In many industrial facilities, equipment begins to show problems long before reaching its expected service life. Bearings that wear repeatedly, seals that lose containment capability, motors that overheat, or pumps that gradually develop higher vibration levels are among the most common industrial mechanical failures.

In these situations, the immediate response is often to repair or replace the affected component. However, replacing a damaged part does not guarantee that the problem has actually been solved.

When a failure recurs, the defective component is often only the visible evidence of a deeper issue related to operating conditions, material selection, design, installation, or maintenance practices.

Root cause analysis, also known as RCA, makes it possible to investigate why a failure actually occurred and define actions that reduce the likelihood of recurrence. ASQ defines RCA as a group of approaches, tools, and techniques used to uncover the underlying causes of problems rather than simply correcting their effects.

A FAILURE DOES NOT ALWAYS BEGIN WITH THE COMPONENT THAT BREAKS

When equipment stops working, attention usually focuses on the part showing the most obvious damage.

If a bearing wears out, the bearing is replaced.
If a seal leaks, a new seal is installed.
If a shaft fractures, the shaft is replaced.

These actions may restore operation temporarily, but they do not necessarily explain what caused the failure.

A damaged bearing, for example, may be the result of:

• Insufficient or improper lubrication
• Lubricant contamination
• Component misalignment
• Loads beyond the original design conditions
• Excessive temperatures
• Vibration originating in another part of the equipment
• Incorrect material selection
• Improper tolerances or fits
• Installation errors

For this reason, a damaged part does not always represent the true cause of the problem. In many cases, it is the final element in a chain of conditions that began long before the equipment stopped operating.

NASA notes that the purpose of identifying a root cause is to correct a problem at its most basic source so it does not recur, rather than simply repairing the visible symptom.

WHY TEMPORARY REPAIRS LEAD TO REPEATED FAILURES

Emergency repairs are necessary when restoring production is the immediate priority. The problem begins when that temporary response becomes the only maintenance strategy.

Repeatedly replacing a component without investigating the origin of the damage can create a cycle of failure, repair, and failure again.

This reactive approach often leads to:

• Higher spare-part consumption
• Increased maintenance labor hours
• Repeated unplanned downtime
• Lost production capacity
• Progressive damage to other components
• Greater personnel exposure to emergency interventions
• Uncertainty about system reliability

Reactive maintenance addresses the asset after the failure has already occurred. In contrast, reliability-centered maintenance seeks to understand equipment functions, failure modes, and the actions required to preserve expected performance.

The difference is not simply about repairing equipment faster. It is about learning from failure in order to make better decisions.

WHAT IS ROOT CAUSE ANALYSIS?

Root cause analysis is a structured process used to identify the technical, human, and organizational factors that contributed to a failure.

Its purpose is not to find a quick or convenient explanation, but to build a cause-and-effect relationship supported by evidence.

An industrial failure analysis should distinguish among three levels:

Symptom

This is the visible or measurable indication of the problem.

Examples include:

• Elevated temperature
• Vibration
• Leakage
• Noise
• Pressure loss
• Irregular wear

Immediate Cause

This is the mechanism that directly produced the damage.

Examples include:

• Metal-to-metal contact
• Fatigue fracture
• Overheating
• Loss of the lubricating film
• Chemical attack
• Component deformation

Root Cause

This is the fundamental condition that allowed the failure mechanism to occur or recur.

Examples include:

• A material incompatible with the fluid or operating temperature
• An inadequate lubrication program
• A design that does not reflect actual operating conditions
• Poor contamination control
• Incorrect installation procedures
• Lack of condition monitoring
• Process changes made without reevaluating the equipment

Confusing these levels can lead to weak corrective actions. “The bearing wore out” describes the damage, but it does not explain why the bearing wore out.

HOW INDUSTRIAL FAILURE ANALYSIS IS PERFORMED

Although the depth of the investigation depends on equipment criticality, an RCA process usually includes several stages.

1. Clearly Define the Problem

The investigation should begin with a precise description of the failure.

It is not enough to record that “the pump failed.” It is necessary to determine:

• Which function was lost
• When the failure occurred
• Under what conditions the equipment was operating
• What changed before the event
• Which components were affected
• What the operational impact was

An ambiguous problem statement can direct the entire investigation toward incorrect conclusions.

2. Preserve and Collect Evidence

Evidence may include:

• Damaged components
• Photographs
• Maintenance history
• Temperature and vibration records
• Lubricant analysis
• Process data
• System alarms
• Operator interviews
• Installation conditions
• Recent production changes

Evidence should be collected before cleaning, repairing, or discarding parts, because those actions can eliminate critical information.

3. Reconstruct the Sequence of Events

The analysis team should establish what happened before, during, and after the failure.

A timeline helps identify relevant changes such as:

• Load increases
• Temperature variations
• Supplier changes
• Lubricant changes
• Recent repairs
• Changes in speed or operating cycles
• Variations in the process fluid

The U.S. Department of Energy notes that identifying causal factors helps reveal control deficiencies and supports timely corrective action.

4. Identify and Validate the Causes

A cause should not be accepted simply because it appears logical. It must be supported by data, inspections, or testing.

Useful questions during this stage include:

• Is there physical evidence supporting this cause?
• Does the cause explain all of the observed damage?
• Can the failure mechanism be reproduced?
• What condition allowed it to develop?
• Why did the existing controls fail to detect it?
• Would the failure occur again if only the damaged part were replaced?

5. Define Corrective Actions

An effective corrective action should reduce or eliminate the possibility of recurrence.

Depending on the findings, this may include:

• Changing a component material
• Correcting alignment or tolerances
• Redesigning part of the equipment
• Modifying operating parameters
• Improving contamination control
• Updating installation procedures
• Implementing condition monitoring
• Adjusting the maintenance strategy
• Training personnel
• Eliminating a critical dependence on external lubrication

6. Verify the Results

The analysis does not end when a correction is implemented.

It is necessary to confirm that the action:

• Eliminated or reduced the failure mechanism
• Did not introduce new risks
• Improved asset performance
• Can be sustained over time
• Can be applied to similar equipment

Without follow-up, even a strong recommendation can become an administrative action with little practical impact.

TOOLS USED IN ROOT CAUSE ANALYSIS

There is no single tool that is appropriate for every failure. Maintenance and reliability engineers select a methodology based on the complexity of the event, equipment criticality, and available evidence.

The Five Whys

This method involves repeatedly asking why a condition occurred until deeper contributing factors are identified.

It is useful for relatively simple problems, but it can lead to overly linear conclusions when a failure has multiple causes.

Ishikawa Diagram

Also known as a cause-and-effect or fishbone diagram, this tool helps organize potential causes into categories such as:

• Machine
• Method
• Material
• Manpower
• Measurement
• Environment

Its primary value is that it broadens the investigation and prevents the team from focusing too early on a single explanation.

Pareto Analysis

Pareto analysis helps identify which failures, assets, or factors account for the highest number of events or the greatest losses.

It is useful for prioritizing investigations, although it does not determine root cause by itself.

Fault Tree Analysis

Fault tree analysis graphically represents the combinations of events that can lead to a failure. It is particularly useful in complex systems involving multiple components or operating conditions.

FMEA

Failure Modes and Effects Analysis evaluates how each element can fail, what the consequences would be, and which controls are already in place.

It is primarily used as a preventive tool during design or maintenance planning. NASA distinguishes tools such as Pareto analysis and FMEA from a complete RCA methodology: they can support the investigation, but they do not replace evidence-based validation of causal relationships.

DMAIC

DMAIC—Define, Measure, Analyze, Improve, and Control—provides a structured framework for solving problems and sustaining improvements over time. ASQ describes it as a systematic approach for improving processes that are not meeting expected performance levels.

THE ROLE OF MATERIALS IN INDUSTRIAL MECHANICAL FAILURES

Root cause analysis does not always conclude that the problem lies in maintenance. In some cases, the equipment receives proper service, but the component material is not suitable for actual operating conditions.

This may occur when there is:

• Dry running
• Intermittent lubrication
• Extreme temperature
• Contact with corrosive fluids
• Low-viscosity media
• Contamination that cannot be fully eliminated
• Sanitary restrictions
• Continuous operation
• Risk of adhesion, galling, or accelerated wear

Under these conditions, increasing lubrication frequency or repeatedly replacing the same part may address the symptom without correcting the underlying cause.

The RCA may show that the system requires a material with different properties.

WHEN THE ROOT CAUSE IS DEPENDENCE ON EXTERNAL LUBRICATION

In certain equipment, recurring failures are associated with dependence on oils or greases that degrade, become contaminated, leak, or stop functioning under extreme operating conditions.

Carbon graphite components may provide an alternative when the analysis shows that the system requires:

• Self-lubrication
• A low coefficient of friction
• Dimensional stability
• Chemical resistance
• Operation at high or low temperatures
• Reduced galling risk
• Oil-free or grease-free operation

However, replacing a material should not be treated as an isolated decision. Load, speed, temperature, operating media, counterface, tolerances, and component geometry must all be evaluated.

Metcar develops carbon graphite materials and components for industrial applications in which conventional lubrication systems or traditional materials present limitations.

Its involvement may begin during the application analysis stage: understanding the failure mechanism, reviewing actual operating conditions, and determining whether a specific carbon graphite grade can help eliminate a recurring cause.

FROM REPLACING COMPONENTS TO IMPROVING INDUSTRIAL RELIABILITY

Equipment does not always fail early because its components are defective. More often, failure occurs because there is a gap between the conditions for which the equipment was designed and the conditions under which it actually operates.

Root cause analysis helps make that gap visible.

By examining evidence, damage mechanisms, operating history, and causal relationships, engineers can move from a strategy based on repeated repairs to one focused on industrial reliability.

The question should no longer be only:

Which component should we replace?

It should also include:

What allowed it to fail, and what must change to prevent it from happening again?

When a failure recurs, the solution is not always to repair faster. It may involve modifying the process, improving controls, redesigning the system, or selecting a material better suited to actual operating conditions.

Is your equipment experiencing recurring wear, leakage, or failures despite regular maintenance?

Analyzing the failure mechanism and actual operating conditions may reveal whether the problem is related to material selection or dependence on external lubrication.

Contact the Metcar team to evaluate your application and determine whether a carbon graphite solution can help improve system reliability.