likely.

Approach

A cross-functional team applies the following actions:

• performs an initial clean-and-inspect exercise;
• marks contamination sources and installs a simple shield;
• creates an operator inspection standard with photographs;
• establishes a controlled changeover parameter sheet;
• colour-codes and shadows tools on a dedicated trolley;
• trains operators to identify abnormal conditions and escalate safely;
• reserves technical diagnosis and energy isolation for authorised technicians;
• introduces a structured shift handover; and
• reviews sensor alarms and repeat work weekly.

The team uses baseline data, verifies the modifications through a controlled trial and updates the standard operating procedures. A maintenance engineer approves the sensor shield to ensure it does not create a hygiene or guarding problem.

Outcome

After eight weeks, short stops decline, seal defects become less frequent and changeover preparation is faster. The improvement is sustained because responsibilities, inspection criteria and escalation routes are explicit. The team continues monitoring rather than declaring permanent success from a short-term result.

Learning points

The case illustrates that TPM combines operator ownership with specialist maintenance, while lean methods remove waiting and motion. Human-factors methods improve the design of tasks and information. Most importantly, performance improvement comes from controlling causes and standardising successful practice, not from pressuring individuals to work faster.

Challenges and limitations

TPM and lean maintenance can fail when implemented as slogans, audit scores or isolated housekeeping projects. Common limitations include:

• copying another organisation’s pillars without adapting them to risk and technology;
• transferring tasks to operators without competence, time or authority;
• pursuing OEE at the expense of safety, quality or customer demand;
• treating all failures as preventable through more inspection;
• excessive paperwork that obscures critical controls;
• weak leadership support during production pressure;
• poor data definitions and inaccurate failure coding; and
• failure to involve contractors, engineers and supply-chain personnel.

A mature programme recognises trade-offs. More maintenance is not always better maintenance; the correct decision depends on risk, failure behaviour, consequence, cost and available control options.

Best-practice implementation sequence

A practical implementation sequence is:

1. establish leadership intent and non-negotiable safety requirements;
2. select a pilot area using criticality and improvement potential;
3. measure current losses with agreed definitions;
4. restore basic equipment conditions;
5. train and involve operators, maintainers and supervisors;
6. implement 5S, autonomous-maintenance standards and planned work;
7. analyse chronic losses and verify countermeasures;
8. improve planning, parts, data and contractor controls;
9. audit outcomes and behaviours rather than paperwork alone; and
10. replicate only after the pilot is stable and lessons are understood.

Chapter Summary

TPM is an organisation-wide approach to maximising equipment effectiveness through total participation, life-cycle thinking and continuous improvement. Its pillars include autonomous maintenance, planned maintenance, focused improvement, quality maintenance, early equipment management, training, safety and office support. OEE combines availability, performance and quality, but must be interpreted with a clear loss analysis and balanced business measures.

Lean maintenance removes waste from maintenance value streams without encouraging unsafe or premature deferral of essential work. 5S makes abnormal conditions visible, while Kaizen provides a disciplined method for incremental improvement. Human factors show that reliability depends on task design, ergonomics, competence, communication and error-tolerant systems. Contractor management requires control from selection through acceptance. A strong reliability culture is supported by just reporting, accurate data, leadership commitment, management of change and responsible use of digital technologies.

Key Terms

TPM: An organisation-wide system for maximising equipment effectiveness across its life cycle.

Autonomous maintenance: Trained operator activities for basic care, inspection and abnormality detection.

OEE: A product of availability, performance efficiency and quality rate.

Availability: The proportion of planned production time during which equipment is operating.

Performance efficiency: Output achieved relative to ideal output during operating time.

Quality rate: The proportion of total output that meets requirements.

Focused improvement: Structured, cross-functional elimination of significant losses.

Planned maintenance: Maintenance scheduled using time, condition, risk or known failure behaviour.

Quality maintenance: Prevention of equipment conditions that cause product defects.

Lean maintenance: Application of lean principles to maintenance work and value streams.

5S: Sort, Set in order, Shine, Standardise and Sustain.

Kaizen: Continuous, incremental improvement based on disciplined problem solving.

Human factors: Study of interactions among people, technology, tasks and organisations.

Ergonomics: Design of work to fit human capabilities and limitations.

Poka-yoke: Mistake-proofing that prevents or immediately detects errors.

Just culture: A culture that supports learning while distinguishing error from reckless conduct.

MTBF: Mean operating time between observed failures of repairable equipment.

MTTR: Mean time required for defined corrective repairs.

Management of change: A controlled process for evaluating and implementing changes.

Reliability culture: Shared organisational behaviours and decisions that support dependable and safe performance.

Review Questions

Short Answer Questions

1. What are the three meanings of “total” in TPM?
2. Define autonomous maintenance and state two activities it may include.
3. What are the five steps of 5S?
4. Why is ergonomics relevant to maintenance reliability?
5. State two controls that should be included in contractor management.

Long Answer Questions

1. Explain the main TPM pillars and discuss how they reinforce one another.
2. Describe how a maintenance value-stream map can reveal waste and improve work execution.
3. Explain why OEE is useful but insufficient as a complete measure of operational performance.
4. Discuss the role of competence, communication and just culture in preventing maintenance error.
5. Develop an implementation plan for introducing TPM in a selected manufacturing area.

Analytical / Application Questions

1. A machine is scheduled for 600 minutes, has 75 minutes of downtime, produces 900 units at an ideal cycle time of 0.5 minutes per unit, and produces 855 good units. Calculate availability, performance, quality and OEE. Identify one limitation of interpreting the result.
2. A technician repeatedly installs a filter incorrectly because two similar housings have unclear labels and the work instruction is difficult to access. Analyse the immediate and latent causes, then propose controls using the hierarchy of error prevention.
3. A plant’s contractor has excellent technical skills but repeatedly closes work orders without recording test results. Design a governance and acceptance process that protects safety, knowledge retention and maintenance quality.

Multiple Choice Questions with Answers

Question 1

Which statement best describes TPM?

A. A programme in which only maintenance technicians are responsible for equipment

B. An organisation-wide system for maximising equipment effectiveness throughout its life cycle

C. A method of replacing all preventive maintenance with run-to-failure

D. A financial system for calculating spare-parts value

Correct Answer: B. An organisation-wide system for maximising equipment effectiveness throughout its life cycle

Question 2

In the OEE calculation, which factor represents the proportion of output meeting requirements?

A. Availability

B. Performance efficiency

C. Quality rate

D. Planned production time

Correct Answer: C. Quality rate

Question 3

Which activity is most appropriate for autonomous maintenance when properly authorised and trained?

A. Live electrical fault diagnosis

B. Cleaning and visual inspection for abnormal conditions

C. Modification of a safety interlock

D. Repair of a pressurised vessel

Correct Answer: B. Cleaning and visual inspection for abnormal conditions

Question 4

Which 5S stage establishes visual and procedural requirements so that good conditions are maintained?

A. Sort

B. Set in order

C. Shine

D. Standardise

Correct Answer: D. Standardise

Question 5

A connector designed so that it can fit only the correct socket is an example of:

A. Poka-yoke

B. Overproduction

C. Run-to-failure maintenance

D. Mean time to repair

Correct Answer: A. Poka-yoke

Question 6

A maintenance value-stream map is primarily used to:

A. Calculate the tensile strength of a component

B. Trace work and information flow to identify delay and waste

C. Replace all risk assessments

D. Determine an operator’s employment grade

Correct Answer: B. Trace work and information flow to identify delay and waste

Question 7

Which response best reflects a just culture?

A. Punish every error regardless of circumstances

B. Suppress near-miss reports to protect performance statistics

C. Distinguish human error from reckless disregard and learn from system weaknesses

D. Allow deliberate violation of safety controls without review

Correct Answer: C. Distinguish human error from reckless disregard and learn from system weaknesses

Question 8

A scheduled production period is 480 minutes, downtime is 60 minutes, and all other OEE factors equal 1. What is availability?

A. 0.125

B. 0.500

C. 0.875

D. 1.143

Correct Answer: C. 0.875

Question 9

Which contractor-control activity should occur before hazardous work begins?

A. Deleting the contractor’s work history

B. Verifying competence, risk controls and required permits or isolations

C. Allowing unrestricted access to the plant

D. Accepting the work without testing

Correct Answer: B. Verifying competence, risk controls and required permits or isolations

Question 10

Why should an AI anomaly-detection result normally be treated as decision support rather than an unquestionable diagnosis?

A. AI systems cannot process numerical data

B. Models may be affected by incomplete data and can produce false alarms or missed detections

C. Condition monitoring never provides useful information

D. Human review always reduces reliability

Correct Answer: B. Models may be affected by incomplete data and can produce false alarms or missed detections