Question 1 of 6: Cost of Quality, Supplier Certification, and ISO 9000

Question 1: Cost of Quality, Supplier Certification, and ISO 9000 (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

(a) The four cost-of-quality categories, and how total COQ is lowered

The cost of quality (COQ) is conventionally split into four categories. Prevention costs are incurred to keep nonconformities from happening in the first place — quality planning and engineering, SPC and process-capability studies, supplier qualification, design reviews, and quality training. Appraisal costs are incurred to find nonconformities that have already occurred — incoming, in-process and final inspection and test, calibration of measuring equipment, and quality audits. Internal failure costs arise when a nonconformity is found before the product leaves the plant — scrap, rework, re-inspection, downtime, and yield loss. External failure costs arise when a nonconformity is found after the product has reached the customer — warranty claims, returns and allowances, complaint handling, product liability, recalls, and lost goodwill/future sales; these are almost always the most expensive category, both because of direct cost and because they damage reputation.

A company lowers its total COQ — not just one category — by deliberately shifting spend upstream, toward prevention. The well-known "1-10-100" rule captures why: a nonconformity that costs one unit of money to prevent at the design/process stage typically costs on the order of ten times as much to catch and rework internally, and a hundred times as much or more once it escapes to the customer. Juran's cost-of-quality curve formalizes the resulting trade-off: as prevention (and to a lesser extent appraisal) spending rises, failure costs fall much faster than prevention spending rises, so total COQ actually decreases even though one of its four components is growing — up to a point of diminishing returns near zero defects, where further prevention spend buys progressively less failure-cost reduction. The methodologies and tools used to execute this shift include statistical process control (control charts) to keep processes in control and reduce internal failures; design of experiments and Taguchi robust-design methods to build insensitivity to variation in at the design stage (prevention, not appraisal); FMEA (failure modes and effects analysis) to anticipate and prevent failure modes before they occur; Six Sigma's DMAIC cycle and TQM more broadly, which institutionalize continuous improvement; poka-yoke (mistake-proofing) to make errors physically impossible rather than merely detectable; and ISO 9001-based quality management systems, which provide the process discipline and documentation that make the other tools stick.

(b) Supplier-producer relations, process-capability testing, and comparison with acceptance sampling

Supplier-producer relations have moved away from the traditional adversarial, lowest-bid, multi-source model toward long-term partnerships with a small number of certified suppliers who share responsibility for quality rather than simply meeting a purchase-order specification. In this modern model the buyer increasingly relies on process-capability certification in place of lot-by-lot incoming inspection: the supplier demonstrates, using its own control-chart history and capability indices ($C_p$, $C_{pk}$), that its process is in statistical control and capable of meeting the specification, and the buyer accepts material on a "ship-to-stock" basis without re-inspecting every lot.

Because $C_{pk}$ is itself estimated from a finite sample, a capability claim carries the same two kinds of statistical risk as an acceptance-sampling decision. The supplier's (producer's) risk is the probability that a genuinely capable process ($C_{pk}$ truly at or above the required minimum) is rejected by the capability test purely because of sampling variation in the estimate — a Type I error, analogous to $\alpha$. The producer's-customer, i.e. the buyer's (consumer's) risk is the probability that a genuinely incapable process is accepted as capable because the sample happened to look better than the true process — a Type II error, analogous to $\beta$. Both risks are controlled the same way they are in any hypothesis test: by choosing an adequately large sample (number of parts/lots in the capability study, which sets the width of the sampling distribution of $\hat C_{pk}$) and by stating the acceptance criterion as a required lower confidence bound on $C_{pk}$ (not just the point estimate), so that a supplier is only certified once there is high confidence the true capability clears the bar.

Compared with traditional acceptance sampling (e.g. MIL-STD-105E), process-capability certification is generally the more effective approach for an ongoing supply relationship. Acceptance sampling is reactive: it screens finished lots after the fact, provides no diagnostic feedback about the process that produced them, becomes prohibitively expensive at very low AQLs (because the required sample size grows quickly), and must be repeated on every lot indefinitely. Capability certification is proactive: it verifies the process itself is stable and capable once, continuously self-monitors via the supplier's own control charts thereafter, and gives the supplier the diagnostic information needed to improve rather than just a pass/fail verdict on a sample. It is therefore preferred wherever an ongoing partnership (rather than a one-off purchase from an unknown source) is in place; acceptance sampling remains appropriate for occasional or unqualified suppliers where no capability history exists.

(c) Purpose and steps of quality certification, and the ISO 9000:2000 revision

Quality certification (typically to ISO 9001) serves both parties in a transaction. For the supplier, it is an independent, internationally recognized credential that opens market access, reduces the number of separate customer audits it must host, and forces internal process discipline. For the producer/buyer, it substitutes third-party assurance for its own incoming inspection and supplier audits, letting it trust that a certified supplier's quality management system (QMS) is capable of consistently meeting specification without re-verifying every shipment.

The main steps in achieving certification are: (1) a management-commitment and gap-analysis phase, comparing current practice against the standard's requirements; (2) documenting the QMS — a quality manual, procedures, and work instructions; (3) implementation and staff training; (4) internal audits and corrective action on any nonconformities found; (5) a formal management review; (6) an external registrar audit, normally in two stages — a documentation review followed by an on-site assessment; (7) award of the certificate; and (8) ongoing surveillance audits (annually or so) with full recertification on a roughly three-year cycle.

The ISO 9000:2000 revision was a substantial rewrite of the 1994 series. It replaced the old prescriptive, twenty-element, function-organized checklist with a process-based model built around the Plan-Do-Check-Act cycle and eight explicit quality-management principles (customer focus, leadership, involvement of people, the process approach, a systems approach to management, continual improvement, factual decision-making, and mutually beneficial supplier relationships). It consolidated three separate standards (ISO 9001/9002/9003, which had covered design+production, production only, and final inspection only) into a single certifiable standard, ISO 9001, with permitted exclusions for clauses that genuinely do not apply. It added explicit, previously-absent requirements to measure customer satisfaction and to demonstrate continual improvement of the QMS, rather than merely maintaining conformance. It increased top-management's documented responsibility for the QMS. And it aligned its structure with ISO 14001 (environmental management), making integrated management systems practical for the first time.