Question 1 of 6: Quality Philosophy, Concurrent Engineering, and Cost of Quality

Question 1: Quality Philosophy, Concurrent Engineering, and Cost of Quality (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) Traditional vs. Taguchi definitions of quality; loss functions; quality–profit–productivity relation

The traditional (conformance-based) definition treats quality as a binary, "goal-post" property: a unit is good if its characteristic falls anywhere inside the specification limits, and defective the instant it steps outside them, no matter how close to the limit. The associated traditional loss function is a step function: $L(y)=0$ for $LSL\le y\le USL$ and $L(y)=L_0$ (a fixed cost) the moment $y$ crosses either limit — it treats a unit produced exactly on target and a unit produced just inside the limit as equally "perfect," and a unit one micron outside the limit as suddenly worthless.

Taguchi's definition reframes quality as "the loss a product imparts to society from the time it is shipped," and models that loss with a continuous quadratic loss function $L(y)=k(y-T)^2$, where $T$ is the target value and $k$ is a cost coefficient fixed by the loss at the specification limit ($k=L_0/(USL-T)^2$ or $L_0/\Delta^2$ for a symmetric tolerance $\Delta$). Under this view, ANY deviation from target incurs a loss that grows smoothly with the square of the deviation — a unit that is in-spec but 90% of the way to the limit is measurably worse than one sitting exactly on target, even though the traditional view would call both "100% good." Taguchi's approach therefore drives design and process improvement toward minimizing variation around the target itself, not merely toward staying inside the tolerance band.

Quality and profit are linked through the cost-of-quality chain: better quality (whether measured the traditional or the Taguchi way) reduces scrap, rework, warranty claims and customer complaints, directly lowering the cost of poor quality; it simultaneously supports premium pricing, repeat business and market-share growth through customer loyalty, raising revenue. Both effects widen margin. Quality and productivity are linked through Deming's chain reaction: improving quality reduces the fraction of output that must be scrapped or reworked, which raises the effective first-pass output per unit of input (labour, material, machine time) for the same nominal capacity — i.e., productivity rises as a direct consequence of fewer defects, not as a separate initiative competing for the same resources.

(b) Concurrent engineering, the Malcolm Baldrige Award, quality certification, and ISO/TS16949

Concurrent engineering (CE) replaces the traditional sequential "over-the-wall" hand-off (design → process engineering → manufacturing → quality) with cross-functional teams that design the product and its downstream manufacturing, quality and service processes simultaneously. Because manufacturability, testability and quality requirements are considered while the design is still fluid, CE catches design-for-quality and design-for-manufacture problems before tooling is committed, sharply cutting late engineering changes, scrap and rework — the very waste categories lean manufacturing targets — while also compressing time-to-market, which is itself a core lean/JIT objective.

The Malcolm Baldrige National Quality Award (MBNQA) is a U.S. award that recognizes performance excellence, judged against seven criteria (leadership; strategic planning; customer focus; measurement, analysis and knowledge management; workforce focus; operations focus; and results). It is fundamentally a self-assessment and improvement framework: an organization applies, is scored against the criteria by examiners, and a small number of winners are recognized annually. This differs from quality certification (e.g. ISO 9001) in a fundamental way: certification is a pass/fail, third-party audit of whether a documented quality management system conforms to a fixed standard, renewed through recurring surveillance audits, and says nothing about how excellent the results are — only that the system exists and is followed. The Baldrige Award is competitive, holistic, and results-oriented, not a recurring conformance certificate.

ISO/TS 16949 (now superseded by IATF 16949) is the automotive-sector-specific quality management standard, built on the ISO 9001 framework with additional automotive requirements: Advanced Product Quality Planning (APQP), the Production Part Approval Process (PPAP), control plans, Measurement Systems Analysis (MSA), and an explicit continuous-improvement/defect-prevention/variation-reduction mandate across the supply chain. Most automotive OEMs require their suppliers to hold this certification, in addition to (not instead of) ISO 9001.

(c) The four cost-of-quality categories; how they move during improvement; Six Sigma and DMAIC

The classic four quality-cost categories are: prevention costs (quality planning, training, process control, supplier qualification — spent to keep defects from occurring); appraisal costs (inspection, testing, calibration — spent to detect defects that do occur); internal failure costs (scrap, rework, downtime for defects caught before shipment); and external failure costs (warranty, returns, complaints, liability, and lost future business for defects that reach the customer).

In a successful improvement effort, the short-term pattern is that appraisal costs often rise first (more inspection is deployed to characterize and catch the problem while root causes are being found), while internal failure costs begin to fall as the first fixes take hold. Over the longer term, sustained investment in prevention (process capability studies, mistake-proofing, supplier development) drives internal AND external failure costs down substantially — historically the largest and most visible savings — and, once the process is demonstrably capable and in control, appraisal costs can also be reduced (less inspection is needed to catch problems that are no longer occurring), while prevention spending is retained at a modest, institutionalized level. The net effect is that total cost of quality falls well below its pre-improvement level even though one category (appraisal) may have temporarily risen along the way.

Six Sigma is a disciplined, statistically driven methodology aimed at reducing process variation to the point where the nearest specification limit sits at least six standard deviations from the mean (nominally 3.4 defects per million opportunities, allowing for a typical $\pm1.5\sigma$ long-term mean shift). Improvement projects follow the DMAIC cycle: Define the problem, project scope and critical-to-quality (CTQ) characteristics; Measure current performance and validate the measurement system (e.g. Gage R&R); Analyze the data to identify root causes (hypothesis tests, regression, cause-and-effect analysis); Improve the process, frequently using designed experiments to optimize the vital few factors; and Control, where control plans and SPC charts are put in place to sustain the gains.