EXIN LSSA Green Belt Practical flashcards

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

DPMO formula

Defects per million opportunities equals defects divided by total opportunities, multiplied by one million. Total opportunities equals units multiplied by opportunities per unit.

Cpk interpretation

Cpk measures how well a stable process fits within specification limits, accounting for centering. Values below 1.0 indicate the process produces defects; 1.33 is a common minimum target.

Rolled throughput yield

RTY multiplies the first-pass yields of every process step. Because yields multiply, a long process with many high-yield steps can still deliver a low overall yield.

Gauge R&R threshold

Gauge repeatability and reproducibility below 10 percent of tolerance or process variation is generally acceptable. Between 10 and 30 percent is conditionally acceptable; above 30 percent requires measurement system action.

Takt time

Takt time equals available production time divided by customer demand rate. It sets the pace each process must meet to satisfy demand without building inventory.

Little's Law

Average work-in-process equals throughput multiplied by average flow time. At constant throughput, reducing WIP proportionally shortens lead time through the process.

Seven wastes

Overproduction, waiting, transport, overprocessing, inventory, motion, and defects, often plus unused talent. Overproduction is considered worst because it generates the other wastes.

SMED principle

Single-minute exchange of die separates setup work into internal activities that require machine stoppage and external activities that can occur while running. Converting internal to external reduces changeover time.

Type I and II errors

Type I error rejects a true null hypothesis and is controlled by alpha. Type II error fails to reject a false null and is related to power, which equals one minus beta.

p-value meaning

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

ANOVA purpose

Analysis of variance tests whether three or more group means are equal. A significant F test shows at least one mean differs; post-hoc tests identify which pairs differ.

R-squared meaning

In regression, R-squared is the proportion of variation in the response explained by the model. Adjusted R-squared penalizes added predictors and is used to compare models.

Chi-square independence test

Compares observed counts in a contingency table with expected counts computed from row and column totals. It tests whether two categorical variables are associated.

Full factorial runs

A two-level full factorial experiment with k factors requires two raised to the k power runs before replication. Adding center points tests for curvature.

FMEA risk priority number

RPN equals severity multiplied by occurrence multiplied by detection. Higher RPN values indicate higher priority for corrective action, though high severity alone can also demand action.

Control chart selection

Use I-MR charts for individual measurements, X-bar and R charts for small subgroups, and p, np, c, or u charts for attribute data depending on sample size constancy.

Control versus specification limits

Control limits are computed from process data, typically at three sigma from the centerline. Specification limits come from customer or design requirements and are not interchangeable with control limits.

Ppk versus Cpk

Ppk uses overall standard deviation and suits initial capability studies. Cpk uses within-subgroup variation and suits ongoing capability assessment of a stable process.

SIPOC diagram

A high-level map showing Suppliers, Inputs, Process, Outputs, and Customers. It defines project scope boundaries before detailed process mapping begins.

SMART goal statement

A project goal that is specific, measurable, achievable, relevant, and time-bound, stating the target metric, baseline, and completion date without naming causes or solutions.

Cost of poor quality

Costs from prevention, appraisal, internal failure, and external failure. External failure costs, such as warranty and recalls, are typically the most damaging to the organization.

Operational definition

A precise statement of what is measured, how, by whom, and the decision rule applied. It makes measurements repeatable across appraisers and shifts.

Five Whys technique

Repeatedly asking why a problem occurs to trace it to a systemic cause. Stop when the cause is actionable and within the team's control, then verify with data.

Fishbone diagram categories

A cause-and-effect diagram commonly organized by Man, Machine, Material, Method, Measurement, and Environment. Teams brainstorm causes in each category before validating them with data.

Theory of constraints

Identify the bottleneck, exploit it, subordinate other processes to its pace, elevate its capacity, then repeat. Improving non-bottleneck steps does not raise system throughput.

OEE calculation

Overall equipment effectiveness equals availability multiplied by performance multiplied by quality. World-class performance is often cited near 85 percent.

Kano model classes

Features are must-be, performance, or delighter. Missing must-be features cause dissatisfaction, while delighters please when present but are not expected.

QFD house of quality

Quality function deployment translates customer requirements into technical characteristics using a relationship matrix, aligning design decisions with what customers value.

Central limit theorem

Sample means tend toward a normal distribution as sample size grows, regardless of population shape. This enables normal-based inference with adequately sized samples.

Normal distribution rule

About 68 percent of values fall within one standard deviation of the mean, 95 percent within two, and 99.7 percent within three.

Kanban pull signal

A card or signal authorizing upstream replenishment based on downstream consumption. Card counts are set from demand during lead time plus a safety buffer.

Heijunka level scheduling

Leveling production volume and product mix over time to smooth demand variation, reduce inventory buffers, and support stable pull systems.