Measurement System Analysis (MSA) Expert (Accredited)

Identify the true value and its substitute, the reference value, measured under the best conditions to estimate bias and assess measurement system variation.

Explore resolution and discrimination in measurement systems, apply the 10 to 1 rule of thumb to match instrument least count to tolerance, and compare vernier calipers with tape.

Explore accuracy and precision in measurement system analysis, distinguishing location variation (accuracy) and width variation (precision), and bias, linearity, stability, repeatability, and reproducibility, with a dartboard analogy.

Explain the difference between accuracy and precision using dart board examples. Identify bias, linearity, and stability as the three accuracy measures to be explored next.

Identify bias as the difference between the observed average and the reference value in a measurement system, caused by calibration errors, wear, and operator or reading mistakes.

Assess how linearity reveals bias variation across an instrument's operating range, distinguish constant bias from true linearity errors, and highlight calibration as a key remedy.

Explore stability in measurement system analysis, which tracks bias drift over time, caused by wear, maintenance, aging, instrument quality, environmental factors, and calibration frequency.

Calibrate your instrument to fix accuracy errors—bias, linearity, and stability—across the full measurement range and with higher frequency for timeliness, then examine precision, the closeness of readings.

Explain reproducibility as variation in the average of measurements across different appraisers using the same gauge, contrasted with repeatability as within-variation by a single operator.

Explore how the normal distribution explains repeatability and reproducibility in measurements, defining mean and standard deviation, and using sigma, z-values, and area under the curve to interpret variability.

Learn how repeatability and reproducibility combine to form Gage R and R, using multiple operators and 25 measurements per piece to illustrate overall GRR and variance concepts.

Analyze how measurement variation affects part acceptance within tolerances, illustrating GRR, repeatability, and reproducibility, and introduce PTR (precision to tolerance ratio) for instrument suitability.

Assess the precision to tolerance ratio (PTR) as a measure of measurement system capability, using six standard deviations, repeatability, and reproducibility to define acceptable ranges (below 10%, 10–30%, above 30%).

Explore the type 1 gage study, a mini grr, evaluating precision and accuracy with a single part and operator through repeated measurements to gauge bias and P to T ratio.

Conduct a type 1 gage study to assess bias and precision using a calibrated gage, a production-representative part, random order, predefined process, and a ten-measurement reference in a controlled environment.

Demonstrates a Type 1 gage study in Minitab using a 0.025 inch reference and 0.0007 tolerance with 50 measurements of coating thickness, highlighting CG, CGK, bias, repeatability, and descriptive statistics.

Apply descriptive statistics from a paint thickness example to perform a type one gage study, calculating mean, standard deviation, min, max, six-sigma range, and tolerance.

Evaluate bias in a type one gage study by comparing the average to the reference value, and determine significance using hypothesis testing and a t-test.

Perform a t-test to determine if the calculated bias is significant, using the null hypothesis that bias is zero, a 95% confidence level, and a p value threshold of 0.05.

Explore process capability concepts, from bias and control limits to specification limits, and learn how CP and CPK evaluate whether a gage measurement system and process meet design requirements.

Learn to compute the potential process capability (CP) and interpret its meaning, compare CP with CPK, and see how a mean shift affects CPK in a film thickness example.

Discover how CPK reveals centering shifts by comparing CPL and CPU, using three sigma; identify when a process is not capable and relate CGK to Type One Gage Study.

Analyze gage capability through CG and CGK in a type one gage study, calculate 20% tolerance spread, and assess measurement system variation against specification limits.

Compute CG and CGK to assess process capability and bias versus precision, using control limits and tolerance, showing CGK accounts for bias and precision whereas CG focuses on dispersion.

In a type one gage study, compute percent variation from repeatability and bias using CG and CGK, with a 20% tolerance. Interpret the Minitab output rather than performing calculations manually.

Explore how the run chart visually represents variation in the measurement system, identifying operator and part variation, and distinguish crossed from nested gage R&R studies.

Create a gage run chart in Minitab from crossed R and R data across parts and three operators. Examine variation within and between parts using two datasets.

Analyze how a gage run chart reveals variation within an operator, between operators, and part-to-part variation, guiding investigations into operator methods and gauge reading.

Explore within and between variation in measurements using a run chart, showing how same operator readings differ and how different operators create between-operator variation, with bias and linearity coming next.

Explore bias and linearity in gage measurement, using five parts across the process range, a controlled reference from ten measurements, and random-order operator data analyzed in Minitab.

Develop a gage bias and linearity study in minitab using part numbers, master values, and measurements, and interpret the bias and linearity reports with process variation considerations.

Interpretation part 1 teaches assessing bias and linearity in the Gage study by analyzing the slope and p values against a 0.05 threshold, identifying significant linearity and bias.

Assess gage linearity and bias using p values, 95% confidence intervals, the regression line Y = A + BX, and R square 71.4% to estimate bias across reference values.

Explore Gage R&R concepts by examining repeatability and reproducibility across three methods: range, average and range, and ANOVA, revealing operator effects and interactions.

The range method gives a quick GRR estimate by averaging the ranges from two appraisers across five parts and dividing by D2 star.

Explore the average and range (X bar R) method for GRR analysis, compare it with ANOVA, and learn to compute repeatability, reproducibility, and two-way interactions using Minitab and Excel.

Learn to perform a gage r&r study in Minitab using average and range, with X-bar R and ANOVA methods, and interpret the output including repeatability and reproducibility.

Explore Gage R&R analysis with the average and range (x-bar R) method, including repeatability, reproducibility, and total Gage R&R calculations, and compare with ANOVA approach through practical examples.

Interpret anova results for gage R&R alongside X-bar R, using run charts and box-and-whisker plots to compare operators A, B, and C, noting operator C measures lower.

Interpret gage R and R ANOVA results by comparing measurement by part with the run chart, highlighting nine readings per part and part-to-part variation.

Interpret the part–operator interaction chart alongside the run chart to confirm no interaction, as operator lines A, B, and C remain parallel, while crossovers would indicate dependency on part.

Assess measurement system performance with gage R and R charts, noting operator variation and interaction. Compare X bar chart and R chart to interpret measurement spread and control limits.

Explore components of variation in a gage R and R study, separating part-to-part variation from measurement system variation, and interpret ANOVA results for significant part and operator effects.

Explore the anova table in measurement system analysis, explaining sum of squares, degrees of freedom, f values, and p values for parts and operators, linking to Gage R and R.

Calculate the number of distinct categories (NDC) by dividing PV (part variation) by GRR (Gage R and R), using 97% confidence intervals and truncation rules.

Assess the Gage R and R value to determine if high GRR comes from repeatability or reproducibility, then apply maintenance, training, or clearer gage marking to improve the measurement system.

Introduces nested gage r&r design as an alternative to crossed studies. Each piece is measured by one operator due to destructive testing, totaling 60 measurements across 3 operators in Minitab.

Explore nested gage r&r for destructive tests, analyzing two readings per part, with Minitab ANOVA, variance components, and box and scatter plots by operator and part.

Explore box and whisker and scatter plots in measurement system analysis, revealing operator differences and part-to-part and measurement variation. See how these plots complement nested gage r and r studies.

Interpret nested gage r&r results with operator and part plots, compare to crossed gage r&r, and identify variation sources, including part 15, while noting x bar and range charts.

Analyze the r chart and x bar chart in the nested gage r and r study, highlighting operator variation, sample range, and how average range sets control limits.

Explore how nested gage r&r splits variation into part-to-part and gage r&r components, note that reproducibility is zero, and see how tolerance or historical standard deviation add bars.

Summarize the outputs of the nested gage r and r study, including the nested anova, variance components, and gage evaluation to interpret p-values, operator interaction, total variation, and aiag guidelines.

Learn how attribute agreement analysis handles nominal and ordinal data to assess measurement system quality, and identify three error types—inconsistency with themselves, with one another, and with the standard.

Demonstrate attribute agreement analysis in Minitab, four inspectors rate items 1–5, compare to the standard, and use Fleiss' kappa and candles coefficient of concordance.

Evaluate consistency across appraisers by analyzing agreement within themselves, calculate appraiser-specific consistency percentages and 95% confidence intervals, and interpret plots to choose reliable operators.

Evaluate appraisers' consistency and correctness against the standard in attribute agreement analysis, showing Amanda as highly consistent and correct with a 95% confidence interval.

Evaluate how well appraisers agree with each other in attribute agreement analysis, noting that 74% of samples show exact matching and identifying inconsistencies where a single rating differs.

Analyze the fourth assessment in attribute agreement analysis, comparing all operators against the standard, showing 74% agreement (50 items, 37 matched) with a 60–85% confidence interval and the kappa value.

Compare consistency and correctness plots from attribute agreement analysis to compare Amanda and Eric, and introduce kappa value tables for deeper analysis.

Define kappa value as the agreement between appraisers on nominal or ordinal data, comparing observed and chance agreement. Use cross tables and examples to interpret values from -1 to 1.

Compare Cohen's kappa and Fleiss kappa to measure agreement among two or more appraisers, and use Kendall's coefficient for ordinal data to assess correlation.

Examine agreement within appraisers using Fleiss CAPA and Kendall's coefficient of concordance for ordinal ratings, highlighting Amanda's perfect consistency and Eric's potential rating inconsistencies.

Evaluate the correctness and agreement with the standard in measurement system analysis, using kappa values and Kandall's correlation to identify strong agreement and investigate anomalies.

Assess Kappa value and Kandall's coefficient of concordance for agreement among appraisers, compare all operators to the standard, and note a dip at 3 with strong overall concordance.