Quality Planning, Customer Requirements and Economics

Introduction

Quality in mechanical engineering is not created by inspection alone. It is designed into products and processes through a disciplined understanding of customer needs, technical requirements, risks, process capability, and economic consequences. A component may conform to its drawing and still fail to satisfy the customer if it is difficult to assemble, unreliable in service, delivered late, or unnecessarily expensive. Conversely, a product may contain excellent features that customers neither need nor value.

Quality planning provides the systematic link between customer expectations and manufacturing activities. It determines what must be achieved, how achievement will be measured, where risks may arise, and which controls are economically justified. Quality Function Deployment (QFD) is one important method for translating the voice of the customer into engineering characteristics and process controls. Quality assurance establishes confidence that quality requirements will be met, whereas quality control provides operational techniques for detecting and controlling variation.

The economic dimension is equally important. Resources devoted to prevention and appraisal can reduce internal and external failure costs, but excessive inspection may add cost without improving the product. Effective quality management therefore seeks the lowest total cost consistent with customer, statutory, safety, and performance requirements.

Learning objectives

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

1. Explain the purpose and stages of quality planning.
2. Translate customer needs into measurable product and process specifications.
3. Construct and interpret the main elements of a QFD House of Quality.
4. Distinguish quality assurance, quality control, inspection, and improvement.
5. Classify prevention, appraisal, internal-failure, and external-failure costs.
6. Evaluate risk using severity, occurrence, and detection considerations.
7. Analyse prevention–appraisal–failure trade-offs and make economically justified decisions.
8. Apply these concepts to a mechanical-engineering product or process.

1. Quality planning as a systems activity

Quality planning is the process of identifying quality objectives and establishing the processes, resources, responsibilities, measurements, and acceptance criteria required to achieve them. It is broader than preparing an inspection plan. A quality plan should answer five questions:

• What does the customer or other interested party require?
• Which product and process characteristics determine satisfaction and safety?
• How will those characteristics be measured or verified?
• Who is responsible for achieving and confirming them?
• What action is required if results do not conform?

Quality planning begins before detailed design and continues through production, delivery, installation, use, maintenance, and disposal. The life-cycle perspective is essential because many quality problems originate in early decisions. For example, a poorly specified tolerance may cause expensive machining, while an inaccessible lubrication point may create field failures even if the manufactured dimensions are correct.

A useful planning sequence is:

1. Identify customers, users, regulators, maintainers, and downstream operations.
2. Capture explicit and implicit requirements.
3. Convert requirements into measurable characteristics.
4. Assess technical feasibility, risk, and regulatory obligations.
5. Design the product and process to meet the characteristics.
6. Establish verification, validation, and control methods.
7. Monitor results and improve the system using feedback.

Verification asks whether specified requirements have been met; validation asks whether the resulting product is suitable for its intended use. A shaft can be verified as having the specified diameter, yet validation may reveal that its fatigue life is inadequate for the actual loading spectrum.

Requirements hierarchy

Customer requirements normally exist at several levels:

A requirement is useful only when its meaning, measurement method, acceptance limit, responsible party, and applicable conditions are clear. Terms such as “high quality,” “smooth,” or “strong” are incomplete until translated into measurable technical language.

2. Voice of the customer and specification development

The voice of the customer (VOC) is the organized representation of customer expectations. It may be obtained through interviews, field observation, warranty data, service records, complaint analysis, competitor comparison, surveys, and direct participation by users in design reviews. Engineers must distinguish what customers say from why they say it. A customer may request a “thicker housing,” while the underlying need is lower vibration or greater impact resistance. Designing directly to the stated solution can prevent better alternatives from being considered.

From need to measurable specification

A robust translation process follows this pattern:

Customer statement → interpreted need → measurable characteristic → target value → tolerance and test method.

For example:

• Statement: “The coupling must be easy to install.”
• Need: Low installation effort and low risk of incorrect assembly.
• Characteristics: Maximum insertion force, assembly time, keying error-proofing, and required tool access.
• Targets: Insertion force ≤ 120 N; assembly time ≤ 3 min; impossible to assemble in the wrong orientation.
• Evidence: Force measurement, timed assembly trials, and mistake-proofing verification.

A specification should include:

• characteristic name and definition;
• nominal or target value;
• allowable limits or tolerance;
• units and reference conditions;
• measurement method and equipment;
• sampling or test conditions;
• traceability requirements; and
• action required for nonconformance.

Tolerances and economic consequences

A tolerance defines the permissible variation around a nominal value. For a bilateral dimensional tolerance, the specification may be written as:

where x is the measured characteristic, x_L is the lower specification limit, and x_U is the upper specification limit. The nominal value is often near the midpoint, but this is not mandatory.

A narrower tolerance can improve function but usually increases manufacturing cost, inspection effort, or scrap. A wider tolerance can reduce cost but may increase assembly variation or field failure. Tolerance decisions should therefore be based on functional analysis, process capability, risk, and total cost rather than habit.

For a normally distributed characteristic with process mean μ and standard deviation σ, capability indices are commonly used when the process is stable and approximately normal:

Here, USL and LSL are the upper and lower specification limits. C_p measures potential capability assuming the process is centred; C_pk also accounts for off-centre operation. These indices are meaningful only when the process is statistically stable, the measurement system is adequate, and the data represent the relevant operating conditions. They are not direct guarantees of conformance or product performance.

3. Quality Function Deployment

Quality Function Deployment is a structured approach for deploying customer requirements through product design, part characteristics, process planning, and production control. Its best-known tool is the House of Quality, a matrix that relates customer requirements, engineering characteristics, competitive priorities, and technical relationships.

QFD does not replace engineering judgement. It makes assumptions and priorities visible, encourages cross-functional communication, and reduces the risk that important customer needs will be overlooked.

Main elements of the House of Quality

1. Customer requirements (“WHATs”): Statements of what the customer wants, expressed in customer language.
2. Importance ratings: Relative importance assigned by customers or a cross-functional team.
3. Engineering characteristics (“HOWs”): Measurable design variables that influence the requirements.
4. Relationship matrix: Strength of the relationship between each WHAT and HOW.
5. Technical correlation roof: Positive or negative interactions among engineering characteristics.
6. Competitive assessment: Comparison of the proposed product with alternatives.
7. Technical priorities and targets: Calculated or judged priorities used to guide design decisions.

A simple weighted priority for engineering characteristic j is:

where P_j is the priority of characteristic j, I_i is the importance of customer requirement i, R_ij is the relationship weight between requirement i and characteristic j, and m is the number of customer requirements. A common qualitative scale is 0 for no relationship, 1 for weak, 3 for medium, and 9 for strong. The scale is not a law of nature; it is a team convention and should be documented.

If relative priorities are desired, they may be normalized:

where W_j is the percentage priority and n is the number of engineering characteristics. The calculation assumes that the importance ratings and relationship weights are sufficiently comparable for weighted addition. It should not be interpreted as a precise measurement of customer utility.

Illustrative QFD calculation

Suppose customers rate three requirements for a lifting-device handle:

The engineering characteristics are grip diameter, required operating force, and corrosion resistance. Using relationship weights 0, 1, 3, and 9:

The priorities are:

• Grip diameter: P₁ = 5×9 + 4×1 + 3×0 = 49
• Operating force: P₂ = 5×3 + 4×9 + 3×1 = 54
• Corrosion resistance: P₃ = 5×0 + 4×0 + 3×9 = 27

The total is 130, so the relative priorities are approximately 37.7%, 41.5%, and 20.8%, respectively. Operating force receives the greatest priority, although grip diameter remains important. The team must then set realistic targets, such as a maximum operating force, a grip diameter range based on ergonomic evidence, and a corrosion-resistance test requirement.

Strengths and limitations of QFD

QFD encourages customer-centred design, exposes conflicts, supports traceability, and provides a defensible basis for prioritization. It is especially useful when multiple departments must agree on requirements.

Limitations include subjective importance ratings, dependence on the quality of customer research, possible double-counting of related requirements, and the danger of treating numerical priorities as objective truth. A large matrix can become bureaucratic. QFD should be updated when requirements, technology, regulations, or field evidence change.

4. Quality assurance, quality control, and improvement

These terms are related but not interchangeable.

Quality assurance (QA) comprises planned and systematic activities that provide confidence that quality requirements will be fulfilled. Examples include documented procedures, design reviews, supplier qualification, audits, training, configuration management, validation planning, and corrective-action systems.

Quality control (QC) comprises operational techniques and activities used to fulfil quality requirements. Examples include measurement, inspection, testing, process monitoring, sampling, calibration, and disposition of nonconforming output.

Inspection is one QC activity. It compares a characteristic with specified requirements, but inspection alone does not create quality and may detect problems only after value has been added.

Quality improvement is the organized effort to increase process effectiveness, reduce variation, remove root causes, or improve customer outcomes. Tools include the Plan–Do–Study–Act cycle, statistical process control, designed experiments, root-cause analysis, and mistake-proofing.

A useful distinction is:

A mature system uses QC data to improve QA and processes. Repeated dimensional nonconformance should not be managed indefinitely by sorting; the process cause should be investigated.

5. Risk-based quality planning

Risk is the effect of uncertainty on objectives. In product quality, risks may involve safety, performance, reliability, delivery, compliance, environmental impact, or cost. Risk-based planning directs stronger controls toward failures with more serious consequences.

Failure Mode and Effects Analysis (FMEA) is a structured method for identifying failure modes, their effects, causes, existing controls, and recommended actions. Traditional FMEA ratings often use scales for severity S, occurrence O, and detection D. A risk priority number may be calculated as:

The product is dimensionless if the three ratings are ordinal scores. It should not be treated as a probability or a physically meaningful risk unit. Two failure modes can have the same RPN while having very different safety implications. High-severity hazards may require action even when occurrence is low, and a low detection score does not make a severe failure acceptable.

Modern risk management therefore considers the individual rating profiles, special characteristics, legal requirements, and whether controls prevent causes or merely detect effects. For safety-critical applications, formal hazard analysis and regulatory requirements take precedence over a simple RPN ranking.

Risk controls generally follow a hierarchy:

1. Eliminate the hazard through design.
2. Reduce the likelihood or consequence through inherently safer design.
3. Add protective devices or interlocks.
4. Establish process controls and inspection.
5. Provide warnings, instructions, and training.

Prevention is normally stronger than detection. A poka-yoke device that prevents incorrect assembly is generally preferable to an end-of-line inspection that discovers the error after assembly.

6. Economics of quality

The cost of quality (COQ) is the cost associated with achieving quality and the cost arising when quality is not achieved. A conventional classification is:

Prevention costs

Prevention costs are incurred to avoid defects and failures. Examples include design reviews, capability studies, supplier development, preventive maintenance, training, process qualification, robust design, and mistake-proofing.

Appraisal costs

Appraisal costs arise from evaluating conformity. Examples include inspection, testing, calibration, measurement-system analysis, audits of product, and laboratory activities.

Internal-failure costs

These occur before delivery to the customer: scrap, rework, retesting, downtime caused by defects, material review, and yield loss.

External-failure costs

These occur after delivery: warranty repair, returns, complaint handling, field service, recalls, liability, lost reputation, and lost future sales.

The accounting identity is:

where C_P is prevention cost, C_A appraisal cost, C_IF internal-failure cost, and C_EF external-failure cost. This is a classification model, not a universal accounting rule; organizations may define categories differently. Opportunity costs and intangible reputation effects are often difficult to measure and may be excluded from formal totals.

Prevention–appraisal–failure trade-off

Increasing prevention can reduce internal and external failure. Increasing appraisal may reduce the number of defective units shipped, but inspection does not necessarily reduce the underlying defect rate. The economically preferred solution is usually near the point where the marginal cost of additional prevention or appraisal equals the marginal reduction in failure cost, subject to safety and compliance constraints.

A simple decision rule compares alternatives using total relevant cost:

where TC is total relevant cost, C₀ is other unchanged cost, and C_F is expected failure cost. For expected failure cost:

where p_r is the probability of failure scenario r, L_r is its loss, and q is the number of scenarios. The probabilities should refer to a defined population and time period, and the losses should include relevant repair, disruption, liability, and other measurable consequences.

Worked economic example

A manufacturer is considering an automated assembly verification system. The system costs £18,000 per year to operate. Without it, expected annual internal and external failure costs are £52,000. With it, expected failure costs fall to £21,000. Other costs are unchanged.

Without the system:

TC₀ = £52,000

With the system:

TC₁ = £18,000 + £21,000 = £39,000

Annual saving:

£52,000 − £39,000 = £13,000

The system is economically attractive on these assumptions, before considering its useful life, discount rate, tax, training, maintenance, and residual value. If the system also prevents a safety-critical escape, the decision cannot be based on cost alone.

For multi-year investments, present value may be used. If a future cash flow CF_t occurs at the end of year t and the discount rate per period is r, its present value is:

The sign convention must be consistent: benefits may be treated as positive and costs as negative, or all cash outflows may be listed separately. The calculation assumes a known discount rate and comparable cash-flow periods.

7. Applied case: quality planning for a gearbox housing

The following is an illustrative case, not a report of a particular company.

Background

A manufacturer develops an aluminium gearbox housing for industrial conveyors. Customers report three priorities: low noise, leak-free operation, and rapid replacement of the housing during maintenance. Previous prototypes met dimensional drawings but showed oil leakage and inconsistent bearing alignment.

Problem analysis

The design team collected warranty data, interviewed maintenance technicians, and observed assembly. The underlying needs were translated as follows:

A QFD workshop identified bearing-bore positional accuracy, sealing-surface flatness, casting integrity, access clearance, and alignment datum design as high-priority engineering characteristics. The technical correlation roof revealed a trade-off: increasing wall thickness could improve stiffness but increase mass and casting cost.

Approach and solution

The team adopted several controls:

• datum features were redesigned so that machining and inspection referenced functional surfaces;
• a pressure-decay test was introduced after machining;
• a fixture prevented incorrect orientation during assembly;
• a measurement-system study verified the adequacy of bore and flatness measurements;
• process capability was monitored for critical bore dimensions;
• supplier controls were added for casting porosity and material certification;
• maintenance technicians participated in validation trials.

The FMEA identified oil leakage as high severity because it could cause equipment downtime and environmental contamination. Rather than relying only on final leak testing, the team improved casting-process control and sealing-surface machining. The final test remained necessary as a containment and verification measure.

Outcome

The revised plan reduced rework and field complaints in the trial production run. More importantly, the organization gained traceability from customer need to engineering characteristic, process control, and evidence of conformity. The case demonstrates that quality planning is most effective when design, manufacturing, suppliers, service personnel, and customers contribute before production is finalized.

Learning points

1. A drawing is only one part of the quality requirement.
2. Customer needs must be converted into measurable characteristics.
3. Prevention and robust design should precede reliance on inspection.
4. Risk priorities should include severity and regulatory consequences, not only an arithmetic ranking.
5. Economic evaluation should include failure costs and the life-cycle consequences of poor quality.

8. Professional practice, governance, and contemporary developments

Quality decisions must respect safety, environmental law, product liability, contractual obligations, and professional ethics. Engineers must not conceal nonconformances, manipulate inspection data, or accept an unverified deviation merely to meet a production target. Traceability is especially important for safety-critical components, where material identity, process history, calibration status, and approval authority may be legally significant.

Digital manufacturing systems increasingly connect requirements management, product-lifecycle management, measurement equipment, manufacturing execution systems, and supplier records. Digital threads can improve traceability, but they introduce risks involving incorrect data linkage, cybersecurity, access control, and undocumented software changes. Artificial intelligence can assist with complaint classification, anomaly detection, and prioritization of inspection, but its output requires validated data, human oversight, explainable decision rules where appropriate, and protection against biased or incomplete training data. AI should support—not replace—engineering accountability.

Best practice includes maintaining a controlled requirements baseline, using cross-functional reviews, validating measurement systems, distinguishing prevention from detection, documenting assumptions in economic analyses, and revisiting quality plans after field feedback or process change. Quality planning is not a one-time document; it is a living control-and-learning system.

Chapter Summary

Quality planning translates customer and stakeholder needs into measurable product and process requirements, targets, responsibilities, verification methods, and responses to nonconformance. Requirements should progress from customer language to engineering characteristics and then to process controls and objective evidence.

QFD provides a structured method for making this translation. Its House of Quality relates customer requirements to engineering characteristics, technical interactions, competitive information, and priorities. Weighted calculations can support decisions, but the resulting numbers depend on subjective ratings and should not replace engineering judgement.

Quality assurance establishes confidence through planned systems and preventive activities. Quality control uses operational measurement, inspection, testing, and process monitoring to achieve and verify conformity. Improvement addresses root causes and raises performance over time.

Risk-based planning, including FMEA, directs attention toward important failure modes. Severity, occurrence, and detection ratings can support prioritization, but RPN is not a probability and cannot override safety or regulatory requirements. Prevention is generally stronger than detection.

The cost of quality includes prevention, appraisal, internal failure, and external failure costs. A sound economic decision compares total relevant life-cycle costs, including expected failure losses, while recognizing that safety, legal compliance, and ethical obligations may justify action beyond the lowest monetary cost.

Key Terms

Voice of the customer: Organized representation of customer expectations and needs.

Quality planning: Establishing objectives, processes, resources, responsibilities, and controls required to achieve quality.

Specification: A documented requirement defining a characteristic, target, limit, condition, or method of verification.

Verification: Confirmation that specified requirements have been fulfilled.

Validation: Confirmation that a product is suitable for its intended use.

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

House of Quality: The principal QFD matrix relating customer requirements to engineering characteristics.

Engineering characteristic: A measurable product or process variable used to satisfy a customer requirement.

Quality assurance: Planned and systematic activities that provide confidence that requirements will be fulfilled.

Quality control: Operational techniques used to achieve and verify conformity.

Inspection: Comparison of a product or characteristic with specified requirements.

Prevention cost: Cost incurred to prevent defects or failures.

Appraisal cost: Cost incurred to evaluate conformity.

Internal-failure cost: Cost arising from nonconformance before delivery.

External-failure cost: Cost arising from nonconformance after delivery.

Failure Mode and Effects Analysis: Structured analysis of failure modes, effects, causes, and controls.

Risk Priority Number: Product of severity, occurrence, and detection ratings in traditional FMEA.

Process capability: The ability of a stable process to produce output within specification limits.

Traceability: Ability to follow the history, application, location, or evidence associated with an item or process.

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

Review Questions

Short Answer Questions

1. What is the difference between verification and validation?
2. Why must customer statements be translated into measurable characteristics?
3. Name the four conventional categories of cost of quality.
4. Distinguish quality assurance from quality control.
5. Why is prevention generally preferred to final inspection?

Long Answer Questions

1. Explain the stages of quality planning from customer identification to process monitoring.
2. Describe the structure and purpose of a QFD House of Quality.
3. Discuss the assumptions and limitations of C_p and C_pk.
4. Explain how FMEA supports risk-based quality planning and why RPN must be interpreted cautiously.
5. Discuss the prevention–appraisal–failure cost trade-off using an engineering example.

Analytical / Application Questions

1. A customer requests a pump that is “quiet, reliable, and easy to maintain.” Convert these statements into at least six measurable requirements, including targets and verification methods.
2. A process has USL = 25.10 mm, LSL = 24.90 mm, μ = 25.04 mm, and σ = 0.020 mm. Calculate C_p and C_pk, then explain what the difference indicates about centring.
3. A proposed prevention project costs £12,000 annually and is expected to reduce internal failure costs by £8,000 and external failure costs by £10,000 per year. Evaluate the direct annual economic effect and identify non-financial factors that could change the decision.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes quality assurance from quality control?

A. Quality assurance is limited to final inspection, while quality control is limited to design.

B. Quality assurance provides planned confidence in processes, while quality control uses operational techniques to achieve and verify conformity.

C. Quality assurance measures dimensions, while quality control manages suppliers only.

D. Quality assurance is performed after delivery, while quality control is performed before design.

Correct Answer: B. Quality assurance provides planned confidence in processes, while quality control uses operational techniques to achieve and verify conformity.

Question 2

In a QFD calculation, customer importance ratings are 5, 4, and 3, and the relationship weights for one engineering characteristic are 9, 1, and 0. What is its weighted priority?

A. 37

B. 45

C. 49

D. 54

Correct Answer: C. 49

Question 3

Which item is normally classified as an external-failure cost?

A. Design review

B. Incoming material inspection

C. Scrap discovered before shipment

D. Warranty repair in the field

Correct Answer: D. Warranty repair in the field

Question 4

For a stable, approximately normal process, what does Cpk account for that Cp does not?

A. Inspection cost

B. Process centring relative to specification limits

C. Customer importance ratings

D. Number of suppliers

Correct Answer: B. Process centring relative to specification limits

Question 5

A traditional FMEA risk priority number is calculated as:

A. S + O + D

B. S − O − D

C. S × O × D

D. S ÷ O ÷ D

Correct Answer: C. S × O × D

Question 6

Which is the strongest example of prevention?

A. Sorting finished parts

B. Retesting rejected parts

C. Adding a fixture that prevents incorrect assembly

D. Issuing a customer warning label after shipment

Correct Answer: C. Adding a fixture that prevents incorrect assembly

Question 7

A quality project costs £18,000 annually and reduces expected failure costs from £52,000 to £21,000. What is the direct annual saving?

A. £5,000

B. £13,000

C. £18,000

D. £31,000

Correct Answer: B. £13,000

Question 8

Which sequence best represents specification development?

A. Test result → customer complaint → nominal value → need

B. Customer statement → interpreted need → measurable characteristic → target and test method

C. Inspection cost → supplier audit → customer statement → drawing

D. Failure mode → invoice → customer need → advertising claim

Correct Answer: B. Customer statement → interpreted need → measurable characteristic → target and test method

Question 9

Why should a high-severity safety risk not automatically be accepted because its RPN is low?

A. RPN is always measured in millimetres.

B. Severity may require action independently of the combined numerical ranking.

C. Occurrence ratings are never used in FMEA.

D. Detection ratings represent product price.

Correct Answer: B. Severity may require action independently of the combined numerical ranking.

Question 10

Which statement about inspection is most accurate?

A. Inspection always removes the root cause of variation.

B. Inspection is identical to quality improvement.

C. Inspection compares characteristics with requirements but may detect defects only after value has been added.

D. Inspection eliminates the need for process control.

Correct Answer: C. Inspection compares characteristics with requirements but may detect defects only after value has been added.