10. Quality in Design, Innovation and Service Delivery

Introduction and Context

Quality is often treated as an inspection outcome, but in advanced quality management it is primarily a design outcome. Inspection can identify defects after they have occurred; it cannot reliably compensate for a product architecture, process, or service system that was poorly conceived. The central proposition of this chapter is that quality must be designed into products, services, technologies, and processes from the earliest stages of innovation and then protected throughout delivery and use.

Design decisions establish much of the eventual cost, performance, reliability, safety, usability, environmental impact, and customer experience of an offering. A late engineering change may require redesign of tooling, supplier contracts, software, training, documentation, and regulatory approvals. Similarly, a service promise that cannot be supported by capacity, information systems, employee capability, or recovery procedures will create dissatisfaction even when employees work diligently. Quality therefore requires integration among customer insight, design thinking, engineering analysis, operations, suppliers, employees, and governance.

This chapter examines the principal methods used to build quality into design and delivery. These include design thinking, quality function deployment (QFD), design for quality and manufacturability, failure mode and effects analysis (FMEA), reliability engineering, robust design, design of experiments (DOE), verification and validation, and service recovery. The methods are complementary rather than interchangeable. QFD translates customer needs into design requirements; FMEA anticipates risks; DOE studies causal relationships; reliability engineering addresses performance over time; verification and validation provide evidence; and service recovery restores value when failure reaches the customer.

Learning Objectives

After studying this chapter, students should be able to:

1. Explain why quality should be designed into products, services, and processes rather than added through inspection.

2. Distinguish design thinking, QFD, FMEA, reliability engineering, robust design, DOE, verification, and validation.

3. Translate customer requirements into measurable technical and service characteristics.

4. Apply risk-based methods to identify, prioritize, and control potential failures.

5. Interpret reliability concepts and select suitable design and testing approaches.

6. Explain how experiments can improve process capability and product performance.

7. Evaluate service failures and formulate effective recovery strategies.

8. Discuss managerial, ethical, digital, and governance issues in quality-oriented innovation.

Quality as an Attribute of the Designed System

A useful distinction is between conformance quality and quality of design. Conformance quality concerns whether an output meets specified requirements. Quality of design concerns whether the requirements themselves reflect legitimate customer, regulatory, operational, and strategic needs. A perfectly manufactured product may still fail if it is inconvenient, unsafe, unaffordable, unreliable, or irrelevant to users.

The design process converts relatively uncertain needs into increasingly precise decisions. At the beginning, customers may express needs in ambiguous language—for example, “the device should be easy to use” or “the service should be dependable.” Designers must interpret these statements, identify measurable attributes, establish priorities, and understand trade-offs. A quality-oriented design process asks not only whether a requirement can be achieved, but also whether it is necessary, verifiable, economically sustainable, and ethically acceptable.

Cross-functional participation is essential because quality is multidimensional. Marketing understands customer segments; engineers understand technical feasibility; operations understands process capability; procurement understands supplier constraints; finance understands cost; legal and compliance specialists understand obligations; and service employees understand actual use conditions. Without such integration, local optimization may produce global failure. For example, a cost reduction in component specification may increase warranty claims and service workload.

Design Thinking: Understanding Before Defining

Design thinking is a human-centred approach to innovation that combines user understanding, problem framing, idea generation, prototyping, and learning. Its value in quality management lies in reducing the risk of solving the wrong problem. Conventional development may begin with a proposed technical solution. Design thinking begins with users, contexts, behaviours, and unmet needs.

The commonly used stages are empathize, define, ideate, prototype, and test. These stages are iterative rather than strictly sequential. Observation, interviews, journey mapping, contextual inquiry, and analysis of complaints can reveal needs that customers cannot easily articulate. The define stage converts observations into a focused problem statement. Ideation deliberately generates alternatives before premature selection. Prototypes make assumptions visible and allow inexpensive learning. Testing with representative users reveals usability, accessibility, emotional, and functional issues.

Design thinking should not be confused with informal creativity. Its discipline depends on evidence, explicit assumptions, rapid learning, and willingness to revise the problem definition. It is especially useful where requirements are uncertain, user experience is important, or innovation involves new behaviours. However, it must eventually be connected to engineering specifications, risk analysis, regulatory requirements, and operational controls. A desirable prototype is not automatically a safe, reliable, scalable, or economically viable product.

Translating the Voice of the Customer through QFD

Quality Function Deployment is a structured method for translating the voice of the customer into technical and operational requirements. Its principal instrument is often called the House of Quality, although QFD may be applied through a series of linked matrices across product, component, process, and production planning.

The initial matrix typically includes customer requirements, their importance, competitive perceptions, technical characteristics, relationships between customer and technical requirements, technical correlations, and target values. The process requires careful interpretation. A customer requirement such as “quiet operation” may be related to vibration, motor speed, enclosure design, acoustic insulation, and control algorithms. The technical team must determine which characteristics can be measured and controlled.

QFD provides several benefits. It makes trade-offs explicit, reduces the risk that important customer needs will be overlooked, and creates a common language across functions. It can also expose conflicts: improving strength may increase weight; increasing speed may reduce energy efficiency; reducing service time may require additional capacity. The purpose is not to eliminate all trade-offs but to make them visible and manage them consciously.

QFD is limited when customer data are weak, requirements are unstable, or relationships are assigned mechanically. Numerical priority scores should support managerial judgment, not replace it. Customer needs involving safety, privacy, accessibility, or legal compliance should not be downgraded merely because their survey importance is low. The method is strongest when supported by interviews, usage data, complaint analysis, benchmarking, and technical expertise.

Design for Quality and Prevention at Source

Design for quality means creating product and process characteristics that make correct performance more likely and incorrect performance less likely. It includes design for manufacturability, assembly, serviceability, maintainability, safety, environment, accessibility, and supply-chain capability.

A design for manufacturability review examines whether specifications can be produced consistently using available equipment, materials, skills, and measurement systems. Excessively tight tolerances may provide no customer benefit while increasing cost and variation. Design for assembly reduces part count, simplifies orientation, prevents incorrect assembly, and supports ergonomic work. Design for serviceability considers access to components, diagnostic information, replacement procedures, and safe maintenance. In software and digital services, equivalent principles include modular architecture, testability, observability, secure defaults, and controlled deployment.

Prevention at source is superior to reliance on final inspection. A connector that can be inserted incorrectly should be redesigned with mistake-proofing, rather than depending on an inspector to detect every error. A digital form should prevent invalid entries where possible, rather than allowing errors to pass to a later administrative review. Such measures reduce dependence on memory, vigilance, and individual heroics.

Design reviews should be evidence-based and staged. Concept reviews examine customer value and feasibility; preliminary reviews examine architecture and major risks; detailed reviews examine specifications, interfaces, tolerances, and controls; and readiness reviews assess whether the design can be released, produced, delivered, supported, and changed under control.

FMEA: Anticipating Failure Before It Reaches the Customer

Failure Mode and Effects Analysis is a systematic method for identifying how a product, process, or service might fail, why it might fail, what effects the failure could produce, and how the risk can be reduced. Design FMEA focuses on product or system design. Process FMEA focuses on manufacturing, transaction, or service processes. Other variants address software, machinery, and healthcare processes.

A typical FMEA identifies the function, potential failure mode, effect, cause, current prevention and detection controls, and recommended action. Teams commonly assess severity, occurrence, and detectability using ordinal scales. The traditional risk priority number is calculated as:

RPN = Severity × Occurrence × Detection

RPN is useful for structuring discussion but has important limitations. Different combinations of scores can produce the same number, although their managerial significance may differ. A low-occurrence failure with catastrophic consequences should not be ignored. Contemporary practice therefore gives priority to severity, critical characteristics, regulatory requirements, and action effectiveness in addition to numerical ranking.

An effective FMEA is not a document completed once to satisfy an audit. It is a living risk-management activity. Actions should preferably change the design or process to eliminate the cause, reduce the likelihood, or make the failure impossible. Detection controls are valuable but generally weaker than prevention. After action implementation, the team should reassess the risk and verify that the control works under realistic conditions.

FMEA also has a social dimension. Team members must be able to report risks without fear of blame. If production targets discourage disclosure, the resulting FMEA may look complete while concealing important knowledge. Psychological safety and leadership behaviour therefore influence the technical quality of risk analysis.

Reliability Engineering and Performance over Time

Reliability is the probability that an item performs its required function under stated conditions for a specified period. It is therefore different from durability, which refers more specifically to the ability to resist deterioration, and availability, which considers both reliability and restoration after failure.

Reliability analysis begins by defining the function, operating environment, mission time, and acceptable failure criteria. Important tools include reliability block diagrams, fault-tree analysis, life testing, accelerated testing, failure data analysis, and maintainability analysis. A series system fails if any critical component fails; a redundant system may continue operating when one component fails. These architectural choices have implications for safety, cost, complexity, and maintenance.

The failure rate of many products is represented conceptually by the bathtub curve: early failures associated with defects or installation problems, a relatively stable period of random failures, and a later period of wear-out. The curve is not universal, and it should not be assumed without evidence. Reliability programmes must use actual failure mechanisms and operating data.

Reliability growth is achieved through iterative testing, root-cause analysis, corrective design changes, and retesting. Merely increasing the duration of testing without learning from failures does not create reliability. Warranty data, field returns, sensor data, maintenance records, and customer complaints can provide valuable evidence, but data quality and reporting bias must be considered.

Reliability decisions have ethical consequences. In products affecting health, transport, infrastructure, or personal safety, acceptable risk cannot be determined solely by cost-benefit calculations. Traceability, configuration control, recall procedures, and transparent communication are essential components of responsible reliability management.

Robust Design and Design of Experiments

Robust design seeks performance that is relatively insensitive to uncontrollable variation, often called noise factors. Noise may arise from temperature, humidity, user behaviour, material variation, equipment ageing, or differences among suppliers. Instead of attempting to control every source of variation, the designer selects settings and configurations that reduce sensitivity.

Taguchi’s approach distinguishes control factors, noise factors, signal factors, and quality characteristics. Although particular statistical formulations should be used carefully, the broader principle is powerful: designs should perform consistently across realistic conditions, not only under nominal laboratory settings. Robustness is closely related to loss prevention because variation around a target can create customer dissatisfaction even when results remain within specification limits.

Design of experiments provides a structured way to study cause-and-effect relationships. Unlike one-factor-at-a-time experimentation, factorial designs can examine several factors simultaneously and reveal interactions. An interaction exists when the effect of one factor depends on the level of another. Screening designs can identify influential factors; response-surface methods can optimize settings; and confirmation runs can test whether the predicted improvement is real.

A sound DOE requires a clearly defined response, appropriate factor ranges, randomization where possible, replication or suitable estimation of error, and attention to measurement-system capability. Statistical significance is not the same as practical significance. A small improvement may be statistically detectable but operationally unimportant, while a clinically or commercially important effect may require a larger sample. Experimental conclusions must also respect physical knowledge and process context.

Verification, Validation, and Controlled Learning

Verification asks whether the design output meets specified requirements: “Was the product or service designed correctly?” Validation asks whether the resulting offering meets intended user needs and is suitable for its actual use environment: “Was the correct product or service designed?”

Verification may include design calculations, code review, inspection, simulation, laboratory testing, and review of technical documentation. Validation may include user trials, pilot operations, field testing, usability evaluation, clinical or operational evaluation, and assessment under representative conditions. Both activities require predefined acceptance criteria, competent methods, controlled configurations, documented results, and treatment of deviations.

A frequent mistake is to validate with ideal users, ideal conditions, or an incomplete version of the service. Realistic validation should include foreseeable misuse, accessibility needs, workload pressures, environmental variation, and interfaces with other systems. For digital products, validation may also examine cybersecurity, privacy, algorithmic performance, explainability, and the consequences of automation errors.

Change control is part of verification and validation. A software update, supplier substitution, material change, or process relocation may invalidate earlier evidence. Impact assessment should determine whether partial or complete re-verification and re-validation are required. Release decisions should be based on evidence and residual risk, not schedule pressure alone.

Quality in Service Delivery and Service Recovery

Services differ from physical products because production and consumption may occur simultaneously, the customer may participate in the process, and outcomes are often variable and intangible. Quality therefore includes technical outcome, reliability, responsiveness, assurance, empathy, accessibility, communication, and the effort required from the customer.

Service design should map the customer journey and the underlying service blueprint. The blueprint distinguishes customer actions, visible employee actions, backstage activities, support processes, systems, and failure points. Measures should include both outcome indicators—such as resolution quality or clinical result—and process indicators—such as waiting time, first-contact resolution, accuracy, and accessibility.

Service recovery is the planned response when a service failure occurs. Effective recovery begins with rapid recognition and acknowledgement. It then requires diagnosis, a fair remedy, communication of next steps, and learning to prevent recurrence. The remedy may involve correction, replacement, refund, priority handling, explanation, or escalation, depending on the nature and impact of the failure. Employees need authority boundaries, information, training, and support to act consistently.

An apology alone is inadequate when the customer has suffered material loss or repeated inconvenience. Conversely, excessive compensation without correcting the underlying process can encourage dependency and conceal systemic problems. Recovery should therefore combine immediate restoration with root-cause analysis, trend monitoring, and redesign. Complaint data should be treated as a source of process knowledge rather than merely a measure of employee performance.

An Illustrative Case: Redesigning a Hospital Appointment Service

Background

The following fictional case illustrates the integration of the methods discussed in this chapter. A regional hospital introduced an online outpatient appointment service intended to reduce telephone traffic and waiting time.

Situation and Problem

Within three months, patients reported duplicate bookings, unclear instructions, inaccessible forms, missed reminders, and difficulty obtaining assistance. The hospital initially proposed adding more call-centre staff. However, complaints continued, and staff reported that patients often arrived without required documents or attended the wrong clinic.

Analysis

A cross-functional team used interviews and journey mapping to understand the patient experience. The team found that “easy booking” involved several distinct needs: finding the correct specialty, selecting a suitable time, understanding preparation requirements, receiving confirmation, and obtaining help when circumstances changed. QFD translated these needs into technical and service characteristics such as search accuracy, appointment confirmation time, reminder delivery rate, accessibility compliance, and escalation response time.

An FMEA identified failure modes including selection of the wrong clinic, incomplete patient information, duplicate submission, reminder failure, and appointment cancellation not reaching the patient. Severity was high for failures that could delay urgent care. The team therefore did not rely solely on RPN; high-severity failures received priority even when their estimated occurrence was moderate.

Usability validation with patients revealed that the original interface used clinical terms unfamiliar to many users. A prototype using plain language, guided questions, visible progress indicators, and a confirmation summary performed better. Verification tested data validation, integration with the hospital scheduling system, security controls, and message delivery. Validation was conducted with older users, users with disabilities, staff, and patients using mobile devices.

Approach and Solution

The hospital introduced a guided booking process, duplicate-booking warnings, multilingual instructions, accessible design, confirmation through more than one channel, and a staffed escalation path. A pilot was run in one specialty before wider deployment. Complaint categories and failed transactions were reviewed weekly, and changes were controlled through documented release procedures.

Outcome

The redesigned service reduced avoidable booking errors and improved staff understanding of recurring failure causes. More importantly, the hospital moved from treating complaints as isolated incidents to managing the appointment process as a designed system. The case does not imply that digitalization alone creates quality; quality resulted from user research, risk analysis, testing, operational readiness, and continuous learning.

Learning Points

The case demonstrates five principles: customer language must be translated carefully; high-severity risks require explicit attention; usability is part of quality; verification and validation answer different questions; and service recovery should generate design and process improvement.

Managerial and Organizational Implications

Quality-oriented design changes managerial priorities. Leaders must protect time for learning before full-scale launch, encourage cross-functional challenge, and allocate resources to prevention even when short-term financial benefits are difficult to show. Governance should define decision rights, risk acceptance authority, evidence requirements, and escalation routes.

Portfolio management is also important. Not every attractive innovation should be pursued. Leaders should assess strategic fit, customer value, technical uncertainty, regulatory exposure, lifecycle cost, sustainability, and organizational capability. Stage-gate reviews can control investment, but gates should be learning points rather than bureaucratic approvals. A project may appropriately stop, pivot, or return to discovery when evidence contradicts its assumptions.

Supplier involvement is essential where external components, cloud services, algorithms, or logistics affect quality. Contracts should define specifications, change notification, audit rights, data responsibilities, incident reporting, and continuity expectations. Suppliers should be involved early enough to influence design rather than merely being asked to meet an already fixed and impractical specification.

Digital Tools, Analytics, and Artificial Intelligence

Digital tools support quality design through computer-aided engineering, simulation, digital twins, requirements-management systems, configuration control, automated testing, process mining, customer analytics, and connected-product monitoring. These tools can reduce iteration time and reveal patterns that manual analysis might miss.

Artificial intelligence can assist in classifying complaints, predicting failures, generating design alternatives, detecting anomalies, and supporting user research. Nevertheless, AI outputs require verification. Training data may be incomplete or biased; models may drift; explanations may be inadequate; and automation may amplify an incorrect assumption. In safety-critical or high-consequence contexts, human accountability, auditability, data governance, cybersecurity, privacy, and model-change control are indispensable.

Analytics should support decisions rather than create false precision. A prediction of failure risk is not itself a root cause, and a dashboard metric is not necessarily a customer outcome. Digital quality systems must preserve the connection between data, physical or service reality, and responsible managerial judgment.

Challenges and Common Mistakes

Several recurring mistakes weaken quality in design and delivery:

• Treating customer statements as complete specifications without investigation.
• Using QFD or FMEA as paperwork exercises rather than decision tools.
• Prioritizing schedule or launch pressure over unresolved high-severity risks.
• Confusing verification with validation.
• Testing only nominal conditions and ideal users.
• Relying on inspection or detection when prevention is feasible.
• Assuming that statistical significance guarantees business or customer significance.
• Ignoring maintenance, support, disposal, accessibility, and lifecycle impacts.
• Failing to reassess risks after design, supplier, software, or process changes.
• Measuring employee compliance while overlooking defective system design.

There are also unavoidable trade-offs. More redundancy may improve reliability but increase cost and complexity. More customization may improve perceived value but reduce process stability. Rapid experimentation may accelerate learning but create governance and data risks. Quality management requires explicit decisions about these trade-offs, supported by evidence and aligned with customer welfare and organizational purpose.

Best Practices for Integrated Quality Design

An effective organization generally:

1. Begins with observed customer and user needs, including the needs of vulnerable and less visible groups.

2. Converts needs into measurable, prioritized, and traceable requirements.

3. Uses cross-functional design reviews and includes operations, service, suppliers, and maintenance perspectives early.

4. Applies FMEA and related risk methods before failures occur and updates them after learning.

5. Designs for manufacturability, usability, serviceability, accessibility, safety, security, and sustainability.

6. Uses DOE, simulation, prototypes, and pilot operations to learn efficiently.

7. Separates and documents verification and validation activities.

8. Tests realistic environments, users, interfaces, and foreseeable misuse.

9. Establishes controlled release, configuration, and change-management practices.

10. Treats complaints, warranty data, and incidents as inputs to redesign and systemic improvement.

11. Uses digital analytics with attention to data quality, bias, privacy, cybersecurity, and human accountability.

12. Reviews quality over the entire lifecycle, from concept through use, maintenance, and retirement.

Chapter Summary

Quality is primarily designed into products, services, and processes; inspection alone cannot overcome weak design. Design thinking helps organizations understand users and frame the right problem. QFD converts the voice of the customer into technical and operational requirements while making trade-offs visible. Design for quality emphasizes prevention, manufacturability, serviceability, safety, accessibility, and mistake-proofing. FMEA anticipates failure modes and prioritizes risk-reduction actions, while reliability engineering addresses performance over time and under stated conditions. Robust design reduces sensitivity to variation, and DOE provides an efficient method for learning about causal factors and interactions. Verification determines whether requirements have been met; validation determines whether the offering is suitable for intended use. In service delivery, journey mapping, service blueprints, recovery processes, and complaint learning are essential. Digital tools and AI can strengthen quality decisions but require governance, evidence, security, privacy, and human accountability. The mature quality organization integrates these methods into a lifecycle system of learning, prevention, controlled change, and responsible innovation.

Key Terms

Design thinking: A human-centred, iterative approach involving user understanding, problem definition, ideation, prototyping, and testing.

Quality Function Deployment (QFD): A structured method for translating customer requirements into technical and operational characteristics.

Voice of the customer: Explicit and implicit customer needs, expectations, preferences, and concerns gathered through suitable evidence.

House of Quality: A matrix commonly used in QFD to relate customer requirements to technical characteristics.

Design for manufacturability: Designing a product so it can be produced consistently and economically with available processes.

Mistake-proofing: Designing a product or process to prevent errors or make them immediately evident.

FMEA: A systematic method for identifying potential failures, their causes and effects, and risk-reduction actions.

Risk priority number: A traditional FMEA score calculated by multiplying severity, occurrence, and detection ratings.

Reliability: The probability that an item performs its required function under stated conditions for a specified time.

Maintainability: The ease and speed with which an item can be restored to a functioning condition.

Availability: The proportion of time an item is capable of performing its required function, considering failures and restoration.

Robust design: Design that maintains performance despite uncontrollable environmental or usage variation.

Design of experiments: A planned statistical approach for studying factor effects and interactions on a response.

Verification: Evaluation of whether specified design requirements have been met.

Validation: Evaluation of whether a product or service is suitable for its intended use and users.

Service blueprint: A representation of customer actions, employee actions, support processes, systems, and service interfaces.

Service recovery: Actions taken to correct a service failure, restore customer value, and learn from the event.

Configuration control: Formal management of approved product, process, software, and document versions.

Reliability growth: Improvement in reliability through iterative testing, failure analysis, corrective action, and retesting.

Review Questions

Short Answer Questions

1. Why is inspection unable to compensate fully for poor quality of design?

2. Distinguish between verification and validation.

3. What is the purpose of QFD?

4. Define reliability and explain how it differs from availability.

5. What is meant by prevention at source?

Long Answer Questions

1. Explain the stages of design thinking and discuss how they contribute to quality-oriented innovation.

2. Describe the structure and uses of an FMEA. What are the limitations of relying exclusively on RPN?

3. Discuss how robust design and design of experiments can reduce variation and improve performance.

4. Explain design for manufacturability, serviceability, accessibility, and safety as components of design for quality.

5. Describe the role of verification, validation, configuration control, and change management in responsible product or service release.

Analytical / Application Questions

1. A company receives repeated complaints that a new appliance is technically compliant but difficult to install and maintain. Design an integrated QFD, FMEA, and design-for-serviceability approach to address the problem.

2. A digital banking service has occasional payment failures, but the failures can create serious consequences for a small number of customers. Explain how severity, occurrence, and detection should be considered, and propose validation and recovery actions.

3. A manufacturer wants to improve product performance by changing three process factors, each at two levels. Explain how a factorial DOE could be used, what evidence should be collected, and how practical significance would be assessed.

22. Multiple Choice Questions with Answers

Question 1

Which statement best captures the distinction between verification and validation?

A. Verification examines user satisfaction, while validation examines production cost.

B. Verification asks whether specified requirements were met, while validation asks whether the offering is suitable for intended use.

C. Verification occurs after launch, while validation occurs only during concept design.

D. Verification identifies failures, while validation eliminates all process variation.

Correct Answer: B. Verification asks whether specified requirements were met, while validation asks whether the offering is suitable for intended use.

Question 2

A QFD team discovers that improving energy efficiency may increase product weight. What is the most appropriate use of QFD in this situation?

A. Ignore the conflict because customer requirements cannot be compared.

B. Make the trade-off visible and evaluate it against prioritized customer and technical requirements.

C. Select the lowest-cost option without further analysis.

D. Replace customer requirements with engineering preferences.

Correct Answer: B. Make the trade-off visible and evaluate it against prioritized customer and technical requirements.

Question 3

Why is a high-severity failure not necessarily safe to ignore when its traditional RPN is low?

A. RPN is based only on customer satisfaction.

B. High severity can make a failure important even when occurrence or detection scores are favourable.

C. Severity is excluded from the RPN calculation.

D. Low RPN always indicates a measurement-system problem.

Correct Answer: B. High severity can make a failure important even when occurrence or detection scores are favourable.

Question 4

Which action is generally the strongest form of risk control in FMEA?

A. Adding a final inspection step

B. Training inspectors to work faster

C. Redesigning the product or process to eliminate the failure cause

D. Increasing the number of reports generated

Correct Answer: C. Redesigning the product or process to eliminate the failure cause

Question 5

A design performs well in the laboratory but poorly when exposed to temperature and user variation. Which approach is most directly relevant?

A. Robust design

B. Final inspection only

C. Complaint closure without analysis

D. Reducing the number of customer requirements

Correct Answer: A. Robust design

Question 6

What is a principal advantage of factorial design over one-factor-at-a-time experimentation?

A. It eliminates the need for a defined response variable.

B. It can reveal interactions among factors while studying several factors efficiently.

C. It guarantees that every result is practically significant.

D. It requires no control of experimental conditions.

Correct Answer: B. It can reveal interactions among factors while studying several factors efficiently.

Question 7

In the hospital appointment case, why were high-severity failures prioritized even when their estimated occurrence was moderate?

A. Because all moderate-occurrence failures automatically have the highest RPN.

B. Because severe consequences require explicit attention beyond mechanical numerical ranking.

C. Because validation is unnecessary for high-severity failures.

D. Because customer interviews cannot identify serious risks.

Correct Answer: B. Because severe consequences require explicit attention beyond mechanical numerical ranking.

Question 8

Which activity is an example of service recovery rather than ordinary service delivery?

A. Designing a new appointment journey before launch

B. Acknowledging a failed booking, correcting it, and investigating its cause

C. Setting a standard appointment capacity

D. Mapping backstage support processes before implementation

Correct Answer: B. Acknowledging a failed booking, correcting it, and investigating its cause

Question 9

Why should AI-based quality predictions not be accepted without human review?

A. AI cannot process any operational data.

B. AI outputs may reflect biased data, model drift, weak explanations, or incorrect assumptions.

C. Human review always produces less reliable decisions.

D. Digital systems cannot be used in quality management.

Correct Answer: B. AI outputs may reflect biased data, model drift, weak explanations, or incorrect assumptions.

Question 10

Which practice best supports lifecycle quality after a product or software change?

A. Assuming previous test evidence remains valid in all circumstances

B. Performing impact assessment and repeating appropriate verification or validation under configuration control

C. Deleting earlier requirements to avoid comparison

D. Relying exclusively on customer complaints after release

Correct Answer: B. Performing impact assessment and repeating appropriate verification or validation under configuration control