Problem Definition, Scoping and RCA Planning

Introduction

Root cause analysis (RCA) is often described as a method for discovering why an undesirable event occurred. That description is incomplete. RCA is also a decision discipline: it determines what problem is being investigated, which evidence is relevant, how much investigation is justified, who should participate, and what constitutes a credible conclusion. Poorly defined investigations may produce technically impressive analyses of the wrong problem. Conversely, a well-scoped investigation can reach a useful conclusion with modest resources.

In Total Quality Management (TQM), RCA is connected to customer focus, process management, evidence-based decision-making, prevention, and continual improvement. The objective is not to identify a person to blame. It is to understand how the system allowed a deviation to occur and to develop controls that prevent recurrence without creating disproportionate cost or bureaucracy.

This chapter presents a structured approach to defining, scoping, and planning an RCA. It examines problem statements, boundaries, baselines, risk and impact, stakeholders, investigation teams, charters, and the selection of appropriate RCA depth. A detailed illustrative case demonstrates how these elements are integrated.

Learning Objectives

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

1. Distinguish symptoms, problems, causes, consequences, and risks.
2. Construct precise, evidence-based problem statements.
3. Define investigation boundaries using the 5W2H framework and process interfaces.
4. Establish baselines and quantify impact where appropriate.
5. Assess risk using severity, occurrence, and detectability considerations.
6. Identify stakeholders and create a cross-functional RCA team.
7. Select an appropriate investigation depth and analytical method.
8. Develop an RCA charter that governs purpose, scope, evidence, roles, timing, and outputs.
9. Recognize ethical, cultural, governance, and technological issues in RCA planning.

1. The Logic of RCA Planning

An RCA begins before any fishbone diagram, fault tree, or “five whys” exercise is used. Planning establishes the analytical frame. A useful logical sequence is:

Observation → Problem definition → Risk and impact assessment → Scope and boundaries → Team and governance → Evidence plan → Analysis → Corrective action → Verification

Each stage constrains the next. If the observation is vague, the problem statement will be vague. If the scope is too broad, the team may collect excessive data without reaching a conclusion. If the scope is too narrow, important contributing conditions may be excluded. If risk is ignored, resources may be allocated according to convenience rather than consequence.

RCA planning should therefore answer six questions:

• What exactly happened or failed?
• What should have happened?
• When, where, and under what conditions did the deviation occur?
• What are the actual and potential consequences?
• Which parts of the system are inside and outside the investigation?
• What level of evidence and analysis is proportionate to the risk?

RCA is not the same as troubleshooting. Troubleshooting seeks to restore operation, often under time pressure. RCA seeks to explain why the failure occurred and why existing controls did not prevent or detect it. The two activities may occur in parallel, but they have different purposes.

2. Foundational Concepts and Terminology

A common source of confusion is the use of “cause” to describe several different ideas. The following distinctions improve analytical precision.

The distinction between a cause and a condition is particularly important. “Operator forgot to inspect” may describe an immediate condition, but it does not explain why forgetting was possible or why the process depended on memory. A deeper inquiry might examine workload, unclear instructions, interface design, training adequacy, supervision, staffing, or conflicting performance targets.

Causal contribution rather than single-cause thinking

Many organizational failures are multicausal. A useful RCA does not assume that every event has one isolated cause. Instead, it identifies a causal structure involving:

• initiating conditions;
• enabling conditions;
• failed or absent controls;
• organizational influences; and
• consequences.

The objective is not to produce the longest possible list of causes. It is to identify evidence-supported causes whose removal or control is likely to prevent recurrence.

3. Constructing a Precise Problem Statement

A problem statement is a neutral, concise description of the deviation. It should not contain an untested explanation. Statements such as “poor training caused defects” or “the supplier was careless” are hypotheses, not problem statements.

A strong problem statement generally includes:

1. What happened?
2. Where did it happen?
3. When did it happen?
4. How large or frequent was the deviation?
5. What requirement or expectation was not met?
6. What is the known impact, if established?

A practical structure is:

Between [date or period], [process/product/service] experienced [observable deviation] at [location or process stage]. The actual performance was [measured result] compared with [required or baseline result], affecting [population, customer, safety, compliance, cost, or delivery dimension].

Example of progressive refinement

Weak: “The packing process is unreliable.”

Improved: “Customer orders are sometimes shipped with incorrect quantities.”

Precise: “During April, 37 of 4,820 retail orders released from the East Distribution Centre contained quantity discrepancies, compared with the required error-free shipment condition. Twenty-nine discrepancies were detected by customers after delivery; eight were detected during internal review.”

The precise version identifies a period, process, location, count, denominator, requirement, and escape information. It does not claim why the discrepancies occurred.

Problem versus objective

The problem statement describes the current undesirable condition. The objective describes the desired future condition. Combining them can create premature commitment to a solution.

• Problem: “Twenty-nine quantity discrepancies escaped internal detection during April.”
• Objective: “Reduce escaped quantity discrepancies to zero sustained occurrences and establish reliable detection before dispatch.”

The objective should be measurable where possible, but the target must not be confused with the cause.

Avoiding judgmental language

Words such as “careless,” “lazy,” “incompetent,” and “uncooperative” are usually poor analytical terms because they attribute moral or personal judgment without specifying observable conditions. Professional RCA replaces them with evidence-based descriptions, for example, “the instruction did not specify the verification sequence” or “the available staffing level was below the workload forecast for four consecutive shifts.”

4. Scoping the Investigation

Scope establishes the boundaries within which evidence will be collected and conclusions will be drawn. Scope is neither a bureaucratic formality nor a way to avoid inconvenient evidence. It is a mechanism for maintaining analytical focus while preserving relevant system context.

4.1 The 5W2H framing tool

The 5W2H questions are useful during initial scoping:

• What: What failed, and what requirement was violated?
• Why: Why is the investigation being initiated now? This question concerns purpose, not presumed cause.
• Where: Which sites, process steps, products, systems, or interfaces are involved?
• When: What period, shift, batch, version, or event window is relevant?
• Who: Which customers, employees, suppliers, functions, or authorities are affected?
• How: How was the deviation detected, and how does it manifest?
• How much: What is the frequency, magnitude, duration, cost, or exposure?

The answers should be supported by initial evidence and marked as confirmed, estimated, or unknown.

4.2 Inclusion and exclusion boundaries

A charter should state what is included and excluded. For example:

Included: order entry, pick-list generation, picking, verification, packing, dispatch release, and relevant system records for the East Distribution Centre during April.

Excluded initially: unrelated product-damage complaints, overseas distribution centres, and historical events before the system upgrade, unless evidence shows that they are causally connected.

Exclusions should be revisited if new evidence indicates that an external condition affects the problem. An exclusion is a planning decision, not a permanent declaration that the topic is irrelevant.

4.3 Process boundaries and interfaces

Failures frequently occur at handoffs: between sales and planning, laboratory and production, supplier and receiving, or maintenance and operations. A process map should therefore include at least one upstream step and one downstream step beyond the visible failure point. This prevents the team from treating an interface failure as an isolated departmental problem.

Boundary questions include:

• Who owns the input and output at each handoff?
• What information is transferred?
• What acceptance criteria apply?
• What happens when information is missing or contradictory?
• Which system is the authoritative record?
• What escalation path exists when a condition exceeds limits?

4.4 Time boundaries and changing conditions

A time boundary should reflect relevant changes such as equipment modifications, software releases, staffing changes, supplier changes, new procedures, or seasonal demand. Comparing periods without considering such changes can produce misleading conclusions.

5. Baselines, Evidence and Quantifying the Deviation

A baseline is the reference condition against which performance is compared. It may be a specification, service-level agreement, control limit, historical performance level, design requirement, regulatory limit, or customer expectation.

A baseline should state:

• the metric and its operational definition;
• the measurement period;
• the population or denominator;
• the data source;
• the required or expected value; and
• known limitations of the measurement system.

For a rate, the general calculation is:

The numerator is the number of defined events, and the denominator is the number of relevant opportunities during the stated period. The result is expressed as a percentage when multiplied by 100. The calculation assumes that numerator and denominator refer to the same population, time period, and event definition.

For example, if 37 orders with discrepancies occurred among 4,820 orders, then:

The rate alone is insufficient. The team should also report that 29 of the 37 discrepancies escaped to customers. The escape proportion is:

This calculation assumes that all 37 discrepancies were identified and that “escaped” is consistently defined as detected after dispatch. It does not establish causation; it establishes the scale and control performance of the observed problem.

Data quality and measurement-system considerations

A baseline may be distorted by inconsistent inspection, missing records, changes in definitions, or selective reporting. Before treating a trend as real, the team should ask:

• Is the measurement method stable?
• Were all opportunities recorded?
• Did the detection threshold change?
• Are duplicate events present?
• Are near misses included or excluded?
• Did reporting behavior change after a management intervention?

In quality analysis, an apparent increase in incidents can reflect improved reporting rather than deteriorating performance. Conversely, a decline can reflect under-reporting. Data validation is therefore part of RCA planning.

6. Risk and Impact Assessment

RCA priority should reflect consequence and exposure, not merely the number of complaints. A rare event may warrant urgent investigation if its potential consequences are severe. A frequent low-impact event may require a process improvement project rather than a full formal RCA.

Impact should be considered across multiple dimensions:

• customer safety and product performance;
• regulatory and legal compliance;
• financial loss and cost of poor quality;
• delivery and continuity;
• reputation and trust;
• employee health and safety;
• environmental effects; and
• recurrence and spread to other products or sites.

A simple qualitative risk matrix can classify severity and likelihood as low, medium, or high. Such a matrix supports prioritization but does not replace professional judgment or formal regulatory requirements.

Where an organization uses a numerical risk priority approach, it must define its scales and limitations. A commonly used form is:

Here, S is a severity rating, O is an occurrence rating, and D is a detectability rating. The ratings are ordinal scores defined by the organization, not physical measurements. Therefore, the product should not be interpreted as a true interval-scale probability or loss value. Different combinations can produce the same RPN, and a very high-severity failure may require action even when its total score is modest. Contemporary practice therefore uses RPN, if at all, alongside explicit severity thresholds and management judgment.

Risk assessment should distinguish:

• actual impact: what has already occurred;
• credible worst-case impact: what could occur under plausible conditions; and
• systemic exposure: where else the same weakness may exist.

7. Stakeholders, Roles and Investigation Teams

RCA is a socio-technical activity. Evidence may be distributed across departments, systems, suppliers, and customers. A team composed only of the department where the event occurred may lack independence and may unintentionally normalize local workarounds.

Stakeholder categories

Stakeholders may include:

• process owners and operators;
• customers or customer representatives;
• quality, risk, compliance, and audit functions;
• engineering, maintenance, information technology, procurement, and supply-chain specialists;
• suppliers and contractors;
• senior decision-makers who allocate resources; and
• regulators or notified bodies where required.

Stakeholder identification should consider both authority and knowledge. A person with formal authority may not understand the work as performed, while an operator may possess essential tacit knowledge without decision rights.

Core RCA roles

A robust charter normally assigns at least these roles:

• Sponsor: authorizes the investigation, removes organizational barriers, and accepts major decisions.
• Facilitator or RCA lead: maintains method discipline, coordinates meetings, and protects neutrality.
• Process owner: provides operational context and owns implementation of approved changes.
• Subject-matter experts: contribute technical, regulatory, data, or human-factors knowledge.
• Recorder or analyst: maintains the evidence log, action register, and decision record.
• Affected-user representative: represents customer, operator, patient, or other stakeholder experience where relevant.

The sponsor should not dictate the conclusion. The process owner should not be the sole investigator when the investigation concerns performance under that owner’s control. Independence must be proportionate to risk.

Psychological safety and just culture

A blame-oriented climate suppresses information. Employees may conceal workarounds, near misses, or deviations if they expect punishment for honest reporting. A just-culture approach distinguishes among human error, at-risk behavior, reckless behavior, and system design weaknesses. This distinction does not eliminate accountability; it improves the accuracy of accountability by separating intentional violations from predictable system-induced error.

Interviews should ask what made sense to people at the time, what constraints existed, and how the process normally works—not merely what the procedure says should happen.

8. Selecting RCA Depth and Method

Not every event requires the same analytical investment. A useful tiered model is shown below.

Method selection should follow the problem, not habit. Common methods include:

• Five whys: useful for a simple causal chain, but weak when causes branch or evidence is uncertain.
• Fishbone or cause-and-effect diagram: useful for organizing hypotheses across categories such as method, machine, material, measurement, environment, and people; it does not prove causes.
• Timeline analysis: useful for reconstructing sequence, decision points, and changing conditions.
• Fault tree analysis: a deductive method that begins with a defined top event and identifies combinations of conditions leading to it.
• Event tree analysis: an inductive method that starts with an initiating event and examines possible consequence paths through controls.
• Barrier analysis: examines preventive and mitigative controls and how they failed, were absent, or were bypassed.
• Change analysis: compares normal and abnormal conditions to identify relevant differences.
• Human-factors analysis: examines task design, workload, interfaces, incentives, competence, and organizational conditions.

A sophisticated investigation may combine methods. For example, a timeline can establish sequence, barrier analysis can examine failed controls, and fault-tree logic can test whether multiple conditions were necessary for the event.

9. The RCA Charter

The charter is the governing document for the investigation. It creates a shared contract among sponsor, team, process owner, and stakeholders. A concise charter should contain:

1. Title and unique investigation identifier.
2. Date opened and sponsor.
3. Problem statement and evidence source.
4. Business, customer, safety, compliance, and ethical significance.
5. Objective and success criteria.
6. Scope, boundaries, assumptions, and exclusions.
7. Initial containment actions.
8. Team members, roles, and independence arrangements.
9. Planned methods and evidence sources.
10. Data-protection, confidentiality, and legal requirements.
11. Milestones, decision gates, and reporting arrangements.
12. Required outputs and action-approval authority.
13. Effectiveness-verification plan.

The charter should be treated as a living document. Changes to the problem definition, scope, or risk classification should be recorded with rationale and approval. Silent scope expansion is a frequent cause of delay and conflict.

Evidence plan

Before collecting evidence, the team should identify:

• records to preserve;
• system logs and version histories;
• samples and physical evidence;
• relevant procedures and training records;
• interviewees and interview sequence;
• data definitions and extraction methods; and
• chain-of-custody or access controls where evidence may be challenged.

Digital systems create special risks. Logs may be overwritten, timestamps may use different time zones, and automated records may reflect system configuration rather than actual work. Data should be preserved promptly and interpreted with knowledge of system architecture.

10. Detailed Illustrative Case: Quantity Discrepancies in a Distribution Centre

The following fictional case is an illustrative teaching scenario.

Background

A retail distributor operates a regional fulfilment centre. Orders are released from an enterprise resource-planning system, picked by warehouse employees, checked at packing stations, and dispatched through a carrier. The contractual requirement is that shipped quantities match confirmed order quantities.

Problem

During April, 37 of 4,820 orders had quantity discrepancies. Twenty-nine were discovered by customers after delivery. The organization initially proposed “retraining pickers,” but the quality manager required a structured RCA before approving that solution.

Initial risk and impact assessment

The actual financial impact was moderate, involving credits, re-shipments, and investigation time. However, the escape rate was high, and the same workflow was used at three other sites. The potential systemic exposure was therefore significant. The investigation was classified as Level 2, with escalation to Level 3 if evidence indicated a broader control failure or regulatory/customer risk.

Charter and scope

The scope included order release, pick-list generation, picking, packing verification, and dispatch release at the East Distribution Centre during April. It included system configuration changes made in March. It initially excluded damage claims and sites using a different warehouse-management system.

The team included the distribution process owner, a quality facilitator, two experienced operators from different shifts, an information-systems analyst, a customer-service representative, and a data analyst. The sponsor was the regional operations director.

Evidence collection

The team reviewed order records, pick lists, scanner transactions, packing-station logs, staffing rosters, training records, customer complaints, and system-change tickets. Interviews were conducted with employees from day and night shifts. The team compared work-as-designed with work-as-performed.

The data showed that discrepancies were concentrated in orders containing multiple units of visually similar products. The new scanner interface displayed a confirmation message after the first scan, but the message did not clearly indicate whether the required quantity had been completed. At peak periods, packers sometimes relied on visual counting because the system queue displayed an “order complete” status before the final verification record synchronized.

Analysis

A simple five-whys chain initially produced “picker failed to count correctly.” The team rejected this as incomplete. Timeline and barrier analysis identified several interacting conditions:

1. Similar products increased identification and counting difficulty.
2. The interface provided weak feedback for partial quantities.
3. Synchronization delays created a temporary mismatch between physical and system status.
4. The packing verification step was designed as a confirmation rather than an independent quantity check.
5. Performance reporting emphasized orders processed per hour but did not include escaped quantity errors.
6. The March system change had been tested for transaction completion, not for usability under peak workload.

The immediate error was a quantity mismatch. The deeper causes involved interface design, verification-barrier weakness, change-control testing, and conflicting performance measures. The evidence did not support a conclusion that lack of training was the primary cause.

Approach and solution

Containment included a temporary independent quantity check for high-risk product families and daily review of escaped discrepancies. Corrective actions were divided into layers:

• redesign the scanner interface to show required, scanned, and remaining quantity;
• prevent dispatch release when the verification record is incomplete;
• test the revised workflow under realistic peak-volume conditions;
• revise performance indicators to include quality escapes;
• update change-control criteria to include human-factors and operational usability testing; and
• provide focused coaching on the revised process rather than generic retraining.

Outcome and verification

After implementation, the organization monitored discrepancy rate, escape proportion, system synchronization exceptions, and compliance with the independent check. Verification required sustained performance over an agreed period and review across shifts. A single week without complaints was not accepted as proof of effectiveness.

Learning points

The case demonstrates that:

• a numerical problem statement is stronger than a general complaint;
• high escape rates may indicate control weakness rather than only execution error;
• cross-functional teams reveal interactions between technology, work design, and management measures;
• corrective actions should address causes and failed barriers, not merely retrain individuals; and
• transferability to other sites should be assessed after local causes are understood.

11. Governance, Ethics and Contemporary Technology

RCA records may contain personal data, confidential customer information, trade secrets, or legally sensitive material. Access should be limited according to legitimate need, retention requirements, and applicable privacy law. Interview notes should distinguish factual evidence from interpretation. Organizations should avoid using RCA as an informal disciplinary investigation unless that purpose is explicitly governed and separated from learning-oriented analysis.

AI and advanced analytics can assist with anomaly detection, text clustering, timeline construction, and large-scale record review. They do not remove the need for human judgment. AI-generated patterns may reflect reporting bias, incomplete data, or historical discrimination. Any AI-supported RCA should document data provenance, model limitations, validation procedures, human review, and the consequences of false positives and false negatives. Automated suggestions must not be treated as verified causes.

Contemporary RCA is increasingly concerned with resilience: how systems adapt under variable demand, degraded resources, and unexpected conditions. This perspective complements failure analysis by examining why operations usually succeed and how normal adaptations may become unsafe under particular combinations of pressure and weak controls.

12. Common Planning Errors and Best Practices

Common errors

• Defining the problem as a presumed cause.
• Beginning with a favorite tool rather than a clear question.
• Treating the first plausible explanation as the root cause.
• Ignoring near misses and successful recoveries.
• Failing to include process interfaces.
• Using data without checking definitions or measurement reliability.
• Making “retraining” the default corrective action.
• Allowing the affected department to investigate itself without appropriate independence.
• Expanding the scope without revising time, resources, and governance.
• Closing actions without verifying sustained effectiveness.

Best practices

• Write the problem statement before selecting the method.
• Separate facts, hypotheses, assumptions, and unknowns.
• Use a baseline with a clear denominator and time period.
• Consider actual harm, potential harm, and systemic exposure.
• Include people who perform the work as well as those who design and govern it.
• Preserve evidence early, especially digital records.
• Examine why controls failed or were absent.
• Design actions that are specific, owned, resourced, and measurable.
• Verify effectiveness under normal and stressed operating conditions.
• Capture transferable learning without assuming every site is identical.

Chapter Summary

Effective RCA begins with disciplined planning. A problem statement should describe an observable deviation against a defined requirement without embedding an untested cause. Scoping establishes process, time, location, population, and interface boundaries while allowing revision when evidence changes the understanding of the event. Baselines make the deviation measurable, but data quality and measurement definitions must be examined before conclusions are drawn.

Risk and impact assessment should consider actual and potential consequences, recurrence, escape, and systemic exposure. Stakeholder identification and cross-functional team design are essential because failures often arise at organizational and technological interfaces. The RCA charter provides governance by defining purpose, scope, roles, methods, evidence, milestones, confidentiality, and effectiveness criteria.

Investigation depth should be proportionate to risk and complexity. Simple events may require a local review, whereas serious or systemic events require independent teams, preserved evidence, barrier analysis, human-factors consideration, and formal governance. Ethical RCA avoids blame, protects legitimate confidentiality, and distinguishes individual accountability from system weakness. Technology and AI can support analysis, but their outputs require validation and human oversight.

Key Terms

Problem statement: Evidence-based description of an undesirable deviation.

Baseline: Reference condition used for comparison.

Scope: Defined extent and limits of an investigation.

Boundary: Point specifying what is included or excluded.

Symptom: Observable indication of an underlying problem.

Immediate cause: Condition closest in time to the event.

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

Latent cause: Underlying organizational, design, or governance weakness.

Escape point: Control that failed to detect or prevent a problem.

Containment: Temporary action that limits current impact.

Corrective action: Action intended to eliminate the cause of a nonconformity.

Preventive control: Control intended to reduce occurrence.

Detective control: Control intended to identify a deviation.

RCA charter: Governing document defining an investigation.

Stakeholder: Person or group affected by, knowledgeable about, or authorized over the investigation.

Barrier analysis: Examination of controls that should prevent or mitigate an event.

Measurement system: Method, people, equipment, and rules used to produce data.

Psychological safety: Climate in which people can report information without unreasonable fear.

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

Systemic exposure: Possibility that the same weakness exists elsewhere.

Review Questions

Short Answer Questions

1. What is the difference between a problem statement and a cause hypothesis?
2. Why should an RCA charter define both inclusion and exclusion boundaries?
3. What is an escape point?
4. Why must a baseline include a denominator when a rate is reported?
5. Name four roles commonly assigned in an RCA team.

Long Answer Questions

1. Explain how a precise problem statement improves the quality of subsequent RCA activities.
2. Discuss the relationship between risk assessment and the selection of RCA depth.
3. Compare five-whys analysis, barrier analysis, fault-tree analysis, and human-factors analysis.
4. Explain why cross-functional teams and psychological safety are important in organizational RCA.
5. Describe the contents and governance value of an RCA charter.

Analytical / Application Questions

1. A hospital records 12 medication discrepancies in 6,000 administrations, of which 9 reached patients. Develop a neutral problem statement, calculate the discrepancy rate and escape proportion, and identify additional evidence required before selecting corrective actions.
2. A manufacturing plant experiences repeated defects after a software-controlled equipment upgrade. Design the investigation scope, identify relevant stakeholders, and recommend an RCA depth with justification.
3. An organization proposes retraining employees as the sole response to a recurring service failure. Critically evaluate this proposal using the concepts of failed barriers, work-as-performed, measurement systems, and effectiveness verification.

Multiple Choice Questions with Answers

Question 1

Which statement is the best example of a precise RCA problem statement?

A. The warehouse team is careless.

B. The packing process is unreliable.

C. During April, 37 of 4,820 orders from the East Distribution Centre had quantity discrepancies.

D. Training should be improved immediately.

Correct Answer: C. During April, 37 of 4,820 orders from the East Distribution Centre had quantity discrepancies.

Question 2

What is the primary purpose of an RCA charter?

A. To assign personal blame before evidence is collected.

B. To define the investigation’s purpose, scope, roles, methods, and outputs.

C. To replace the need for data collection.

D. To guarantee that one root cause will be found.

Correct Answer: B. To define the investigation’s purpose, scope, roles, methods, and outputs.

Question 3

If 37 discrepancies occurred among 4,820 orders, what was the approximate discrepancy rate?

A. 0.077%

B. 0.768%

C. 7.68%

D. 76.8%

Correct Answer: B. 0.768%

Question 4

Which item is an escape point?

A. The original customer order

B. A control that should have detected the error but did not

C. The final business impact

D. A proposed corrective action

Correct Answer: B. A control that should have detected the error but did not

Question 5

Why should a problem statement avoid phrases such as “careless employee”?

A. Such phrases are always legally prohibited.

B. They describe an untested judgment rather than an observable condition.

C. They make the problem statement too quantitative.

D. They identify too many possible causes.

Correct Answer: B. They describe an untested judgment rather than an observable condition.

Question 6

Which RCA method begins with a defined top event and works backward through combinations of conditions?

A. Event tree analysis

B. Fault tree analysis

C. Benchmarking

D. Control charting

Correct Answer: B. Fault tree analysis

Question 7

In the expression RPN = S × O × D, what do the symbols represent?

A. Scope, output, and duration

B. Severity, occurrence, and detectability

C. Safety, organization, and design

D. Sample, observation, and deviation

Correct Answer: B. Severity, occurrence, and detectability

Question 8

Which consideration most strongly supports escalating an RCA to a formal systemic investigation?

A. The event is easy to explain locally.

B. The event has serious potential consequences or cross-site exposure.

C. The investigation team prefers a longer report.

D. The process owner requests no external involvement.

Correct Answer: B. The event has serious potential consequences or cross-site exposure.

Question 9

What is the main limitation of using five-whys analysis alone for complex failures?

A. It cannot be written in a charter.

B. It may oversimplify branching and interacting causes.

C. It requires probability calculations.

D. It can only be used for safety events.

Correct Answer: B. It may oversimplify branching and interacting causes.

Question 10

Which action best demonstrates effectiveness verification?

A. Closing the action immediately after training is delivered.

B. Confirming sustained improvement using defined measures under relevant operating conditions.

C. Obtaining verbal agreement that the problem is unlikely to recur.

D. Replacing the investigation team with a manager.

Correct Answer: B. Confirming sustained improvement using defined measures under relevant operating conditions.