CQE flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

Cost of poor quality

The sum of internal failure costs (scrap, rework found before shipment) and external failure costs (returns, warranty, recalls after shipment). It excludes prevention and appraisal spending, which are conformance costs.

Juran Trilogy

Juran's model organizing quality management into three processes: quality planning (designing to meet customer needs), quality control (meeting goals during operation), and quality improvement (breaking through to new performance levels).

Transformational leadership

A leadership style in which the leader inspires followers through idealized influence, intellectual stimulation, individualized consideration, and inspirational motivation, moving them beyond reward-for-performance exchange toward shared goals and higher performance.

Tuckman team stages

The five predictable stages of team development: forming (orientation), storming (conflict over roles and approach), norming (agreement on rules), performing (task focus), and adjourning (disbanding after completion).

Herzberg two-factor theory

Herzberg's finding that hygiene factors such as pay, working conditions, and policy prevent dissatisfaction but do not motivate, while motivators such as achievement, recognition, responsibility, and growth create genuine satisfaction.

Second-party audit

An audit performed by an organization on its supplier, or by a customer on its supplier, to qualify or monitor a supply chain partner. It is distinct from internal first-party and independent third-party certification audits.

ISO 9001 clause structure

ISO 9001:2015 is organized into ten clauses: scope, normative references, terms and definitions, context of the organization, leadership, planning, support, operation, performance evaluation, and improvement.

Corrective action

Action taken to eliminate the cause of a detected nonconformity so it does not recur. It follows containment and root-cause analysis and is verified for effectiveness, unlike correction, which only fixes the defect.

Baldrige Criteria categories

The seven Malcolm Baldrige National Quality Award categories: leadership; strategy; customers; measurement, analysis and knowledge management; workforce; operations; and results, scored out of a possible 1000 points.

Design verification

Confirmation, usually by review, analysis, or test, that design outputs meet the specified design input requirements. It answers whether the product was built right to the specification.

Design validation

Confirmation through testing on production units or with actual users that the resulting product or service fulfills intended use and customer needs. It answers whether the right product was designed.

Quality function deployment

A structured planning method that translates customer requirements, the whats, into technical design requirements, the hows, using a relationship matrix known as the house of quality with competitive evaluations.

Taguchi loss function

A quadratic model, L = k(y - T)^2, showing that financial loss grows with the square of deviation from the target value T, so reducing variation around target reduces loss even within tolerance.

Series system reliability

For independent components arranged so that all must function, system reliability equals the product of individual reliabilities (Rs = R1 x R2 x ... x Rn), so it is always lower than the weakest component.

Process capability index Cpk

Cpk = min[(USL - mean)/3 sigma, (mean - LSL)/3 sigma]. It measures how well a centered or shifted process fits within specification limits; values of 1.33 or greater are typically required.

X-bar and R chart limits

For subgroups of n = 2 to 10, X-bar limits are the grand mean plus or minus A2 times the average range, and R limits are D3 and D4 times the average range, using constants from tables.

Gage R&R acceptance

Measurement system study rule: percent GRR below 10% of total variation is acceptable, 10% to 30% may be acceptable depending on application and cost, and above 30% is unacceptable and requires improvement.

Producer's risk

The probability, denoted alpha, that a lot of acceptable quality at or better than the Acceptable Quality Level is rejected by the sampling plan. It corresponds to a Type I statistical error.

Rule one out of control

The first Western Electric and Nelson sensitizing rule: a single plotted point falls beyond either 3-sigma control limit, indicating a special cause that should be investigated immediately.

Seven basic quality tools

The seven classic graphical tools: Pareto chart, cause-and-effect (fishbone) diagram, check sheet, control chart, histogram, scatter diagram, and stratification or flowchart. They are simple aids for problem solving.

Poka-yoke

A Japanese term for mistake-proofing: a device or design feature that prevents an error from being made or detects it immediately, so that defects cannot pass to the next operation.

Overall equipment effectiveness

OEE = availability x performance x quality, where availability is uptime over planned time, performance is actual versus ideal speed, and quality is good units over total units produced.

DMAIC phases

The Six Sigma improvement roadmap: Define the problem and customer requirements, Measure baseline performance, Analyze root causes, Improve the process with verified solutions, and Control the gains with monitoring.

Central limit theorem

For samples of size n drawn from any distribution, the sampling distribution of the mean approaches normal as n grows, with mean mu and standard error sigma/sqrt(n); n of 30 or more usually suffices.

Type I error

Rejecting a true null hypothesis, also called producer's or alpha risk. Its probability is the significance level alpha, commonly 0.05, and reducing alpha lowers the chance of false alarms.

Coefficient of variation

The ratio of the standard deviation to the mean, CV = s / x-bar, usually expressed as a percentage. It compares relative variability across data sets with different units or very different means.

Fractional factorial design

A designed experiment using only a fraction, such as 2^(k-p), of the runs of a full factorial, trading the ability to estimate some high-order interactions for fewer trials; aliasing confounds those effects.

Coefficient of determination

R-squared, the proportion of total variation in the response variable that is explained by the regression model. It ranges from 0 to 1, and higher values indicate a better-fitting model.

Risk treatment options

The standard responses to an identified risk: avoid the activity, reduce (mitigate) its likelihood or impact, transfer or share it through insurance or contract, or accept and monitor it within risk tolerance.

Fault tree analysis

A top-down deductive technique that starts with an undesired top event and works backward through AND and OR logic gates to identify combinations of basic failures, called minimal cut sets, that cause it.

Risk priority number

In FMEA, RPN = severity x occurrence x detection, each scored from 1 to 10. Higher values flag failure modes needing action, though high severity should also be considered independently of the total.

HAZOP

Hazard and operability study: a team-based, systematic review of a process that applies guide words such as no, more, less, and reverse to each node parameter to find deviations and their consequences.