8. Quality Planning, Product Design and Reliability

Introduction and Context

Quality is most effectively created before production begins. Inspection can detect some defects, but it cannot compensate for a product whose requirements are unclear, whose design is inherently fragile, or whose process is incapable of meeting specifications. Quality planning therefore connects customer expectations, product architecture, manufacturing processes, supplier capability, measurement systems, and field performance.

In production management, this connection is especially important because product design decisions determine a large proportion of later cost, manufacturability, reliability, serviceability, and customer satisfaction. A design that is difficult to assemble may generate recurring process defects. A material selected without adequate environmental analysis may produce premature failures. Similarly, a process that appears acceptable during pilot production may not remain stable under volume, shift, supplier, or seasonal variation.

This chapter examines the integrated methods used to prevent such problems. It focuses on quality planning, design reviews, Quality Function Deployment (QFD), Design Failure Mode and Effects Analysis (DFMEA), Process FMEA (PFMEA), control plans, Advanced Product Quality Planning (APQP), Production Part Approval Process (PPAP), and reliability engineering. These methods are considered not as isolated forms, but as a linked management system for designing products and processes that are capable, robust, maintainable, and economically sustainable.

Learning Objectives

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

1. Explain how quality planning integrates customer, design, process, supplier, and service requirements.

2. Apply QFD to translate the voice of the customer into measurable technical characteristics.

3. Distinguish DFMEA from PFMEA and use both to prioritize preventive action.

4. Explain the relationships among APQP, control plans, process capability, and PPAP.

5. Assess reliability, availability, maintainability, failure rate, and life-cycle performance.

6. Select appropriate reliability testing and accelerated testing approaches.

7. Evaluate design-for-quality and design-for-manufacturability decisions.

8. Identify managerial, ethical, regulatory, and digital challenges in quality planning.

Quality Planning as an Integrating Discipline

Quality planning is the structured activity of deciding what quality must be achieved, how it will be designed into the product, how it will be produced consistently, and how evidence of conformity will be generated. It begins with customer and stakeholder requirements and extends through design, industrialization, production, delivery, use, maintenance, and disposal.

A useful quality plan answers five questions:

• What does the customer or regulator require?
• Which product characteristics are critical to function, safety, compliance, and satisfaction?
• Which process conditions control those characteristics?
• How will risks be prevented, detected, and responded to?
• What objective evidence demonstrates that the product and process are capable?

Quality planning is consequently cross-functional. Marketing may understand customer needs, engineering controls product characteristics, manufacturing engineering designs the process, procurement manages suppliers, quality engineering evaluates risk and capability, maintenance protects equipment performance, and service personnel provide field information. A plan that excludes any of these perspectives is likely to contain blind spots.

The central principle is prevention. Prevention is generally more effective than detection because defects become progressively more expensive as they move from design to production, distribution, installation, and customer use. Prevention does not mean eliminating all inspection. Rather, inspection is positioned intelligently within a system that emphasizes robust design, capable processes, mistake-proofing, and rapid feedback.

From Customer Need to Engineering Requirement: QFD

Quality Function Deployment is a method for systematically translating the voice of the customer into product and process requirements. Its best-known representation is the House of Quality, although QFD is better understood as a sequence of linked deployment matrices rather than a single form.

The first stage identifies customer requirements, often called the “whats.” These may include safety, low noise, ease of use, appearance, durability, response time, or serviceability. Because customers frequently express needs in informal or ambiguous language, the team must clarify the intended meaning without replacing the customer’s perspective with engineering assumptions.

The second stage identifies technical characteristics, or the “hows,” that can be measured and controlled. For example, “quiet operation” may be translated into sound pressure level at defined operating conditions; “long life” may become a specified number of cycles or operating hours; “easy to assemble” may become maximum assembly time or allowable insertion force.

The relationship matrix indicates how strongly each technical characteristic affects each customer requirement. A roof matrix may then identify positive or negative correlations among technical characteristics. For example, increasing structural stiffness may improve durability but increase weight, cost, or energy consumption. Such conflicts are valuable because they expose trade-offs early, when alternatives are still available.

QFD also supports competitive assessment, target setting, and prioritization. A high-priority customer requirement should not automatically produce an extreme engineering target. Targets must be tested against feasibility, cost, safety, regulatory requirements, and process capability. QFD is therefore a decision-support method, not a mechanical scoring exercise.

Design Reviews and Design for Quality

A design review is a formal, evidence-based examination of a design at a planned stage. It should include people who can challenge assumptions and identify consequences outside the originating function. Typical review questions include:

• Are all customer, statutory, regulatory, and interface requirements defined?
• Are critical characteristics identified and measurable?
• Can the product be manufactured, inspected, transported, installed, maintained, and safely disposed of?
• Have foreseeable misuse, environmental conditions, and variation been considered?
• Are test methods representative of actual use?
• Have unresolved risks been assigned owners and due dates?

Design for Quality means selecting product concepts, materials, tolerances, interfaces, and verification methods that reduce the likelihood of failure. It includes design for manufacturability and assembly (DFMA), design for serviceability, design for safety, design for environment, and design for robustness.

Tolerance design illustrates the importance of this approach. A nominal dimension may meet its target while the combined variation of several dimensions causes interference, leakage, or poor fit. Designers therefore need tolerance stack-up analysis, statistical variation analysis where appropriate, and clear identification of special characteristics. Overly tight tolerances can make production unnecessarily expensive; overly loose tolerances can impair function. The objective is not maximum precision but adequate, economically achievable precision.

Robust design seeks performance that is relatively insensitive to noise factors such as temperature, vibration, operator technique, material variation, and power fluctuations. Design experiments, simulation, material selection, and parameter optimization can be used to find settings that reduce sensitivity. Robustness is particularly valuable because not all variation can be eliminated economically.

DFMEA and PFMEA: Anticipating Failure

Failure Mode and Effects Analysis is a structured method for identifying potential failures, understanding their effects and causes, and prioritizing risk-reduction action. It is most useful when performed early and revised as knowledge develops.

DFMEA examines the product design. A failure mode is the way in which the design might fail, such as fracture, leakage, loss of electrical continuity, excessive wear, or incorrect software response. The effect describes what the user, next assembly, system, or regulator experiences. The cause identifies the mechanism, such as inadequate strength, corrosion, poor heat dissipation, or an incompatible interface. Existing prevention and detection controls are then recorded.

PFMEA examines how the manufacturing or service process might fail. Examples include incorrect component orientation, inadequate torque, contamination, incorrect software loading, missed inspection, or damage during handling. The analysis considers the effect on the product and customer, the process cause, prevention controls, and detection controls.

Traditional FMEA uses rankings for severity, occurrence, and detection. These rankings are useful for discussion, but their multiplication into a Risk Priority Number should not be treated as an absolute measure of risk. A low-occurrence failure with catastrophic severity may deserve immediate attention even if its numerical score is modest. Contemporary practice therefore emphasizes severity, action priority, evidence-based occurrence estimates, and the effectiveness of preventive controls.

An effective FMEA is not a document completed by one person. It is a team analysis involving design, production, quality, maintenance, suppliers, testing, and service where relevant. It should lead to action: redesign, error-proofing, stronger process controls, improved test methods, supplier development, or revised maintenance. A completed form with no implemented action is not risk management.

APQP, Control Plans and PPAP

Advanced Product Quality Planning is a disciplined framework for coordinating product and process development. Although associated strongly with the automotive sector, its logic is applicable to many industries. APQP commonly progresses through five broad phases:

1. Planning and definition of the programme.

2. Product design and development.

3. Process design and development.

4. Product and process validation.

5. Feedback, assessment, and corrective action.

The phases overlap. Customer requirements, feasibility analysis, preliminary process assumptions, design risk, supplier readiness, testing, and launch timing must be managed together rather than as independent departmental activities.

The control plan converts risk analysis and process knowledge into an operational control system. It identifies each process step, the product or process characteristic to be controlled, the specification or target, the measurement method, sample size and frequency, responsible personnel, and reaction plan. It should distinguish prevention controls from detection controls. For example, a poka-yoke device preventing incorrect orientation is a preventive control, while final visual inspection is a detection control.

A reaction plan must be specific. It should state what happens when a result is out of control or a specification is violated: stop the process when appropriate, segregate suspect product, identify the last known acceptable condition, investigate the cause, correct the process, and verify disposition. “Inform supervisor” alone is not an adequate reaction plan.

PPAP is an approval and evidence process used to demonstrate that the supplier understands the customer’s requirements and that the production process can produce conforming parts at the intended rate under actual production conditions. Typical evidence may include design records, approved specifications, process flow diagrams, DFMEA and PFMEA, control plans, measurement system studies, dimensional results, material or performance test results, initial process studies, sample parts, and a formal submission warrant.

PPAP should not be confused with a guarantee of lifetime quality. It provides evidence of readiness at a defined point and under defined conditions. Changes in material, tooling, location, supplier, software, process parameters, or inspection methods may require reassessment and customer approval.

Process Capability and Measurement Confidence

A control plan is meaningful only if the measurement system and process are capable. Measurement system analysis examines repeatability, reproducibility, bias, stability, linearity, and resolution. If a gauge cannot distinguish acceptable from unacceptable product reliably, process decisions based on it are questionable.

Process capability compares process variation and centering with specification limits. Capability indices such as Cp and Cpk are useful when the process is stable and the data assumptions are appropriate. Cp reflects potential capability based on spread, while Cpk also considers displacement from the target or specification center. Capability indices do not prove that a process is stable; control charts and process knowledge are needed first.

Managers should avoid treating a capability index as a universal pass-or-fail truth. Non-normal data, autocorrelation, mixed populations, short runs, measurement error, and unrepresentative samples can produce misleading results. Capability is a property of a process under stated conditions, not a permanent attribute of a machine.

Reliability, Availability and Maintainability

Reliability is the probability that an item performs its required function for a specified time under stated conditions. The definition contains four important elements: function, time, operating conditions, and probability. A product can be reliable in one environment and unreliable in another.

For a constant failure rate, the reliability function is often represented as:

R(t) = e^(-λt)

where λ is the failure rate and t is operating time. The mean time to failure for an appropriate non-repairable exponential model is approximately 1/λ. Real products may not have a constant failure rate. The familiar bathtub curve describes an early-life period of decreasing failure rate, a useful-life period with approximately random failures, and a wear-out period with increasing failure rate. The model is a conceptual aid, not a universal description of every product.

Availability concerns whether an item is ready for use when required. In a simplified steady-state form:

Availability = MTBF / (MTBF + MTTR)

where MTBF is mean time between failures and MTTR is mean time to repair. High reliability does not automatically produce high availability if repairs are slow, spare parts are unavailable, diagnostics are poor, or maintenance access is difficult.

Maintainability is the probability that a failed item can be restored to a specified condition within a stated time using defined procedures and resources. Design features that improve maintainability include modular replacement, accessible fasteners, diagnostic codes, standard tools, clear service instructions, and safe isolation points. Reliability, availability, and maintainability should be considered together because customers generally purchase useful service, not merely survival without failure.

Life-cycle performance includes acquisition, operation, maintenance, energy, downtime, training, consumables, disposal, and environmental costs. A low purchase price may be economically inferior if it produces high operating cost or frequent interruptions. Life-cycle costing supports more rational design and procurement decisions.

Reliability Testing and Accelerated Testing

Reliability testing may be conducted during development, before launch, or during production monitoring. Common approaches include life testing, environmental testing, vibration and shock testing, thermal cycling, humidity testing, endurance testing, software stress testing, and field-return analysis. Test conditions must represent credible use and credible failure mechanisms.

Accelerated testing increases stress, temperature, duty cycle, or other factors to obtain information in less time. It is valid only when the acceleration mechanism is understood and the failure mode remains relevant. For temperature-driven processes, an Arrhenius-type relationship may be appropriate; for mechanical fatigue, stress-life or strain-life models may be more suitable. Simply increasing every stress may create unrealistic failures and lead to incorrect conclusions.

Test planning should define objectives, sample size, censoring rules, confidence requirements, failure criteria, environmental conditions, instrumentation, and data analysis methods. Zero failures in a small test is not proof of infinite reliability. Statistical confidence depends on sample size, exposure time, assumptions, and the reliability demonstration objective.

Reliability growth occurs when design or process changes remove failure causes discovered during testing. A disciplined failure reporting and corrective-action system connects test results to engineering changes, verification, and production controls.

An Illustrative Case: Improving Reliability in an Industrial Pump

Background

The following fictional case concerns Apex Fluid Systems, a manufacturer of electrically driven industrial pumps. The company was preparing a new pump for chemical-processing customers. The customer required high availability, resistance to corrosive fluids, easy seal replacement, and documented process capability.

Situation and Problem

Prototype testing showed acceptable flow performance, but several units developed seal leakage after extended thermal cycling. Assembly records also showed variation in shaft-seal compression. The launch schedule encouraged the team to add final leakage inspection rather than reconsider the design and process.

Analysis

A cross-functional QFD review confirmed that “no leakage,” “long service interval,” and “quick maintenance” were high-priority customer requirements. The DFMEA identified seal degradation, incorrect compression, shaft surface damage, and thermal expansion mismatch as potential causes. The PFMEA identified incorrect seal orientation, uncontrolled press force, contamination, and inadequate verification of compression.

A tolerance stack-up showed that the combined variation of the shaft, housing, seal, and retaining feature could produce excessive compression in some units and insufficient compression in others. The measurement system study also found that the existing leakage test was unstable because fixture temperature was not controlled.

Approach and Solution

The design team changed the retaining geometry, specified a more compatible seal material, and introduced a controlled surface-finish requirement. Manufacturing engineering developed a guided assembly fixture with orientation poka-yoke and force-displacement monitoring. The control plan added verification of critical dimensions, environmental conditioning before leakage testing, and a defined reaction plan for abnormal results.

The team revised the DFMEA and PFMEA, conducted design and process validation, and repeated reliability testing using representative chemical exposure and thermal cycling. Supplier approval activities were updated, and PPAP evidence included capability studies for the seal housing diameter and assembly force.

Outcome

The revised design completed the planned validation without the earlier leakage mode. More importantly, the organisation reduced dependence on final inspection by controlling the causes of variation. Maintenance personnel reported easier seal replacement because the redesigned access features reduced disassembly steps.

Learning Points

The case demonstrates that reliability problems often originate in interactions among design, materials, tolerances, process variation, measurement conditions, and maintenance requirements. Adding inspection at the end would not have addressed these causes. QFD, FMEA, design review, capability analysis, control planning, and validation were effective because they were used as a connected system.

Challenges, Risks and Common Mistakes

Several weaknesses repeatedly reduce the value of quality planning. Teams may begin FMEA after the design is effectively frozen, copy old documents without reviewing changed risks, or assign rankings without evidence. QFD may become a mathematical exercise in which relationship scores are debated more than customer needs. Control plans may list inspections but omit prevention, reaction criteria, or ownership. PPAP packages may be assembled retrospectively, obscuring unresolved launch risks.

Other risks include supplier process changes made without notification, inadequate traceability, unrepresentative reliability tests, overreliance on capability indices, and failure to learn from warranty or field-return data. Excessive documentation can also become counterproductive if employees cannot understand or use it. The objective is controlled knowledge and effective action, not paperwork volume.

Ethically, engineers and managers must report test failures honestly, avoid suppressing unfavourable data to meet launch targets, and protect users from known safety risks. Regulatory and contractual requirements may govern traceability, material declarations, functional safety, cybersecurity, and change approval. Product claims concerning life, safety, or reliability should be supported by appropriate evidence and clearly stated assumptions.

Digital Quality Planning and Emerging Practice

Digital engineering tools increasingly connect requirements management, computer-aided design, simulation, FMEA, process planning, manufacturing execution systems, and field data. A digital thread can preserve the relationship between a customer requirement, a critical product characteristic, a process parameter, an inspection result, and a field failure.

Analytics can identify patterns in warranty data, detect drift in process parameters, and support predictive maintenance. Machine learning may assist anomaly detection or failure prediction, but it does not eliminate the need for engineering judgement. Models can be affected by biased data, changing operating conditions, undocumented design changes, and false alarms. Data governance, cybersecurity, access control, model validation, and human accountability are therefore essential.

Emerging practice also emphasizes sustainability and circularity. Designers increasingly consider energy use, repairability, recycled content, disassembly, remanufacturing, and end-of-life recovery. These objectives can conflict with weight, cost, durability, or recyclability, making early cross-functional planning more important rather than less.

Best Practices for Professional Application

Effective organisations generally follow these principles:

• involve customers, suppliers, operations, service, and quality early;
• translate qualitative needs into measurable and verifiable requirements;
• review designs for function, variation, safety, manufacture, assembly, service, and environment;
• use DFMEA and PFMEA to drive preventive action rather than merely document risk;
• validate measurement systems before interpreting process data;
• confirm process stability before using capability indices;
• design control plans with explicit prevention controls and reaction plans;
• test products under representative stresses and analyse failures systematically;
• maintain configuration control and reassess changes;
• connect warranty, maintenance, and field data to design and process learning;
• treat reliability, maintainability, availability, cost, and sustainability as linked life-cycle decisions.

Chapter Summary

Quality planning integrates customer requirements, product design, process development, supplier readiness, production control, validation, and field learning. QFD translates the voice of the customer into measurable technical targets and exposes design trade-offs. Design reviews and design-for-quality methods address manufacturability, tolerance, robustness, safety, serviceability, and life-cycle performance.

DFMEA anticipates product-design failures, while PFMEA examines process failures. APQP coordinates development activities; the control plan converts identified risks into operational controls; and PPAP provides evidence of production readiness. Measurement system analysis and process capability studies are essential for interpreting control data correctly.

Reliability concerns successful function over time and under stated conditions. Availability additionally depends on repair time and support resources, while maintainability concerns the ease and speed of restoration. Reliability testing, accelerated testing, field-data analysis, and reliability growth help establish and improve life-cycle performance. Digital tools can strengthen traceability and analytics, but they require sound data, governance, and engineering judgement. The central lesson is that quality and reliability are designed into systems through coordinated decisions, not inspected into finished products.

Key Terms

Quality planning: Structured planning to achieve specified product, process, and service quality.

Voice of the customer: Explicit and implicit customer needs, expectations, and preferences.

Quality Function Deployment: A method for translating customer requirements into technical and process requirements.

House of Quality: A matrix-based representation commonly used in the first stage of QFD.

Design review: A formal evaluation of design adequacy, risks, interfaces, and readiness.

DFMA: Design for manufacture and assembly; designing products for efficient, consistent production and assembly.

DFMEA: Analysis of potential failures arising from product design decisions.

PFMEA: Analysis of potential failures arising from manufacturing or service processes.

Critical characteristic: A product or process characteristic requiring special control because of its effect on function, safety, regulation, or customer requirements.

APQP: Advanced Product Quality Planning, a structured framework for coordinated product and process development.

Control plan: A document specifying process controls, measurements, frequencies, responsibilities, and reaction plans.

PPAP: Production Part Approval Process, which provides evidence that production requirements are understood and achievable.

Poka-yoke: A mistake-proofing method that prevents or immediately detects incorrect action.

Process capability: The ability of a stable process to meet specified requirements.

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

Availability: Probability that an item is ready for use when required.

Maintainability: Probability that an item can be restored within a specified time and conditions.

MTBF: Mean time between failures for a repairable item.

MTTR: Mean time to repair or restore an item.

Accelerated testing: Testing under increased but technically justified stress to obtain reliability information more quickly.

Review Questions

Short Answer Questions

1. Why is prevention generally more effective than final inspection in quality management?

2. What is the difference between a customer requirement and a technical characteristic in QFD?

3. Distinguish DFMEA from PFMEA.

4. What information should a control plan contain?

5. Define reliability, availability, and maintainability.

Long Answer Questions

1. Explain how QFD, design reviews, and tolerance analysis support design for quality.

2. Describe the relationship among APQP, FMEA, control plans, process capability, and PPAP.

3. Explain the reliability bathtub curve and discuss its usefulness and limitations.

4. Discuss the design and managerial factors that influence availability and maintainability.

5. Explain the conditions under which accelerated reliability testing produces valid conclusions.

Analytical / Application Questions

1. A process has a high Cpk, but customer complaints remain frequent. Identify possible reasons and propose an investigation plan.

2. A product has few failures during use but requires long repair times and frequently lacks spare parts. Analyse its reliability and availability performance and recommend improvements.

3. Develop a preliminary QFD-to-control-plan logic for a product whose major customer needs are safety, low operating noise, long life, and easy maintenance.

22. Multiple Choice Questions with Answers

Question 1

Which activity most directly translates customer statements such as “quiet operation” into measurable engineering targets?

A. Production Part Approval Process

B. Quality Function Deployment

C. Process capability analysis

D. Maintenance scheduling

Correct Answer: B. Quality Function Deployment

Question 2

A failure involving incorrect component orientation during assembly would primarily be examined in which analysis?

A. DFMEA

B. PFMEA

C. QFD competitive assessment

D. Life-cycle costing

Correct Answer: B. PFMEA

Question 3

What is the main purpose of a control-plan reaction plan?

A. To define the product’s marketing position

B. To calculate mean time between failures

C. To specify actions when a process or result becomes unacceptable

D. To replace the need for process capability studies

Correct Answer: C. To specify actions when a process or result becomes unacceptable

Question 4

Which statement best distinguishes availability from reliability?

A. Availability considers readiness for use and can be reduced by long repair times

B. Availability applies only to non-repairable products

C. Reliability measures repair speed, whereas availability measures design cost

D. Reliability and availability are identical for all production systems

Correct Answer: A. Availability considers readiness for use and can be reduced by long repair times

Question 5

Why should a capability index not be interpreted before process stability is assessed?

A. Capability indices apply only to service processes

B. An unstable process may produce a misleading estimate of its long-term performance

C. Stable processes cannot have specification limits

D. Capability indices measure customer satisfaction directly

Correct Answer: B. An unstable process may produce a misleading estimate of its long-term performance

Question 6

A team records a low numerical FMEA risk score for a possible failure that could cause a catastrophic safety consequence. What is the most appropriate response?

A. Ignore it because the calculated score is low

B. Prioritize it based only on occurrence

C. Evaluate its severity and implement suitable risk reduction despite the composite score

D. Remove it from the FMEA until a failure occurs

Correct Answer: C. Evaluate its severity and implement suitable risk reduction despite the composite score

Question 7

Which situation is the strongest example of a valid preventive control?

A. Final visual inspection of every finished unit

B. A fixture that prevents a component from being installed in the wrong orientation

C. Sorting returned products after customer complaints

D. Reviewing warranty data once a year

Correct Answer: B. A fixture that prevents a component from being installed in the wrong orientation

Question 8

What is a major risk in accelerated reliability testing?

A. The test may create unrealistic failure mechanisms that do not represent field use

B. The test always requires a larger sample than normal-life testing

C. Acceleration eliminates the need for statistical analysis

D. Accelerated testing can measure only appearance defects

Correct Answer: A. The test may create unrealistic failure mechanisms that do not represent field use

Question 9

In the illustrative pump case, why was the leakage test itself investigated?

A. The customer had removed leakage requirements

B. The measurement conditions were unstable and could obscure true process performance

C. Reliability testing cannot be used for pumps

D. The design team wanted to eliminate all inspections

Correct Answer: B. The measurement conditions were unstable and could obscure true process performance

Question 10

Which change most directly improves maintainability?

A. Increasing product complexity without adding diagnostics

B. Making service components accessible and providing clear isolation and replacement procedures

C. Reducing all preventive maintenance activities

D. Using tighter tolerances on every component

Correct Answer: B. Making service components accessible and providing clear isolation and replacement procedures