ICGB flashcards
ICGB · Build recall with a focused review session.
32 cards. Read the term and the definition here. The deck at the top of this page is optional.
- Project charter
- A document approved by the sponsor that states the business case, problem and goal statements, scope, team, and timeline, authorizing the project to begin.
- SMART goal statement
- A goal written to be specific, measurable, achievable, relevant, and time-bound, including metric, baseline, target, and deadline so success can be verified.
- SIPOC diagram
- A high-level map of Suppliers, Inputs, Process, Outputs, and Customers that defines process boundaries without detailing internal steps.
- Kano model
- A framework classifying requirements as must-be, performance, attractive, indifferent, or reverse, based on how their presence or absence affects customer satisfaction.
- Cost of poor quality
- The sum of internal failure, external failure, appraisal, and prevention costs, used to quantify the financial impact of poor quality and prioritize improvement.
- Gage R&R
- A measurement system analysis that quantifies repeatability and reproducibility variation; percent GRR under 10 is acceptable, 10 to 30 conditional, over 30 unacceptable.
- Cp and Cpk
- Cp equals specification width divided by six sigma and ignores centering; Cpk divides the nearer specification distance by three sigma and accounts for off-center processes.
- DPMO
- Defects per million opportunities, calculated as defects divided by units times opportunities, then multiplied by one million, used to express process performance.
- Rolled throughput yield
- The product of each process step's first-pass yield, exposing hidden factory losses that final inspection yield can mask.
- Attribute agreement kappa
- A statistic correcting observed agreement for chance agreement; above 0.75 is good, 0.40 to 0.75 fair, and below 0.40 poor.
- Hypothesis test logic
- The null hypothesis assumes no difference or effect; the p-value is the probability of data at least as extreme given the null; reject the null when p is below alpha.
- Type I and Type II errors
- Type I error rejects a true null hypothesis and equals alpha; Type II error fails to reject a false null and equals beta; power is one minus beta.
- ANOVA F test
- A test comparing between-group variation to within-group variation for three or more means; a significant result requires post-hoc pairwise comparison to locate differences.
- Multi-Vari analysis
- A graphical technique stratifying variation by within-piece, piece-to-piece, and time-to-time sources to narrow the search for dominant causes.
- FMEA RPN
- The risk priority number equals severity times occurrence times detection, each rated one to ten; high severity items are addressed regardless of RPN.
- Full factorial design
- An experiment with k factors at two levels requiring two to the k runs, estimating all main effects and interactions but growing rapidly with factor count.
- Resolution IV design
- A fractional factorial design in which main effects are clear of two-factor interactions, but two-factor interactions are confounded with each other.
- Response surface methodology
- A set of designs, such as central composite or Box-Behnken, that fit quadratic models to locate optimum factor settings after screening.
- SMED
- Single-minute exchange of dies, a lean method that converts internal setup performed while the machine is stopped into external setup performed while it runs.
- Poka-yoke
- An error-proofing device or method that prevents a mistake from occurring or detects it immediately at the source rather than downstream.
- Control plan
- A document listing process step, characteristic, specification, measurement method, sample size and frequency, control method, and reaction plan for each controlled item.
- Control versus specification limits
- Control limits are calculated from process variation and detect special cause; specification limits are set by the customer or design and define acceptability.
- Western Electric rules
- Signal rules flagging special cause, including one point beyond three sigma, eight consecutive points on one side, and six consecutive trending points.
- Standard work
- Documented current best method, takt time, and required inventory that sustains improvement gains and provides a baseline for further improvement.
- Reaction plan
- A defined action and owner triggered when a control chart signals special cause, typically stop, contain, and escalate to the process owner.
- DMAIC roadmap
- The Define, Measure, Analyze, Improve, Control sequence for improving existing processes, with tollgate reviews verifying deliverables before each phase advance.
- Y equals f of X
- The model treating the output Y as a function of input X variables; Measure quantifies Y, Analyze finds vital X, Improve optimizes X, Control monitors both.
- Pareto analysis
- A ranking of causes by frequency or cost applying the 80/20 rule to identify the vital few problems that deserve priority action.
- Little's Law
- Average inventory equals average throughput multiplied by average flow time, linking work-in-process, completion rate, and cycle time.
- Takt time
- Available production time divided by customer demand, setting the pace at which a process must complete one unit to meet demand.
- Run chart
- A time-ordered plot of data with a centerline that detects trends and shifts before control limits are established from sufficient data.
- Confidence interval
- A range estimate around a statistic whose width decreases as sample size increases and as variability decreases; a 95 percent interval excludes the true value five percent of the time.