Lesson 3: Predict potential failures and implement measures to prevent mechanical defects.

2: Recognise Early Signs of Physical Degradation, Such as Fatigue Cracking, Creep, or Erosion, Through Targeted Non-Destructive Inspections

Early physical degradation is one of the most important indicators of declining mechanical integrity. Mechanical components rarely fail without any preceding change in condition. Fatigue cracks may develop gradually from repeated loading, creep can cause progressive deformation during prolonged exposure to elevated temperature and stress, while erosion can progressively remove material from surfaces exposed to high-velocity fluids or abrasive particles. If these mechanisms are identified early, engineering teams can often intervene before a defect develops into a major component failure, loss of containment, production interruption or extensive equipment damage.

Targeted non-destructive inspection (NDI/NDT) provides a systematic means of detecting, characterising and monitoring degradation without unnecessarily destroying or removing the component from service. The effectiveness of an inspection programme depends on selecting the appropriate technique for the suspected degradation mechanism, understanding the limitations of the inspection method, defining suitable inspection locations, and interpreting results against appropriate acceptance criteria. For mechanical QA/QC professionals, NDT should therefore be treated as an evidence-based integrity-management activity rather than simply a routine testing exercise.

The objective is not merely to identify visible defects. A well-designed inspection programme seeks to detect early indicators, establish whether degradation is active, determine its extent, assess its significance, and provide reliable evidence for decisions such as continued operation, increased monitoring, repair, replacement or further engineering assessment. This makes targeted NDT an essential part of predictive mechanical quality control and reliability management.

Understanding Physical Degradation in Mechanical Components

Physical degradation occurs when the condition or performance of a mechanical component progressively deteriorates during manufacturing, operation, maintenance or service.

The degradation mechanism depends on factors such as:

• Applied mechanical loads.
• Cyclic loading.
• Temperature.
• Pressure.
• Fluid velocity.
• Chemical environment.
• Surface condition.
• Material properties.
• Manufacturing quality.
• Operating duration.
• Maintenance practices.
• Environmental exposure.

Common degradation mechanisms include:

• Fatigue cracking.
• Creep deformation.
• Erosion.
• Corrosion.
• Wear.
• Fretting.
• Thermal damage.
• Surface cracking.
• Material loss.
• Distortion.
• Delamination in relevant materials.

Different degradation mechanisms produce different physical indications. Consequently, the inspection method must be selected according to the type of defect that is reasonably expected.

What Is Targeted Non-Destructive Inspection?

Non-destructive inspection refers to techniques used to examine materials, components or assemblies without causing unacceptable damage to their future serviceability.

Targeted NDT means that inspection resources are directed towards:

• Known high-risk locations.
• Areas with high stress concentration.
• Previous defect locations.
• Welded joints.
• Heat-affected zones.
• High-temperature regions.
• High-velocity flow areas.
• Areas exposed to cyclic loading.
• Locations identified through FMEA.
• Components showing abnormal operating trends.

This targeted approach is more effective than applying identical inspection methods and frequencies to every component regardless of risk.

Key Concepts and Definitions

Recognising Fatigue Cracking

Fatigue cracking develops when a component is subjected to repeated or fluctuating stresses over time. The applied stress may remain below the material’s static strength, yet repeated cycling can initiate and propagate a crack.

Fatigue is particularly important in:

• Rotating shafts.
• Gear components.
• Couplings.
• Welded structures.
• Pressure-containing components.
• Crankshafts.
• Connecting rods.
• Mechanical linkages.
• Components subject to vibration.

The fatigue process can broadly be understood as:

Cyclic loading → crack initiation → crack propagation → reduced section → potential failure

The early stages can be difficult to detect visually because cracks may be small and located in highly stressed areas.

Typical Causes of Fatigue Cracking

Potential fatigue contributors include:

• Repeated loading.
• Vibration.
• Stress concentration.
• Sharp geometric transitions.
• Keyways.
• Weld toes.
• Surface damage.
• Poor surface finish.
• Misalignment.
• Repeated start-stop cycles.
• Torque fluctuations.
• Pressure cycling.

FMEA can help identify these locations as priority inspection zones.

Where to Look for Fatigue Cracks

Targeted inspection should focus on areas where cyclic stresses are concentrated.

Examples include:

• Shaft shoulders.
• Keyways.
• Weld toes.
• Fillet transitions.
• Gear teeth.
• Thread roots.
• Bolt holes.
• Coupling interfaces.
• Abrupt geometry changes.
• Areas with previous repairs.

The inspection location should be based on engineering understanding of the load path and likely crack initiation mechanism.

Non-Destructive Methods for Detecting Fatigue Cracking

Several NDT methods may be appropriate depending on the component and expected defect.

Visual Inspection

Visual examination is often the first inspection stage.

It can identify:

• Surface cracks that are large enough to see.
• Corrosion.
• Surface damage.
• Deformation.
• Wear.
• Leakage.
• Discolouration.
• Distortion.

However, visual inspection has limitations. Very small cracks, subsurface defects and defects hidden beneath coatings may not be detectable.

Liquid Penetrant Testing

Liquid penetrant testing is commonly used to identify surface-breaking discontinuities in suitable non-porous materials.

The process generally involves:

1. Preparing and cleaning the surface.
2. Applying penetrant.
3. Allowing appropriate dwell time.
4. Removing excess penetrant.
5. Applying developer where required.
6. Examining the resulting indications.
7. Evaluating the indication against applicable criteria.
8. Recording the findings.

It can be useful for detecting fine surface-breaking cracks around:

• Welds.
• Machined components.
• Shafts.
• Fillets.
• Complex surfaces.

Magnetic Particle Testing

Magnetic particle testing can identify surface and near-surface discontinuities in suitable ferromagnetic materials.

It is particularly useful for components such as:

• Carbon steel shafts.
• Ferromagnetic fabricated components.
• Welded mechanical structures.
• Forged components.

The method uses a magnetic field and suitable particles to reveal discontinuities that interrupt the magnetic field.

Ultrasonic Testing

Ultrasonic testing uses high-frequency sound waves to examine material condition and identify internal or surface-connected discontinuities.

It can be useful for:

• Weld inspection.
• Thickness measurement.
• Crack detection.
• Internal discontinuity assessment.
• Remaining wall-thickness evaluation.

Ultrasonic techniques can provide valuable information where a visual or surface-only method is insufficient.

Radiographic Testing

Radiographic examination can provide information about internal discontinuities by producing an image associated with variations in material attenuation.

Depending on the application, it can identify certain internal discontinuities such as:

• Porosity.
• Inclusions.
• Certain volumetric weld defects.

The technique requires controlled application and competent interpretation.

Recognising Creep Degradation

Creep is time-dependent deformation that can occur when a material is exposed to sustained stress, particularly at elevated temperature.

Creep is especially relevant to:

• High-temperature piping.
• Pressure-containing equipment.
• Boilers.
• High-temperature process equipment.
• Turbine components.
• Furnace components.
• Hot mechanical structures.

Creep can gradually cause:

• Dimensional changes.
• Permanent deformation.
• Localised damage.
• Cracking.
• Loss of load-carrying capacity.

Unlike a sudden overload, creep damage may develop over an extended period.

Early Indicators of Creep

Potential signs include:

• Permanent dimensional change.
• Bulging.
• Localised deformation.
• Distortion.
• Wall-thickness changes.
• Surface cracking.
• Changes around welds.
• Increasing deformation over time.

The significance of these indicators depends on the component, material, temperature, stress and service history.

Targeted Inspection for Creep

A targeted creep inspection programme may focus on:

• High-temperature zones.
• Welded joints.
• Areas with high stress.
• Locations with previous deformation.
• Components with long service exposure.
• Regions where temperature is highest.

Useful inspection activities may include:

• Dimensional measurement.
• Visual inspection.
• Ultrasonic examination.
• Surface examination.
• Metallurgical assessment where appropriate.
• Thickness measurement.
• Replication or other specialist techniques where specified by the applicable integrity programme.

Understanding Erosion

Erosion occurs when material is progressively removed from a surface through mechanical interaction with a moving fluid, particles or another material.

Erosion can be significant in:

• Pipe bends.
• Elbows.
• Valve internals.
• Pump components.
• Impellers.
• Nozzles.
• Flow restrictions.
• Areas with high fluid velocity.

Erosion may become more severe where:

• Flow velocity increases.
• Solid particles are present.
• Fluid direction changes abruptly.
• Turbulence is high.
• Materials are relatively susceptible to material loss.

Early Signs of Erosion

Potential indicators include:

• Localised wall-thickness reduction.
• Pitting-like material loss.
• Surface roughening.
• Reduced component dimensions.
• Localised thinning at bends.
• Increased leakage risk.
• Changes in flow performance.

Thickness monitoring can be particularly useful for tracking erosion-related material loss.

Ultrasonic Thickness Measurement for Erosion

Ultrasonic thickness measurement can provide quantitative evidence of wall-thickness condition.

The basic concept involves:

• Selecting measurement locations.
• Establishing a baseline thickness.
• Taking repeat measurements.
• Comparing measurements over time.
• Calculating or estimating the degradation trend where appropriate.
• Evaluating remaining thickness against applicable requirements.

For example, if a pipe section shows a progressive decrease in measured wall thickness, the inspection team should determine whether the trend is consistent with erosion, corrosion or another degradation mechanism.

Baseline Inspection

A baseline inspection establishes an initial condition against which future results can be compared.

A good baseline should document:

• Component identity.
• Inspection location.
• Measurement method.
• Material where relevant.
• Initial measurement.
• Date.
• Equipment condition.
• Relevant operating conditions.

Without a reliable baseline, identifying gradual degradation becomes more difficult.

Trend-Based Degradation Assessment

Repeated inspection results can be more valuable than a single measurement because they allow engineers to identify trends.

For example:

The sequence indicates progressive material loss.

The engineer should not automatically assume the loss is caused by erosion. Other mechanisms, measurement uncertainty and inspection conditions should also be considered.

The key principle is:

Single measurement = condition snapshot

Repeated measurements = degradation trend

Selecting the Correct NDT Method

The inspection method should match the expected defect mechanism.

For surface-breaking cracks

Potential methods include:

• Visual examination.
• Liquid penetrant testing.
• Magnetic particle testing for suitable ferromagnetic materials.

For internal defects

Potential methods include:

• Ultrasonic testing.
• Radiographic testing.

For wall-thickness loss

Potential methods include:

• Ultrasonic thickness measurement.
• Appropriate dimensional inspection.

For deformation

Potential methods include:

• Dimensional measurement.
• Visual inspection.
• Geometric surveys where appropriate.

Selecting an unsuitable method may produce false confidence because the inspection may not be capable of detecting the relevant defect.

Risk-Based Targeting of Inspection Locations

A risk-based approach directs greater attention to components where failure consequences or degradation likelihood are higher.

Priority locations may include:

• High-stress areas.
• High-temperature areas.
• High-pressure boundaries.
• Areas exposed to aggressive fluids.
• Components with repeated cyclic loading.
• Previously failed components.
• Locations showing abnormal trends.
• Areas with known manufacturing weaknesses.

This makes inspection resources more effective.

Linking FMEA with NDT

FMEA and targeted NDT work particularly well together.

For example:

FMEA identifies: fatigue cracking as a significant failure mode.

FMEA identifies cause: cyclic loading at a shaft shoulder.

NDT strategy: targeted surface inspection of the shaft shoulder.

Monitoring strategy: vibration trending.

Preventive action: investigate alignment and loading.

This demonstrates how predictive analysis can directly inform inspection planning.

Recognising Degradation Through Condition Trends

Inspection results should ideally be considered alongside operating data.

Relevant trends may include:

• Increasing vibration.
• Increasing temperature.
• Increasing leakage.
• Increasing pressure drop.
• Reducing flow.
• Increasing wall loss.
• Increasing dimensional deviation.
• Increasing maintenance frequency.

When several indicators change simultaneously, the likelihood of developing degradation may increase.

For example:

Increasing vibration + bearing temperature + lubrication deterioration = stronger indication of possible bearing degradation

The combination of evidence can be more informative than any single measurement.

Practical Example: Fatigue Crack in a Rotating Shaft

A rotating shaft has operated continuously for several years. Vibration data show a gradual increase, although the shaft remains within normal operating speed.

FMEA identifies the shaft shoulder as a potential fatigue initiation location because of the geometry and repeated loading.

A targeted surface inspection is therefore conducted.

The inspection identifies a small surface indication near the shoulder.

The QA/QC response should involve:

• Recording the indication.
• Confirming its characteristics using the appropriate inspection method.
• Comparing it with applicable acceptance criteria.
• Reviewing operating history.
• Assessing fatigue significance.
• Escalating for engineering evaluation where required.

The important point is that the defect was identified before a major shaft failure occurred.

Practical Example: Creep in High-Temperature Piping

A high-temperature process line has operated for an extended period. Inspection records show slight dimensional changes in a high-temperature section.

The engineering team identifies the area as creep-sensitive because it experiences:

• Elevated temperature.
• Sustained stress.
• Long operating duration.

A targeted inspection programme is established to monitor:

• Dimensional changes.
• Wall condition.
• Surface cracking.
• Weld regions.

Repeated measurements provide evidence of whether deformation is stable or increasing.

Practical Example: Erosion at a Pipe Elbow

A process line transports fluid containing abrasive particles. The line contains several changes in flow direction.

Inspection data show that an elbow is experiencing faster wall-thickness reduction than straight sections.

The QA/QC team identifies the elbow as an erosion-prone location.

Actions may include:

• Increased thickness-monitoring frequency.
• Review of flow velocity.
• Examination of particle concentration.
• Evaluation of material suitability.
• Review of elbow geometry.
• Engineering assessment of remaining thickness.

This demonstrates how targeted NDT can focus resources on the locations most likely to degrade.

Inspection Planning Process

Step 1: Understand the Equipment

Identify:

• Component.
• Material.
• Function.
• Operating conditions.
• Service history.

Step 2: Identify Degradation Mechanisms

Consider:

• Fatigue.
• Creep.
• Erosion.
• Corrosion.
• Wear.
• Thermal degradation.

Step 3: Identify High-Risk Locations

Review:

• Stress concentrations.
• Welds.
• High-temperature areas.
• Flow restrictions.
• Previous defect locations.

Step 4: Select Suitable NDT Methods

Match the method to the suspected defect.

Step 5: Establish Baseline Data

Record:

• Initial condition.
• Measurements.
• Inspection method.
• Location.

Step 6: Perform Targeted Inspection

Conduct the inspection using appropriate procedures and competent personnel.

Step 7: Interpret Indications

Determine:

• Location.
• Size where applicable.
• Type.
• Significance.
• Trend.

Step 8: Compare with Acceptance Criteria

Determine whether the indication is:

• Acceptable.
• Requires monitoring.
• Requires further examination.
• Requires repair.
• Requires replacement.
• Requires engineering assessment.

Step 9: Establish Follow-Up

Where degradation is identified:

• Increase monitoring.
• Investigate the cause.
• Evaluate remaining integrity.
• Implement preventive controls.

Step 10: Update the Risk Assessment

Inspection findings should feed back into:

• FMEA.
• Maintenance plans.
• Inspection plans.
• Risk assessments.
• Engineering specifications.

Importance of Competent NDT Personnel

NDT results are only as reliable as the inspection process and interpretation supporting them.

Inspection quality depends on:

• Appropriate method selection.
• Correct equipment.
• Suitable procedures.
• Adequate surface preparation.
• Correct calibration.
• Appropriate sensitivity.
• Competent personnel.
• Correct interpretation.
• Traceable reporting.

A technically unsuitable inspection method or poorly controlled inspection can produce false confidence.

Acceptance Criteria and Engineering Judgement

Finding an indication does not automatically mean that a component must be rejected.

The significance of an indication depends on:

• Defect type.
• Defect size.
• Defect location.
• Component function.
• Loading.
• Material.
• Operating environment.
• Applicable acceptance criteria.
• Consequence of failure.

The inspection result should therefore be assessed against the applicable technical and engineering requirements.

Common Inspection Errors

Common weaknesses include:

• Selecting an inspection method without considering the failure mechanism.
• Inspecting only easily accessible locations.
• Ignoring previous defect locations.
• Failing to establish baseline measurements.
• Comparing measurements taken under inconsistent conditions.
• Ignoring measurement uncertainty.
• Treating every indication as a critical defect.
• Treating every indication as harmless.
• Failing to trend results.
• Ignoring changes in operating conditions.
• Poor inspection documentation.
• Failing to update risk assessments after findings.

Benefits of Targeted NDT

A properly designed targeted inspection programme can:

• Detect degradation at an early stage.
• Reduce unexpected component failures.
• Improve mechanical reliability.
• Support predictive maintenance.
• Reduce unnecessary component replacement.
• Improve inspection efficiency.
• Focus resources on critical areas.
• Support evidence-based decisions.
• Identify developing fatigue damage.
• Monitor creep-sensitive equipment.
• Detect material loss from erosion.
• Improve asset integrity management.
• Support continuous improvement.
• Reduce unplanned downtime.

Case Study: Integrated Degradation Detection

Background

A manufacturing plant operates a high-speed rotating assembly connected to a process pipeline. The system has operated for several years under variable loading conditions.

Recent operating data show:

• Gradually increasing vibration.
• Slightly increased bearing temperature.
• Occasional changes in operating load.

Risk Identification

The FMEA identifies several potential failure mechanisms:

• Fatigue cracking.
• Bearing degradation.
• Misalignment.
• Coupling deterioration.

The shaft shoulder and coupling interface are identified as high-risk locations.

Targeted Inspection

The inspection team conducts appropriate surface and dimensional examinations at the identified locations. Additional condition-monitoring data are reviewed.

A small indication is identified near a high-stress shaft transition.

Engineering Assessment

The finding is evaluated according to:

• Location.
• Size.
• Loading.
• Material.
• Operating history.
• Applicable acceptance criteria.

The team determines that further engineering evaluation is required before continued unrestricted operation.

Preventive Measures

Potential controls include:

• Detailed defect characterisation.
• Temporary operating controls where justified.
• Increased vibration monitoring.
• Alignment verification.
• Repair or replacement based on engineering assessment.
• Review of the FMEA.

Lesson from the Case

The case demonstrates the value of combining predictive risk analysis with targeted NDT. The physical indication alone does not tell the complete story. The significance becomes clearer when inspection results are combined with loading history, condition trends and failure-mode analysis.

Integrating NDT Findings into Continuous Improvement

NDT findings should not remain isolated within inspection reports. They should be fed into the wider quality and reliability system.

Useful actions include:

• Updating defect databases.
• Revising inspection frequencies.
• Updating FMEA rankings.
• Revising maintenance strategies.
• Reviewing manufacturing processes.
• Improving component design.
• Investigating root causes.
• Monitoring recurring degradation.

This creates a feedback cycle:

Inspection → Finding → Analysis → Corrective action → Monitoring → Verification

Final Professional Perspective

Recognising early physical degradation requires engineers to understand both the failure mechanism and the capabilities of the inspection method used to detect it. Fatigue cracking, creep and erosion have different causes, different physical manifestations and different inspection requirements. Consequently, there is no single NDT technique that can reliably identify every type of degradation in every component.

The most effective approach is targeted and risk-based. FMEA, operating data, historical inspection results and condition-monitoring trends can be used to identify where degradation is most likely to occur. Appropriate NDT methods can then be directed towards those locations to establish reliable evidence of component condition. Repeated inspection provides additional value because it allows engineers to distinguish an isolated condition from a developing degradation trend.

Conclusion

Targeted non-destructive inspection provides an essential mechanism for identifying early signs of mechanical degradation before they develop into major failures. Fatigue cracking, creep and erosion can progressively reduce component integrity, but their early indicators can often be detected through appropriate combinations of visual examination, surface inspection, ultrasonic techniques, thickness measurements, radiographic examination and other specialised methods. The correct inspection method must always be matched to the expected degradation mechanism, component material and operating environment.

For mechanical QA/QC and reliability professionals, the greatest value comes from combining NDT with predictive risk assessment, FMEA, historical inspection data and operational condition monitoring. This enables inspection resources to be concentrated on high-risk locations, degradation trends to be identified earlier, and engineering decisions to be supported by objective evidence. When inspection findings are systematically fed back into maintenance, design, process control and continuous-improvement activities, organisations can reduce unexpected failures, extend reliable component service, improve asset integrity and maintain consistently high mechanical quality and safety standards.

3. Design a Proactive Preventive Action Plan to Modify Equipment Setups or Operating Guidelines Before a Predicted Mechanical Failure Occurs

Proactive preventive action is a fundamental principle of modern mechanical QA/QC and reliability management. Instead of waiting for a mechanical component to fail and then responding to the resulting defect, production interruption, equipment damage or safety concern, a proactive approach uses available engineering evidence to identify developing risks and implement controls before failure occurs. In complex industrial environments, this requires the integration of inspection findings, condition-monitoring data, FMEA results, maintenance history, operating trends, engineering calculations and quality records.

A proactive preventive action plan translates predicted failure risks into practical engineering controls. These controls may involve modifying equipment setup, adjusting operating parameters, improving alignment, changing lubrication practices, revising inspection intervals, introducing additional monitoring, modifying maintenance instructions, or updating operating guidelines. The objective is not simply to make equipment operate differently; it is to reduce the likelihood of a specific predicted failure while maintaining required production, quality, safety and performance requirements.

For mechanical QA/QC professionals, the quality of a preventive action plan depends on the strength of the evidence supporting the predicted failure, the suitability of the proposed intervention, clear ownership of actions, controlled implementation and verification of effectiveness. A technically sound action that is poorly implemented may introduce new risks. Therefore, preventive action must be treated as a controlled engineering change rather than an informal operational adjustment.

Understanding Proactive Preventive Action

A proactive preventive action is a planned intervention introduced because available evidence indicates that a potential problem could occur if existing conditions continue.

The basic logic is:

Evidence → Predicted failure → Risk assessment → Preventive action → Controlled implementation → Verification

Examples include:

• Increasing vibration indicates developing shaft or bearing problems.
• Repeated overheating indicates inadequate cooling or lubrication.
• Progressive wall-thickness loss indicates potential future loss of containment.
• Increasing tool wear predicts dimensional defects.
• Repeated coupling misalignment predicts premature bearing or shaft damage.
• Repeated pressure surges indicate a potential pressure-related failure.
• Increasing fatigue indications suggest the need for intervention before crack growth becomes critical.

The intervention should occur before the predicted failure reaches an unacceptable condition.

Purpose of a Preventive Action Plan

A preventive action plan provides a structured method for controlling identified mechanical risks.

Its purposes include:

• Preventing predicted component failure.
• Reducing defect recurrence.
• Protecting equipment integrity.
• Maintaining production reliability.
• Reducing unplanned downtime.
• Controlling operating risks.
• Improving maintenance effectiveness.
• Protecting product quality.
• Establishing clear responsibilities.
• Providing traceable evidence of action.
• Verifying whether the intervention worked.

A strong plan connects each action directly to an identified failure mechanism rather than introducing generic maintenance activities without technical justification.

Key Concepts and Definitions

Difference Between Reactive and Proactive Approaches

Reactive maintenance generally follows this sequence:

Failure → Investigation → Repair → Restart

Although this approach can be necessary when unexpected failures occur, it provides limited opportunity to prevent the initial event.

A proactive approach follows:

Warning indicator → Analysis → Risk assessment → Preventive action → Verification

For example, if a bearing temperature increases gradually, a reactive approach may wait until the bearing fails. A proactive approach investigates the temperature trend, determines the likely cause, and introduces an intervention before failure.

Possible interventions may include:

• Lubrication correction.
• Alignment verification.
• Load reduction.
• Cooling improvement.
• Bearing inspection.
• Increased monitoring.

Identifying the Predicted Failure

The first stage in designing the action plan is to define exactly what failure is being predicted.

The description should identify:

• Component.
• Failure mode.
• Location.
• Evidence.
• Operating condition.
• Expected consequence.

For example:

Weak statement:

“Bearing may fail.”

Stronger engineering statement:

“Bearing temperature has increased progressively over three inspection periods while vibration has also increased, indicating a developing lubrication, alignment or bearing-condition problem that could lead to premature bearing failure.”

The stronger statement provides a basis for action.

Sources of Evidence for Predictive Action

A preventive action plan should be based on reliable technical evidence.

Potential sources include:

• FMEA results.
• Inspection reports.
• NDT findings.
• Vibration measurements.
• Temperature records.
• Pressure records.
• Historical breakdown data.
• Dimensional inspection results.
• Lubrication records.
• Maintenance history.
• Material test results.
• Production quality data.
• Operator observations.
• Equipment performance data.

The greater the consequence of the predicted failure, the stronger the evidence and engineering justification should generally be.

Establishing the Failure Mechanism

Before selecting an intervention, the team should understand why failure is predicted.

Possible mechanisms include:

Fatigue

Repeated cyclic loading can initiate and propagate cracks.

Potential controls include:

• Reducing cyclic loads.
• Improving alignment.
• Controlling vibration.
• Reviewing component geometry.
• Increasing inspection frequency.

Wear

Repeated mechanical contact can progressively remove material.

Potential controls include:

• Improving lubrication.
• Controlling contact conditions.
• Replacing worn components.
• Improving material selection.
• Monitoring dimensional loss.

Creep

Long-term exposure to elevated temperature and sustained stress can cause deformation.

Potential controls include:

• Reducing operating temperature where feasible.
• Controlling loads.
• Monitoring deformation.
• Increasing targeted inspection.
• Reviewing component suitability.

Erosion

High-velocity fluids or particles can progressively remove material.

Potential controls include:

• Reducing flow velocity where practicable.
• Modifying flow paths.
• Increasing wall-thickness monitoring.
• Reviewing material suitability.
• Replacing vulnerable components.

Misalignment

Misalignment can increase loads on shafts, bearings and couplings.

Potential controls include:

• Precision alignment.
• Improved mounting.
• Foundation correction.
• Alignment verification after maintenance.

Defining Preventive Action Objectives

Each action should have a clear objective.

Examples include:

• Reduce vibration to an acceptable operating range.
• Stabilise bearing temperature.
• Reduce wall-thickness loss.
• Prevent repeated coupling misalignment.
• Reduce excessive tool wear.
• Maintain dimensional capability.
• Prevent pressure excursions.
• Reduce cyclic loading.
• Improve lubrication condition.

Objectives should be measurable wherever possible.

Designing Equipment Setup Modifications

Equipment setup changes can be highly effective when the predicted failure is associated with configuration or installation conditions.

Potential setup modifications include:

• Realigning rotating equipment.
• Adjusting component clearances.
• Correcting coupling alignment.
• Improving equipment support.
• Adjusting belt or chain tension.
• Correcting bearing installation.
• Improving lubrication delivery.
• Modifying cooling arrangements.
• Improving component positioning.
• Reconfiguring flow paths.

The modification should be based on evidence rather than trial-and-error adjustment.

Example: Correcting Rotating Equipment Alignment

Suppose vibration data show a gradual increase in a rotating machine.

Inspection indicates:

• Shaft vibration increasing.
• Bearing temperature increasing.
• Coupling alignment outside the preferred condition.

The preventive action plan could include:

1. Stop the equipment under controlled conditions.
2. Verify alignment measurements.
3. Correct alignment.
4. Inspect coupling condition.
5. Verify bearing condition.
6. Restart under controlled conditions.
7. Record vibration and temperature.
8. Compare results with baseline data.

The intervention is successful only if the expected improvement is demonstrated.

Modifying Operating Guidelines

Sometimes equipment does not need immediate physical modification. The risk may be reduced by controlling how the equipment is operated.

Operating guidelines may define:

• Maximum operating speed.
• Maximum pressure.
• Maximum temperature.
• Permitted load.
• Start-up procedure.
• Shutdown procedure.
• Ramp-up rate.
• Ramp-down rate.
• Maximum vibration.
• Inspection trigger levels.

For example, if repeated pressure surges occur during start-up, a revised start-up procedure may introduce a controlled pressure increase rather than allowing rapid pressurisation.

Establishing Operating Limits

Operating limits should be based on:

• Equipment design capability.
• Manufacturer information.
• Engineering calculations.
• Plant operating conditions.
• Historical performance.
• Applicable technical requirements.
• Identified degradation mechanisms.

Limits should not simply be selected because they are convenient for production.

A limit should have a technical basis.

Trigger Points and Action Levels

A proactive action plan becomes much stronger when it defines clear trigger points.

For example:

These values are illustrative. Actual limits should be established from the applicable equipment requirements, engineering basis, manufacturer information and approved plant procedures.

Prioritising Preventive Actions

Not every predicted failure requires the same level of intervention.

Priority should consider:

• Severity of failure.
• Likelihood.
• Detectability.
• Failure consequences.
• Equipment criticality.
• Production impact.
• Safety implications.
• Environmental consequences.
• Availability of alternative equipment.

High-consequence risks should receive appropriate priority even where the probability appears relatively low.

Developing an Action Ownership Structure

Each preventive action should have a clearly identified owner.

Responsibilities may include:

QA/QC Engineer

• Verify quality evidence.
• Review inspection findings.
• Confirm technical requirements.
• Monitor implementation evidence.

Mechanical Engineer

• Assess mechanical integrity.
• Confirm engineering suitability.
• Develop technical modifications.

Maintenance Engineer

• Plan equipment intervention.
• Coordinate maintenance work.
• Verify equipment condition.

Operations Team

• Implement revised operating instructions.
• Monitor operating parameters.
• Report abnormal conditions.

Reliability Engineer

• Analyse trends.
• Review failure probability.
• Evaluate effectiveness.

Supervisor

• Coordinate work execution.
• Confirm procedural compliance.
• Escalate deviations.

Creating a Controlled Action Plan

A practical preventive action register can contain:

• Action number.
• Equipment identification.
• Failure mode.
• Evidence.
• Risk level.
• Preventive action.
• Responsible person.
• Target date.
• Required shutdown.
• Verification method.
• Completion status.
• Effectiveness result.

This creates traceability between prediction and intervention.

Risk Assessment Before Implementing Changes

A preventive modification should itself be assessed for risk.

A change that solves one problem may create another.

For example, reducing pump speed may reduce vibration but also reduce required process flow.

Changing lubrication may reduce bearing temperature but introduce compatibility concerns if the lubricant is unsuitable.

Therefore, the team should consider:

• New mechanical loads.
• Process impact.
• Quality impact.
• Safety implications.
• Equipment compatibility.
• Maintenance implications.
• Production requirements.

Management of Engineering Change

Significant equipment modifications should be controlled through the organisation’s engineering or management-of-change process where applicable.

This may involve:

• Technical review.
• Drawing updates.
• Procedure revision.
• Risk assessment.
• Approval.
• Inspection.
• Testing.
• Commissioning.
• Documentation.

The purpose is to ensure that changes are properly evaluated rather than introduced informally.

Preventive Action for Fatigue Risk

Consider a rotating shaft showing increasing vibration and evidence of cyclic loading.

A proactive action plan might include:

• Verify shaft alignment.
• Inspect high-stress transitions.
• Review operating speed.
• Review torque fluctuations.
• Conduct targeted NDT.
• Establish vibration monitoring.
• Define escalation criteria.
• Review maintenance frequency.

The plan should address the underlying contributors rather than simply replacing the shaft repeatedly.

Preventive Action for Erosion Risk

For a pipe elbow showing accelerated material loss:

Potential actions include:

• Increase thickness monitoring.
• Review flow velocity.
• Review particle concentration.
• Examine flow conditions.
• Assess material suitability.
• Consider geometry modification.
• Establish minimum acceptable thickness.
• Define inspection trigger points.

The objective is to prevent loss of containment rather than simply respond after a leak occurs.

Preventive Action for Creep Risk

For high-temperature equipment showing progressive deformation:

Potential measures may include:

• Review operating temperature.
• Review sustained loads.
• Monitor dimensional changes.
• Inspect high-risk weld regions.
• Establish periodic condition assessments.
• Review remaining integrity.
• Control operating conditions within approved limits.

Creep-related interventions require careful engineering evaluation because changes in operating temperature or load may affect production requirements.

Preventive Action for Bearing Degradation

If vibration and temperature trends indicate developing bearing degradation:

Potential actions include:

• Verify lubrication condition.
• Check lubricant quantity.
• Inspect bearing installation.
• Verify shaft alignment.
• Review loading.
• Increase vibration monitoring.
• Establish temperature trigger levels.
• Plan controlled replacement if required.

The action should be proportional to the evidence and risk.

Verification of Preventive Action

Implementation alone does not prove success.

The team should establish how effectiveness will be demonstrated.

Verification may involve:

• Post-maintenance vibration measurements.
• Temperature comparison.
• Dimensional inspection.
• Pressure monitoring.
• Flow measurement.
• NDT follow-up.
• Defect-rate comparison.
• Production performance comparison.

For example:

Before intervention: vibration = increasing trend.

After intervention: vibration returns to stable baseline.

This provides stronger evidence that the intervention has addressed the predicted mechanism.

Comparing Before-and-After Data

A useful preventive action review should compare:

• Baseline condition.
• Intervention date.
• Post-intervention condition.
• Operating conditions.
• Inspection results.
• Failure indicators.

The comparison should account for changes in operating load. A component may appear improved simply because it is operating under a lower load.

Effectiveness Review

A preventive action can be classified as effective when evidence demonstrates that:

• The predicted failure risk has been reduced.
• The warning indicator has stabilised.
• The defect trend has improved.
• The underlying cause has been controlled.
• No unacceptable new risk has been introduced.

If the expected improvement does not occur, the action should be reconsidered.

Practical Case Study: Preventing a Bearing Failure

Background

A production machine uses a high-speed rotating shaft supported by bearings.

Routine monitoring identifies:

• Increasing vibration.
• Gradual temperature increase.
• Increasing lubrication contamination.

No catastrophic failure has occurred.

Predictive Assessment

The FMEA identifies bearing degradation as a significant potential failure mode.

Potential causes include:

• Lubricant contamination.
• Misalignment.
• Excessive loading.

Preventive Action Plan

The engineering team develops the following actions:

• Inspect lubrication condition.
• Replace contaminated lubricant where appropriate.
• Verify alignment.
• Inspect bearing condition.
• Increase vibration monitoring.
• Establish temperature warning levels.
• Review operating load.

Implementation

The equipment is taken through a controlled maintenance intervention. Alignment is corrected and lubrication condition is restored.

Verification

Following restart:

• Vibration decreases.
• Temperature stabilises.
• No abnormal trend is observed.

Outcome

The intervention prevents the predicted bearing failure from progressing to an unplanned breakdown.

The case demonstrates that proactive action is most effective when prediction, intervention and verification are connected.

Practical Case Study: Preventing Dimensional Defects

A machining process produces mechanical components with gradually increasing dimensional variation.

Historical data show:

• Stable dimensions initially.
• Increasing variation as tool usage increases.
• Increased rejection rates near the end of tool life.

Predicted Failure

The likely failure mechanism is excessive tool wear causing dimensional drift.

Preventive Action

The team introduces:

• Tool-life monitoring.
• Earlier tool replacement.
• Controlled machining parameters.
• Increased dimensional checks near expected tool-life limits.

Verification

The team compares:

• Pre-change rejection rate.
• Post-change rejection rate.
• Dimensional variation.
• Tool replacement frequency.

If dimensional variation decreases and rejection rates improve without unacceptable production losses, the action demonstrates effectiveness.

Preventive Action Plan Workflow

A structured workflow can be represented as:

1. Detect warning signal

↓

2. Validate evidence

↓

3. Identify predicted failure

↓

4. Determine failure mechanism

↓

5. Assess risk

↓

6. Select preventive intervention

↓

7. Review change implications

↓

8. Approve and implement

↓

9. Verify performance

↓

10. Monitor for recurrence

This workflow supports controlled and evidence-based intervention.

Common Mistakes in Preventive Action Planning

Acting on unverified data

A single abnormal reading may result from measurement error. Before changing equipment or operating procedures, the evidence should be validated.

Treating symptoms instead of causes

Reducing vibration without investigating misalignment may temporarily hide the problem.

Making uncontrolled operating changes

Operators should not introduce informal limits without engineering review where formal control is required.

Failing to define effectiveness criteria

An action without a measurable success condition is difficult to evaluate.

Ignoring production requirements

A technically effective intervention may create unacceptable production consequences if the wider system is not considered.

Failing to update documentation

Changes to equipment setup or operating limits should be reflected in relevant controlled documentation.

Benefits of Proactive Preventive Action

A well-designed preventive action plan can provide:

• Earlier failure prevention.
• Improved mechanical reliability.
• Reduced unplanned downtime.
• Lower repair costs.
• Reduced defect recurrence.
• Improved production stability.
• Better equipment availability.
• Improved maintenance planning.
• Stronger QA/QC control.
• Better risk visibility.
• More effective use of inspection resources.
• Improved operational consistency.
• Better evidence-based decision-making.
• Increased confidence in equipment performance.

Integrating Preventive Actions with FMEA

FMEA should not end when the risk ranking is completed.

The relationship can continue as:

FMEA → Failure prediction → Preventive action → Verification → FMEA update

For example, if FMEA identifies bearing overheating as a significant risk, the preventive action may involve improved lubrication and condition monitoring. After implementation, the results should be reviewed and the FMEA updated where appropriate.

This creates a continuous improvement loop.

Integrating Preventive Actions with NDT

NDT findings can also initiate preventive action.

For example:

NDT detects early crack → Engineering evaluates significance → Operating condition reviewed → Preventive action implemented → Follow-up inspection conducted

This approach ensures that NDT findings lead to meaningful risk reduction rather than simply being recorded in inspection reports.

Integrating Operating Guidelines with Quality Control

Operating guidelines should clearly communicate important controls to personnel responsible for equipment operation.

Depending on the equipment, guidelines may address:

• Start-up sequence.
• Shutdown sequence.
• Maximum pressure.
• Maximum temperature.
• Maximum speed.
• Load restrictions.
• Vibration limits.
• Inspection requirements.
• Escalation requirements.

The objective is to ensure that operating behaviour does not unintentionally accelerate the predicted failure mechanism.

Professional Decision-Making Principles

When designing preventive action, engineers should consider five fundamental questions:

Is the evidence reliable?

Confirm that the data are accurate, representative and sufficiently consistent.

Is the predicted failure credible?

Connect the observed trend to a technically plausible failure mechanism.

Is the proposed action appropriate?

The intervention should address the underlying cause or meaningfully reduce risk.

Could the intervention introduce another risk?

Consider system-level effects.

How will effectiveness be demonstrated?

Define measurable evidence before implementing the action.

Quality Records and Traceability

Preventive action should be supported by appropriate records.

Records may include:

• Inspection reports.
• Trend graphs.
• NDT reports.
• Maintenance records.
• Engineering calculations.
• Risk assessments.
• Revised procedures.
• Equipment setup records.
• Photographic evidence where appropriate.
• Verification results.
• Effectiveness review.

Traceability enables future engineering teams to understand why the action was taken and whether it achieved the intended result.

Conclusion

A proactive preventive action plan transforms mechanical failure prediction into practical engineering intervention. By combining FMEA, NDT findings, condition-monitoring data, historical inspection records and operating trends, QA/QC and reliability teams can identify credible failure mechanisms before they result in equipment breakdown or significant mechanical defects. Preventive actions may involve equipment realignment, lubrication improvement, operating-limit adjustment, inspection-frequency changes, condition monitoring, maintenance modification, process control or engineering redesign, depending on the identified risk.

The effectiveness of proactive prevention depends on disciplined implementation. Each action should have a clearly defined technical basis, responsible owner, implementation method, verification requirement and effectiveness criterion. Changes to equipment setups or operating guidelines should be properly reviewed and controlled, particularly where they may affect safety, mechanical integrity, production performance or other parts of the process. After implementation, actual performance should be compared with the baseline condition to determine whether the predicted risk has been reduced.

When this approach is embedded within a broader mechanical QA/QC and reliability system, organisations can move from reactive failure response towards predictive and preventive control. The result is better component reliability, fewer recurring defects, reduced unplanned downtime, stronger process stability and more effective use of engineering and inspection resources. Most importantly, proactive preventive action allows mechanical problems to be addressed while there is still an opportunity to control them, rather than waiting until a developing defect becomes an operational failure.