CSSGB flashcards

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

Six Sigma definition

A data-driven methodology that reduces process variation and defects to 3.4 DPMO, using DMAIC for improving existing processes and DFSS for designing new ones.

Y equals f of x

The foundational Six Sigma model in which Y, the process output or critical-to-quality characteristic, is a function of the input variables x that must be found and controlled.

1.5 sigma shift

An adjustment applied to long-term sigma estimates so that a 6-sigma process equals 3.4 DPMO; it accounts for typical process drift over time.

Cost of poor quality

The combined cost of internal failures, external failures, appraisal, and prevention activities; it typically equals 15-25% of revenue in a 3-4 sigma organization.

Green Belt role

A part-time improvement team leader who applies Six Sigma tools to projects inside their own work area while continuing to perform regular job duties.

Eight lean wastes

Defects, overproduction, waiting, non-utilized talent, transportation, inventory, motion, and excess processing; the set is recalled with the acronym DOWNTIME.

Project charter

A document that authorizes a project by stating the business case, problem, goal, scope, team and roles, milestones, and expected financial benefit.

SIPOC diagram

A high-level map of Suppliers, Inputs, Process, Outputs, and Customers, created during Define to establish and communicate the boundaries of project scope.

Kano model

It classifies customer requirements as dissatisfiers (must-be), satisfiers (performance), or delighters (attractive); delighters create satisfaction far beyond their cost.

SMART goal

A goal criterion requiring the statement to be Specific, Measurable, Achievable, Relevant, and Time-bound, usually expressed as a quantified target improvement.

RACI matrix

A responsibility chart that assigns each task exactly one Accountable owner, plus those Responsible for the work, Consulted for input, and Informed of results.

Repeatability vs reproducibility

Repeatability is variation when one appraiser measures the same parts repeatedly; reproducibility is the variation between different appraisers measuring the same parts.

Gage R&R acceptance

Percent study variation below 10% is acceptable, between 10% and 30% is marginal and judged by risk and cost, and above 30% is unacceptable.

Cp versus Cpk

Cp compares specification width to six sigma and ignores centering; Cpk accounts for how far the process mean lies from the nearest specification limit.

Rolled throughput yield

The product of the first-pass yields of every step in a multi-step process, showing the probability that a unit passes without any rework.

DPMO

Defects per million opportunities, computed as defects divided by the product of units and opportunities per unit, then multiplied by one million.

Attribute agreement Kappa

A chance-corrected measure of agreement among appraisers; above 0.90 is excellent, 0.70 to 0.90 is acceptable, and below 0.70 requires improvement.

Type I error

The alpha error: rejecting a true null hypothesis and concluding a difference exists when it does not; its probability equals the chosen significance level alpha.

Statistical power

The probability of correctly rejecting a false null hypothesis, equal to 1 minus beta; larger samples and larger true effects both increase power.

Paired t-test

A hypothesis test for two related measurements taken on the same units, such as before-and-after readings, which analyzes the differences rather than the raw values.

ANOVA

Analysis of variance: it tests whether three or more group means differ by comparing between-group variation with within-group variation using an F-statistic.

Correlation coefficient

A value between -1 and +1 that measures the strength and direction of linear association between two variables; it does not establish causation.

Full factorial design

An experiment that runs every combination of factor levels; k factors at two levels need 2^k runs and allow estimation of main effects and interactions.

Main effect

The average change in the response when one factor moves from its low to its high level, averaged across all levels of the other factors.

Risk Priority Number

In FMEA, Severity multiplied by Occurrence multiplied by Detection; it ranks failure modes so that the highest scores receive corrective action first.

Poka-yoke

Mistake proofing: a device or design feature that prevents an error from occurring or prevents a defect from passing to the next process step.

Blocking in DOE

A design technique that groups experimental runs to remove the effect of a known nuisance variable, such as shift or machine, from the error estimate.

Control plan

A document specifying for each process step the characteristic and specification, measurement method, sample size and frequency, control method, and reaction plan.

Control vs specification limits

Control limits are computed from process data at plus or minus three sigma; specification limits are customer-defined tolerances and are never plotted as control chart limits.

X-bar and R chart

A paired SPC chart for continuous data using subgroups of about two to eight, plotting subgroup averages to monitor centering and subgroup ranges to monitor variation.

Western Electric rules

Out-of-control signals including one point beyond 3 sigma, two of three beyond 2 sigma, four of five beyond 1 sigma, and eight consecutive points on one side.

Out-of-control action plan

The documented response an operator follows when a control chart signals, including containment of suspect product, who to notify, and the corrective steps.

Rational subgroup

A sample chosen so that variation within the subgroup reflects only common cause, while shifts between subgroups reflect the process changes being monitored.