CQPA key terms

The total money an organization loses because work is done wrong the first time, including scrap, rework, warranty claims, recalls, and complaint handling. It excludes prevention and appraisal spending.

The sum of all money spent on prevention, appraisal, internal failure, and external failure. It reframes quality as an economic category rather than a purely technical one, letting analysts compare improvement spending against failure losses.

Money spent before production to stop defects from occurring, such as training, supplier development, process design review, and quality planning. It is an investment category and usually rises as failure costs fall.

Money spent to evaluate whether product or process conforms, including inspection, testing, calibration, audits, and lab work. It detects defects already present rather than preventing them.

Losses discovered after the customer receives the product: warranty, returns, field service, liability, and lost goodwill. These are the most expensive failure category because they include damage to reputation.

Losses found before shipment, such as scrap, rework, re-inspection, retest, and downtime caused by defects. It is visible inside the plant and often the first target of improvement projects.

Corrective action taken on nonconforming product so it can meet requirements. Rework adds labor and cycle time without adding customer value, and repeated rework on the same unit signals a process control problem.

Nonconforming material that is discarded because it cannot be economically repaired to meet requirements. Scrap cost includes material, labor, and overhead already invested in the unit.

Structured input gathered from customers about needs, expectations, and satisfaction, collected through interviews, surveys, complaints, and observation. It is translated into measurable requirements that drive design and improvement priorities.

A measurable product or service characteristic that directly affects what the customer values, derived from the voice of the customer. CTQs convert vague desires into specifications a process can be controlled against.

A planning method that translates customer requirements into design characteristics and process controls using matrices, often called the house of quality. It forces tradeoffs among competing technical features to be made explicit early in design.

A framework classifying features as must-be, performance, or attractive based on how their presence affects satisfaction. Must-be features cause dissatisfaction when missing but do not delight when present; attractive features delight when present.

A scored measure of how well delivered performance meets customer expectations, tracked over time to detect trends. It complements hard metrics like complaints and returns by capturing perception.

A widely used criteria set for organizational performance management covering leadership, strategy, customers, measurement, workforce, and operations. Organizations use it for self-assessment rather than certification.

An international standard specifying requirements for a quality management system, centered on process approach, risk-based thinking, and continual improvement. Certification is granted by third-party auditors, not by ISO itself.

The documented set of processes, responsibilities, and resources an organization uses to direct and control quality. It provides the structure within which policies, procedures, records, and improvement activities operate.

A scheduled top-management evaluation of QMS performance, inputs such as audit results and customer feedback, and outputs such as resource decisions and improvement actions. It demonstrates leadership involvement and closes the improvement loop.

The practice of ensuring the right people use the current approved version of a document while obsolete copies are removed or marked. It covers creation, approval, distribution, revision, and retention.

An independent, planned examination of processes or the QMS against defined criteria, performed by trained personnel who do not audit their own work. Findings feed corrective action and management review.

A documented result of an audit, commonly graded as conforming, nonconformity, observation, or opportunity for improvement. Each nonconformity requires a response describing containment, cause, and corrective action.

Action taken to eliminate the cause of an existing nonconformity so it does not recur. It differs from containment, which only protects the customer while the cause is still under investigation.

Action taken to eliminate the cause of a potential nonconformity before it occurs, often triggered by risk analysis or trend data. It is forward-looking, whereas corrective action responds to something that already happened.

An immediate, temporary measure that stops nonconforming product from reaching the customer, such as sorting, quarantining, or 100 percent inspection. It buys time for root cause analysis but does not fix the process.

A systematic investigation to identify the underlying condition that, if removed, prevents recurrence of a problem. Common tools include the five whys, fishbone diagrams, and fault tree analysis.

An iterative questioning technique that asks why repeatedly, typically five times, to move from symptom to underlying cause. Its simplicity is a strength, but it depends on honest, evidence-based answers at each step.

A visual tool that sorts potential causes of a problem into categories, often people, machine, material, method, measurement, and environment. Also called a fishbone or Ishikawa diagram, it broadens thinking before data collection.

A bar chart of causes ordered by frequency or cost, paired with a cumulative line, used to separate the vital few from the trivial many. It guides where improvement effort will yield the largest return.

A measure of whether a stable process can meet specification limits. Cp compares spread to tolerance width; Cpk also accounts for how far the process mean sits from the target, so it is always the lower of the two.

Capability indices calculated from overall long-term variation rather than within-subgroup variation. They are used when the process is not yet shown to be in statistical control and are typically smaller than Cp and Cpk.

The use of control charts and other statistics to monitor a process over time and distinguish common-cause variation from special-cause variation. It supports decisions about when to adjust a process and when to leave it alone.

A time-ordered plot of a process statistic with a center line and statistically derived upper and lower control limits. Points outside the limits or nonrandom patterns signal special-cause variation.

Routine, inherent variation produced by the ordinary interaction of process inputs. Reducing it requires changing the system itself, not reacting to individual points, which would only increase variation.

Variation from an identifiable, nonroutine source such as a bad lot, a shift change, or a machine fault. It appears as points outside control limits or unusual patterns and calls for investigation of that specific event.

A set of studies that quantify how much observed variation comes from the measurement system rather than the product, covering repeatability, reproducibility, bias, linearity, and stability. It must precede capability studies, since bad data cannot validate a process.

A study separating measurement variation into repeatability, the same operator measuring the same part twice, and reproducibility, different operators measuring the same part. High values mean the gauge cannot reliably distinguish good from bad parts.

Comparing an instrument against a reference standard of known accuracy and adjusting or documenting the result. It establishes traceability and is a prerequisite for trusting any measurement used in decisions.

Deciding whether to accept a lot by inspecting a sample rather than every unit, using a plan that defines sample size and accept or reject criteria. It carries producer and consumer risk and is inappropriate for safety-critical characteristics.

A graph showing the probability of accepting a lot for each possible quality level under a given sampling plan. It quantifies producer risk, rejecting a good lot, and consumer risk, accepting a bad lot.

The quality level that a sampling plan is designed to accept most of the time, used as a reference point in acceptance sampling standards. It is not a license to ship defects and should not be applied to critical characteristics.