3. Failure Analysis and Root-Cause Problem Solving

Introduction and Context

Maintenance management is not limited to restoring equipment after a breakdown. Its more valuable purpose is to understand why failures occur, determine how they can be prevented, and improve the reliability of the production system. A repair that merely replaces a failed component may return an asset to service, but it does not necessarily remove the conditions that caused the failure. If those conditions remain, the organization experiences recurrence, lost production, safety exposure, emergency expenditure, and declining confidence in maintenance.

Failure analysis provides a disciplined method for learning from equipment deterioration and breakdown. It connects observable symptoms—such as vibration, leakage, overheating, fracture, or loss of output—with physical mechanisms, human actions, design weaknesses, operating conditions, and management-system deficiencies. Root-cause problem solving then converts that knowledge into corrective and preventive actions.

The subject is important in production management because failure causes are rarely confined to the maintenance department. A pump may fail because of poor lubrication, but the underlying cause may be an unsuitable procurement specification, inadequate training, contamination in the workplace, an unrealistic production schedule, or a maintenance procedure that does not define lubricant quantity. Effective analysis therefore requires technical reasoning as well as cross-functional decision-making.

Learning Objectives

After studying this chapter, the learner should be able to:

1. Distinguish failure modes, failure causes, failure mechanisms, and failure effects.

2. Explain common physical mechanisms of equipment failure.

3. Select and apply Pareto analysis, the five whys, fishbone diagrams, fault-tree analysis, FMEA, and FMECA.

4. Conduct a structured root-cause analysis using reliable evidence.

5. Develop corrective actions that address recurrence rather than symptoms.

6. Evaluate the managerial, ethical, technological, and organizational dimensions of failure investigation.

7. Interpret a practical failure case and recommend an appropriate improvement plan.

The Logic of Failure Analysis

A failure is the loss of an item’s ability to perform a required function. The definition is functional: an asset can fail even when it has not physically broken. A control valve that remains mechanically intact but cannot regulate flow has failed its required function. Similarly, a production line that operates at only 60 percent of its required rate may be functionally failed even though it continues to run.

Several terms must be separated carefully:

• Failure effect is the consequence observed at the equipment, system, production, safety, or business level.
• Failure mode is the specific manner in which an item fails, such as bearing seizure, seal leakage, electrical open circuit, or excessive shaft runout.
• Failure cause is the circumstance or condition that initiates the failure, such as contamination, overload, misalignment, or incorrect installation.
• Failure mechanism is the physical or chemical process by which damage develops, such as fatigue, wear, corrosion, erosion, or thermal degradation.
• Root cause is a fundamental cause that, if eliminated or controlled, prevents recurrence or significantly reduces its probability.
• Contributing cause is a condition that increases the likelihood or severity of failure but may not independently produce it.

For example, the effect may be “centrifugal pump unable to deliver required flow.” The failure mode may be “impeller damage.” The physical mechanism may be cavitation erosion. Immediate causes may include low suction pressure and air entrainment. Contributing causes may include an undersized suction line, blocked strainer, and inadequate operating instructions. A deeper organizational root cause may be that the original design review did not examine minimum available suction pressure.

A useful investigation moves from symptom to evidence, from evidence to mechanism, and from mechanism to controllable causes. It avoids stopping at statements such as “operator error” or “component failed,” because such statements describe an event without explaining why the system allowed it to occur.

Physical Failure Mechanisms

Understanding physical mechanisms is central to credible analysis. The same failure mode can arise from different mechanisms, and the corrective action depends on identifying the correct one.

Fatigue and Fracture

Fatigue occurs when repeated or fluctuating stresses initiate and propagate a crack, often at a stress concentration. Shafts, gears, springs, welded structures, and rotating components are vulnerable. Fatigue may occur even when each individual load is below the material’s static strength. Evidence may include beach marks, crack origin at a keyway or sharp corner, and progressive crack growth.

Overload fracture occurs when the applied stress exceeds the component’s immediate strength. It is usually associated with rapid deformation or a relatively sudden break. Confusing overload with fatigue can lead to incorrect action: increasing material strength will not solve a fatigue problem caused by misalignment or cyclic vibration.

Wear

Wear is the progressive loss of material from interacting surfaces. Abrasive wear occurs when hard particles or rough surfaces cut or plough a softer surface. Adhesive wear results when surfaces locally weld and tear apart, often under inadequate lubrication. Erosive wear is caused by high-velocity particles or fluid. Fretting wear arises from small-amplitude oscillatory motion at contacting surfaces.

The investigation should examine material compatibility, surface hardness, lubrication, contamination, load, speed, clearance, and alignment. A worn bearing is not automatically evidence of poor bearing quality.

Corrosion and Chemical Degradation

Corrosion involves chemical or electrochemical reaction between a material and its environment. Uniform corrosion reduces thickness gradually, while pitting, crevice corrosion, galvanic corrosion, and stress-corrosion cracking can produce localized and unexpectedly severe damage. Temperature, moisture, chemical concentration, coating condition, and material selection are important variables.

Thermal Failure

Excessive temperature can degrade insulation, lubricants, seals, polymers, and electronic components. Thermal failure may result from overload, inadequate cooling, blocked ventilation, high ambient temperature, excessive friction, or incorrect electrical connection. Temperature is often a symptom rather than the complete cause; the analyst must determine why heat generation exceeded heat removal.

Electrical and Electronic Failure

Electrical failures include insulation breakdown, short circuits, open circuits, loose connections, contact wear, overvoltage, and phase imbalance. Electronic systems may also fail because of power-quality disturbances, moisture, thermal cycling, electromagnetic interference, software defects, or component obsolescence. Evidence should be preserved before resetting or replacing the failed device, since diagnostic data may otherwise be lost.

Deformation, Leakage, and Deterioration

Plastic deformation, buckling, seal degradation, leakage, and loss of dimensional accuracy may result from overload, pressure excursions, chemical attack, incorrect assembly, vibration, or aging. These failures often involve interaction between design, operation, and maintenance rather than a single isolated cause.

A Structured Root-Cause Investigation

A sound investigation begins by defining the problem precisely. “The machine keeps breaking down” is too vague. A useful problem statement identifies the asset, function, failure mode, location, time, operating condition, and measurable consequence. For example: “Between 1 and 15 June, the No. 2 process pump lost discharge pressure three times after approximately 40 hours of operation, causing 9.5 hours of production interruption.”

The investigation should then establish a fact base. Relevant evidence may include operator logs, work orders, trend data, alarms, photographs, failed parts, lubrication records, inspection results, process measurements, purchase specifications, and interviews. Interviews should seek facts and sequence rather than blame. The team should distinguish observed evidence from assumptions and record contradictory evidence rather than suppressing it.

A practical investigation sequence is:

1. Contain the immediate risk. Protect people, the environment, equipment, and product. Preserve failed parts where possible.

2. Define the failure. State the required function, actual performance, time, location, and impact.

3. Reconstruct the event. Establish what changed before, during, and after failure.

4. Identify the failure mode and mechanism. Examine the physical evidence and operating data.

5. Test possible causes. Use analysis tools, inspections, measurements, and controlled comparisons.

6. Identify root and contributing causes. Include technical, human, procedural, design, and management factors.

7. Select actions. Prefer actions that eliminate or control the cause and are proportionate to risk.

8. Verify effectiveness. Define measurable acceptance criteria and review recurrence.

9. Capture organizational learning. Update procedures, designs, training, spares, standards, and maintenance strategies.

Pareto Analysis: Choosing Where to Investigate

Pareto analysis ranks failure categories by frequency, downtime, cost, safety consequence, or another relevant measure. It is useful when an organization has many failure records but limited analytical resources. A bar chart arranged from largest to smallest category, often accompanied by a cumulative-percentage curve, helps identify the “vital few” contributors.

The method must be used thoughtfully. Frequency alone can be misleading. A minor recurring fault may produce many work orders, while a rare turbine failure may create much greater business or safety risk. Organizations should therefore construct separate Pareto analyses for frequency, downtime, maintenance cost, and criticality, or use a weighted risk measure.

Pareto analysis identifies priority; it does not establish causation. A high-ranked category still requires technical investigation.

The Five Whys

The five whys is an iterative questioning method used to move from an immediate event toward deeper causes. The number five is not a rule. The questioning should stop when the team reaches a cause that is supported by evidence and can be effectively controlled.

Example:

• Why did the motor trip? Because its winding temperature exceeded the protection limit.
• Why did the winding temperature rise? Because the motor was overloaded.
• Why was it overloaded? Because the driven pump required excessive torque.
• Why did the pump require excessive torque? Because its impeller was rubbing against the casing.
• Why was there rubbing? Because the shaft was misaligned after installation and no alignment verification was required in the work procedure.

The final statement may still require deeper examination, such as why the procedure lacked the requirement or why supervisors accepted work without alignment records. The method is powerful for simple causal chains but weaker when several causes interact. It should not be used as a substitute for evidence or as a way to force every failure into one linear explanation.

Fishbone Diagrams

A fishbone, or cause-and-effect, diagram organizes possible causes under categories. Common categories include machine, method, material, measurement, people, environment, and management. The categories should be adapted to the problem rather than applied mechanically.

The diagram is primarily a structured brainstorming and classification tool. It helps a team avoid focusing exclusively on the failed component. Each proposed cause must subsequently be tested using records, inspection, measurement, or controlled observation. An attractive diagram containing unverified opinions is not a root-cause analysis.

Fault-Tree Analysis

Fault-tree analysis (FTA) begins with an undesired top event and works backward through combinations of causes. Logical gates represent relationships. An AND gate means that multiple conditions must occur together; an OR gate means that any one of several conditions can produce the event.

For example, “pump fails to deliver flow” may result from “pump unavailable” OR “pump running but insufficient capacity.” “Pump unavailable” may result from motor trip OR coupling failure OR control-system inhibit. The analysis can identify minimal cut sets—smallest combinations of basic events sufficient to produce the top event.

FTA is particularly valuable for high-consequence systems, protection functions, and failures involving multiple pathways. It requires clear system boundaries and reliable estimates if quantitative probabilities are calculated. A fault tree should not imply that all causes are independent when common-cause failures, such as loss of power or poor maintenance, affect several branches.

FMEA and FMECA

Failure Mode and Effects Analysis (FMEA) is a prospective method. It asks what can fail, what effect the failure will have, what causes may produce it, and what controls or actions are appropriate. It is used during design, process development, commissioning, and maintenance-strategy review.

A typical FMEA worksheet contains item or function, failure mode, effect, cause, existing controls, detection method, severity, occurrence, and recommended action. FMECA extends FMEA by adding criticality analysis, usually through a risk ranking based on severity and likelihood, sometimes including detectability.

Traditional risk priority numbers multiply ordinal ratings for severity, occurrence, and detection. This can obscure important differences: two very different risk profiles may produce the same product, and a low occurrence rating should not automatically neutralize catastrophic severity. A risk matrix, criticality ranking, or decision rules that prioritize safety and regulatory consequences can supplement or replace a simple numerical product.

FMEA is most effective when it is treated as a living engineering document. It should be revised after design changes, repeated failures, process changes, and significant learning from field performance. It should not become a form completed only to satisfy an audit.

From Corrective Action to Prevention

Corrective action should be matched to the cause. Replacing a component is appropriate when the component itself is defective, but replacement alone is weak when the defect resulted from contamination, overload, design error, or incorrect installation. The hierarchy of controls is useful: eliminate the hazard or failure opportunity where possible; redesign to make the system more robust; introduce protective devices and interlocks; improve procedures and training; and use inspection or warnings as supporting controls.

A strong action statement identifies the action, owner, due date, required resources, and verification method. “Train operators” is incomplete. A stronger action is: “Revise the pump start-up procedure to include suction-valve position and minimum-flow verification; train all shift operators; audit ten start-ups over four weeks with zero deviations as the acceptance criterion.”

Effectiveness verification should measure the outcome, not merely completion. Suitable indicators include recurrence rate, mean time between failures, defect frequency, alarm history, energy use, product quality, and audit compliance. A failure that does not recur during a short observation period is not necessarily permanently solved; the review period should reflect the operating cycle and risk.

Application in Maintenance and Production Management

Failure analysis supports several managerial decisions. It helps determine whether a component should be redesigned, whether a preventive task is justified, whether condition monitoring can detect degradation, whether a spare should be stocked, and whether an asset remains economically viable. It also informs maintenance budgets by distinguishing recurring symptoms from systemic problems.

Production managers must protect investigation time. Under pressure to restore output, teams may discard evidence, restart equipment without recording conditions, or accept temporary repairs as permanent solutions. A balanced response separates immediate containment from the later analysis required for prevention.

Human performance should be examined without adopting a blame-centered approach. An operator’s action may be the final visible link in a chain involving ambiguous displays, conflicting priorities, inadequate staffing, poor access, or insufficient training. Accountability remains important, particularly for deliberate violations, but fair investigation asks what conditions made the error possible and what system controls were absent.

Digital Tools, Data, and Emerging Practice

Computerized maintenance management systems provide work-order histories, failure codes, labor records, and parts consumption. When failure codes are standardized, these data support Pareto analysis and reliability improvement. Poorly designed codes, however, create false precision. “Mechanical failure” is not sufficiently informative to support causal learning.

Condition-monitoring technologies—vibration analysis, thermography, oil analysis, motor-current signature analysis, acoustic emission, and process analytics—can reveal degradation before functional failure. Their value depends on sensor placement, data quality, alarm configuration, and competent interpretation. Predictive analytics and machine learning can identify patterns in high-dimensional data, but correlation does not prove root cause. Models may also inherit bias from incomplete maintenance records or changes in operating regime.

Digital systems should therefore support, not replace, engineering judgment. Analysts must preserve data integrity, control access, document model assumptions, and avoid using automated recommendations without considering safety and process context. Cybersecurity is also relevant when connected assets influence protective or control functions.

Illustrative Case Study: Repeated Failure of a Cooling-Water Pump

Background

A chemical-processing facility used two electrically driven centrifugal pumps in a duty-standby arrangement to circulate cooling water. The pumps were nominally interchangeable. Over six months, the duty pump experienced four mechanical-seal failures, each causing leakage, an unplanned changeover, and approximately two hours of reduced production.

Situation and Initial Interpretation

The initial maintenance response replaced the mechanical seal. A maintenance report classified each event as “seal failure.” Because the pump was restored quickly, no detailed analysis was conducted. A Pareto chart later showed that mechanical-seal failures were the largest contributor to pump downtime.

Analysis

A cross-functional team reviewed work orders, operating logs, alignment records, seal photographs, vibration readings, and process data. The failed seals showed heat checking and circumferential wear rather than a single manufacturing defect. Operators reported that the pump occasionally lost discharge pressure during start-up.

Using a fishbone diagram, the team considered pump design, seal material, water quality, alignment, suction conditions, operating procedure, and instrumentation. The five-whys analysis proceeded as follows:

• Why did the seal fail? It operated with excessive heat and intermittent dry running.
• Why did dry running occur? The pump lost adequate liquid at the seal during start-up.
• Why was liquid supply unstable? A suction strainer partially blocked and the minimum-flow line was not opened during the start sequence.
• Why was the strainer condition not detected and the line omitted from start-up? The procedure did not specify differential-pressure inspection or minimum-flow verification.
• Why did the procedure omit these controls? The operating standard had been copied from a similar but differently configured pump, and the design-change review had not involved operations and maintenance.

The analysis also found that the standby pump had a different suction arrangement and did not experience the same failure. This comparison provided useful evidence against the hypothesis that the seal material alone was defective.

Strategy and Solution

Immediate actions included cleaning the strainer, inspecting the suction line, replacing damaged seals, and verifying alignment. Preventive actions included revising the start-up procedure, installing a differential-pressure indicator across the strainer, adding a low-flow alarm, defining minimum-flow requirements, and updating the pump FMEA. Operators and technicians were trained using a practical demonstration. A four-month effectiveness review was established, covering seal leakage, suction pressure, alarm events, and start-up compliance.

Outcome

The facility recorded no further seal failure during the review period, and start-up deviations were identified before damage occurred. The investigation also led to a broader review of procedures copied from other assets. The important improvement was not the seal replacement; it was the combination of design information, operating control, inspection, and documented verification.

Learning Points

The case illustrates that a repeated component failure may be a system failure. Failure codes should distinguish the failed part from the mechanism and cause. Comparison with a similar asset can eliminate weak hypotheses. Corrective actions are strongest when they combine immediate repair, engineering control, procedural improvement, competence development, and effectiveness measurement.

Challenges, Limitations, and Professional Considerations

Root-cause analysis can fail because of incomplete records, confirmation bias, production pressure, premature conclusions, poor facilitation, or inadequate technical knowledge. Teams may stop at the first plausible explanation, confuse correlation with causation, or select an action that is easy to document rather than effective to implement.

There are also limits to certainty. Some failures destroy the evidence needed to identify the initiating event. Multiple causes may interact, and changing operating conditions may make recurrence difficult to reproduce. Conclusions should therefore express the strength of evidence and identify unresolved assumptions.

Ethically, investigation reports must be accurate, fair, and transparent. Safety-critical findings should not be softened to protect performance indicators or individual reputations. Evidence must be preserved, confidential information handled appropriately, and regulatory reporting completed where required. The investigator should distinguish intentional misconduct from human error and system weakness. Professional integrity requires reporting uncertainty rather than presenting speculation as fact.

Best Practices for Effective Failure Analysis

Effective organizations generally adopt the following practices:

• Define failure in functional and measurable terms.
• Preserve failed parts, settings, alarms, and operating data.
• Separate symptoms, failure modes, mechanisms, causes, and effects.
• Use Pareto analysis to prioritize, not to prove causation.
• Combine qualitative tools with physical inspection and quantitative evidence.
• Involve operations, maintenance, engineering, quality, safety, and suppliers when appropriate.
• Give priority to safety, environmental, legal, and high-consequence risks.
• Design actions that change equipment, process, information, or controls—not only worker behavior.
• Assign accountable owners and verify effectiveness after implementation.
• Feed learning into FMEA, maintenance plans, procedures, training, design standards, and spare-parts policy.
• Treat failure data as an organizational learning asset.

Chapter Summary

Failure analysis explains how and why an asset loses its required function. A meaningful investigation distinguishes failure effects, modes, causes, mechanisms, contributing conditions, and root causes. Physical mechanisms such as fatigue, wear, corrosion, thermal degradation, electrical breakdown, and deformation provide the technical foundation for analysis.

Pareto analysis helps prioritize recurring or costly problems. The five whys explores causal chains, fishbone diagrams organize possible causes, fault-tree analysis models combinations leading to an undesired event, and FMEA/FMECA anticipate failure modes and rank their significance. These tools are complementary rather than interchangeable.

A structured investigation defines the problem, preserves evidence, reconstructs the event, tests causes, selects proportionate corrective actions, and verifies effectiveness. In production management, successful root-cause problem solving requires cross-functional participation, fair treatment of human performance, reliable data, and integration with maintenance strategy and design governance. Digital analytics can extend detection and prioritization, but engineering judgment and evidence remain essential.

Key Terms

Failure: Loss of an item’s ability to perform a required function.

Failure effect: Consequence produced by a failure at equipment, process, safety, or business level.

Failure mode: Specific manner in which an item fails.

Failure cause: Condition or event that initiates a failure.

Failure mechanism: Physical or chemical process through which damage develops.

Root cause: Fundamental controllable cause whose removal prevents recurrence or reduces its probability.

Contributing cause: Condition that increases the likelihood or severity of failure.

Failure analysis: Systematic examination of failure evidence to determine mode, mechanism, and causes.

Root-cause analysis: Structured process for identifying underlying causes and developing preventive action.

Pareto analysis: Ranking of problem categories to identify the largest contributors to a selected measure.

Five whys: Iterative questioning method used to move from an event toward deeper causes.

Fishbone diagram: Cause-and-effect diagram that organizes possible causes into categories.

Fault-tree analysis: Deductive method that models how combinations of basic events produce a top event.

FMEA: Prospective analysis of possible failure modes, effects, causes, and controls.

FMECA: FMEA extended with criticality assessment or prioritization.

Minimal cut set: Smallest combination of basic events sufficient to produce a fault-tree top event.

Corrective action: Action intended to remove or control a cause of an identified problem.

Effectiveness verification: Evidence-based confirmation that an action achieved its intended result.

Common-cause failure: Failure in which one condition affects multiple components or protective paths.

Condition monitoring: Measurement of equipment or process variables to detect developing degradation.

Review Questions

Short Answer Questions

1. Distinguish a failure mode from a failure mechanism using an example.

2. Why is Pareto analysis insufficient by itself to establish root cause?

3. What is the purpose of an AND gate in fault-tree analysis?

4. Why should corrective actions include effectiveness verification?

5. State two limitations of the five-whys method.

Long Answer Questions

1. Explain the relationship among failure effect, failure mode, failure cause, failure mechanism, and root cause.

2. Describe a structured root-cause investigation from problem definition through organizational learning.

3. Compare Pareto analysis, fishbone diagrams, and fault-tree analysis in terms of purpose and application.

4. Explain how FMEA and FMECA support maintenance and production-management decisions.

5. Discuss the role of human performance, organizational systems, and ethics in failure investigations.

Analytical / Application Questions

1. A gearbox repeatedly overheats after maintenance. Develop a fishbone analysis and identify the evidence required to test the main branches.

2. A production line has many minor sensor faults but one rare failure that creates major safety consequences. Explain how you would prioritize investigation using Pareto and risk-based reasoning.

3. Using the cooling-water pump case, propose three additional controls and define measurable criteria for verifying their effectiveness.

22. Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes a failure mode from a failure mechanism?

A. A failure mode is the business consequence, while a mechanism is the repair cost.

B. A failure mode is how the item fails, while a mechanism is the physical process producing the damage.

C. A failure mode is always the root cause, while a mechanism is always a human error.

D. A failure mode is a preventive task, while a mechanism is an inspection method.

Correct Answer: B. A failure mode is how the item fails, while a mechanism is the physical process producing the damage.

Question 2

A Pareto chart shows that seal failures account for the greatest pump downtime. What is the most appropriate interpretation?

A. Seal failure is conclusively the root cause of all pump problems.

B. Seal failures should be prioritized for further investigation.

C. All other failure categories can be ignored.

D. The seal material must immediately be replaced with a stronger material.

Correct Answer: B. Seal failures should be prioritized for further investigation.

Question 3

In a fault tree, an AND gate indicates that:

A. Any one input event can produce the top event.

B. The input events are unrelated to the top event.

C. The input events must occur together for the output event to occur.

D. The top event has already been prevented.

Correct Answer: C. The input events must occur together for the output event to occur.

Question 4

Which corrective action best addresses the root cause of repeated pump dry running caused by an inadequate start-up procedure?

A. Replace the seal after every start-up.

B. Increase the number of emergency spare seals.

C. Revise the procedure, add flow verification, train personnel, and audit compliance.

D. Record each failure as operator error.

Correct Answer: C. Revise the procedure, add flow verification, train personnel, and audit compliance.

Question 5

Why can multiplying ordinal severity, occurrence, and detection ratings in FMECA be misleading?

A. It prevents all failures from being recorded.

B. Different risk profiles can produce the same numerical product.

C. It measures only production output.

D. It can be used only for electrical equipment.

Correct Answer: B. Different risk profiles can produce the same numerical product.

Question 6

Which evidence would most strongly help distinguish fatigue fracture from sudden overload fracture?

A. The colour of the equipment nameplate.

B. The number of operators on the shift.

C. The presence of progressive crack-growth features and the crack origin.

D. The age of the maintenance department.

Correct Answer: C. The presence of progressive crack-growth features and the crack origin.

Question 7

A team uses a fishbone diagram and lists poor lubrication as a possible cause. What should it do next?

A. Treat poor lubrication as proven because it appears on the diagram.

B. Delete all other possible causes.

C. Test the possibility using lubrication records, lubricant analysis, inspection, and operating evidence.

D. Assign blame to the last technician who serviced the equipment.

Correct Answer: C. Test the possibility using lubrication records, lubricant analysis, inspection, and operating evidence.

Question 8

In the illustrative pump case, comparison with the standby pump was valuable because it:

A. Proved that all mechanical seals were defective.

B. Provided a contrasting operating and design condition that helped eliminate weak hypotheses.

C. Removed the need to examine operating procedures.

D. Showed that preventive maintenance was unnecessary.

Correct Answer: B. Provided a contrasting operating and design condition that helped eliminate weak hypotheses.

Question 9

Which statement about digital predictive analytics is most accurate?

A. A statistical correlation automatically proves physical root cause.

B. Machine-learning output can replace engineering judgment in safety decisions.

C. Analytics can identify patterns, but data quality, context, and causal validation remain necessary.

D. Digital tools eliminate the need to preserve failed components.

Correct Answer: C. Analytics can identify patterns, but data quality, context, and causal validation remain necessary.

Question 10

Which investigation practice is most consistent with professional ethics?

A. Suppressing uncertainty to make the report appear decisive.

B. Focusing exclusively on individual blame.

C. Preserving evidence, distinguishing facts from assumptions, and reporting uncertainty honestly.

D. Ignoring regulatory consequences until production has recovered.

Correct Answer: C. Preserving evidence, distinguishing facts from assumptions, and reporting uncertainty honestly.