EXIN LSSA Six Sigma Black Belt flashcards

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

DPMO at six sigma

Defects per million opportunities at six sigma equals 3.4 when the conventional 1.5 sigma long-term shift is applied; without the shift, six sigma corresponds to about 0.002 defects per million.

DMAIC vs DMADV

DMAIC improves an existing process that is not meeting requirements. DMADV (Design for Six Sigma) creates a new product or process when no baseline exists or redesign is required.

Value-added criteria

A step is value-added only if the customer would pay for it, it physically changes the product or information form, fit, or function, and it is performed correctly the first time.

Eight wastes acronym

TIMWOODS: Transport, Inventory, Motion, Waiting, Overproduction, Overprocessing, Defects, and Skills or talent unused. Overproduction is usually considered the worst waste because it hides the others.

Champion and sponsor role

The champion selects and funds projects, removes organizational barriers, and owns the business result. The sponsor or process owner accepts the improved process at handoff and sustains the gains.

Cost of poor quality

Four categories: internal failure, external failure, appraisal, and prevention. Prevention and appraisal are control costs; internal and external failures are the cost of failing to meet requirements.

Kaizen versus kaikaku

Kaizen is continuous, incremental improvement performed by the people doing the work. Kaikaku is radical, breakthrough redesign of a process, often led by a cross-functional team.

Problem statement content

States the measurable gap between current and required performance, including metric, baseline, time frame, and location, without naming a cause or prescribing a solution.

SIPOC diagram purpose

Suppliers, Inputs, Process, Outputs, Customers. It defines project scope with five to seven high-level process steps and links supplier inputs to customer outputs before detailed mapping.

CTQ tree structure

Translates a verbatim voice-of-customer need into a quality driver and then into a measurable characteristic with a target and tolerance that can be tracked in the process.

Kano model categories

Must-be requirements cause dissatisfaction when absent but do not delight when present. Performance requirements scale linearly with satisfaction. Delighters create disproportionate satisfaction when present but are not expected.

SMART goal statement

Specific, Measurable, Achievable, Relevant, and Time-bound. In a charter it quantifies the improvement against the baseline and states the deadline for achieving it.

Pareto chart interpretation

Ranks causes by frequency or cost in descending bars with a cumulative percentage line. It focuses effort on the few categories that account for most of the problem.

RACI matrix meaning

Responsible performs the work, Accountable is the single decision owner, Consulted provides two-way input, Informed receives one-way updates. Only one person should be Accountable per deliverable.

Repeatability versus reproducibility

Repeatability is variation when one operator measures the same item repeatedly with the same gauge. Reproducibility is variation between different operators using the same gauge on the same items.

Gage R&R acceptance criteria

Percent Gage R&R below 10 percent is acceptable, 10 to 30 percent is conditionally acceptable depending on risk and cost, and above 30 percent is unacceptable and requires gauge improvement.

Measurement discrimination rule

The measurement increment should be no larger than one tenth of the process tolerance or of the total process variation, giving at least ten distinct readings across the range.

Cp and Cpk formulas

Cp equals (USL minus LSL) divided by six sigma. Cpk equals the smaller of (USL minus mean) or (mean minus LSL), each divided by three sigma. Cpk penalizes off-center processes.

Ppk versus Cpk

Cpk uses within-subgroup short-term sigma and reflects potential capability. Ppk uses the overall long-term standard deviation and reflects actual performance. Ppk is normally lower than Cpk.

Attribute agreement kappa

Kappa adjusts agreement for chance. Values above 0.75 indicate good agreement, 0.40 to 0.75 is marginal, and below 0.40 indicates poor agreement among appraisers or against a standard.

Sample size for a mean

n equals (z times s divided by E) squared, where z is the standard normal value for the confidence level, s is the standard deviation estimate, and E is the desired margin of error.

Hypothesis test p-value

The probability of observing data at least as extreme as the sample if the null hypothesis is true. Reject the null hypothesis when the p-value is less than the chosen significance level alpha.

Type I and Type II error

Type I error, alpha, rejects a true null hypothesis, which is producer risk. Type II error, beta, fails to reject a false null hypothesis, which is consumer risk. Power equals one minus beta.

Test choice by data type

Continuous response with categorical factor: two-sample t-test for two levels, ANOVA for three or more. Two categorical variables: chi-square. Two continuous variables: correlation or regression.

ANOVA F statistic

F equals mean square between groups divided by mean square within groups. A large F with p below alpha indicates at least one group mean differs, requiring post-hoc pairwise comparison.

Paired t-test condition

Used when observations are naturally matched, such as before-and-after on the same unit. It tests whether the mean of the within-pair differences equals zero.

Coefficient of determination r-squared

The proportion of variation in the response explained by the regression model. A high r-squared does not prove causation and does not by itself confirm that model assumptions are satisfied.

FMEA risk priority number

RPN equals Severity times Occurrence times Detection, each scored typically from one to ten. Higher RPN means higher priority, but high-severity items are addressed regardless of RPN.

DOE design resolution

Resolution III aliases main effects with two-factor interactions. Resolution IV clears main effects but aliases two-factor interactions with each other. Resolution V clears main effects and two-factor interactions.

Poka-yoke categories

Prevention devices stop an error from occurring, using contact, fixed-value, or motion-step methods. Detection devices signal that an error occurred. Prevention at the source is preferred.

SPC chart selection rule

Continuous data uses X-bar and R for small subgroups, X-bar and s for larger subgroups, and individuals with moving range for single observations. Attribute data uses p, np, c, or u charts.

Control limits versus specifications

Control limits are computed as plus or minus three sigma from process data and describe what the process actually does. Specification limits come from customer requirements and are never drawn on a control chart.