Certified Six Sigma Black Belt Exam Certified Six Sigma Black Belt Measurement System Analysis 1 — Questions and Answers
Question 1: In a Gage R&R study, the 'reproducibility' component specifically measures variation due to:
- The measurement instrument's inherent mechanical limitations
- Different operators measuring the same part with the same gage (Correct answer)
- Environmental factors such as temperature and humidity
- Part-to-part variation within the process
Correct answer: Different operators measuring the same part with the same gage
Reproducibility captures the variation introduced when different appraisers (operators) measure the same part using the same measurement device, reflecting differences in how people use the instrument.
Question 2: A measurement system has a %Gage R&R of 22%. According to AIAG guidelines, how should this system be classified?
- Acceptable — no action required
- Marginally acceptable — may be used based on application importance and cost (Correct answer)
- Unacceptable — the measurement system must be immediately replaced
- Excellent — the system exceeds all quality standards
Correct answer: Marginally acceptable — may be used based on application importance and cost
AIAG guidelines classify %Gage R&R between 10% and 30% as marginal. The decision to accept or reject the system depends on factors such as the importance of the application, cost of repair, and risk of misclassification.
Question 3: Which statistical metric in an MSA describes the ability of a measurement system to detect and distinguish between consecutive part values?
- Bias
- Linearity
- Number of Distinct Categories (ndc) (Correct answer)
- Stability
Correct answer: Number of Distinct Categories (ndc)
The number of distinct categories (ndc) indicates how many groups of parts the measurement system can reliably differentiate. An ndc of 5 or more is required for the system to be considered adequate for process control.
Question 4: During a linearity study, a technician measures reference standards across the operating range and finds that bias increases consistently as the measured value increases. This indicates:
- The measurement system has poor repeatability at high values
- The gage has a non-constant bias across its operating range, indicating a linearity error (Correct answer)
- The gage needs a zero-point calibration adjustment only
- Reproducibility error is present among multiple operators
Correct answer: The gage has a non-constant bias across its operating range, indicating a linearity error
Linearity error occurs when the bias of the measurement system is not consistent across the full operating range — meaning the gage reads more or less accurately depending on the magnitude of the value being measured.
Question 5: A Six Sigma Black Belt conducts a stability study by measuring the same master part daily over 30 days. The control chart shows the measurements drifting upward over time. What does this most likely indicate?
- High reproducibility variation among operators
- The process mean has shifted, inflating the MSA results
- The measurement system is losing calibration or wearing over time (Correct answer)
- Part-to-part variation is too high for the gage to detect
Correct answer: The measurement system is losing calibration or wearing over time
Stability measures how consistent a measurement system's bias is over time. An upward drift on a stability control chart suggests the gage is losing calibration, wearing, or experiencing environmental degradation — all of which compromise measurement accuracy.
Question 6: In an attribute MSA, a Black Belt asks three appraisers to evaluate 50 parts twice each and compares their accept/reject decisions. Which metric is most appropriate for quantifying agreement between appraisers?
- Percent R&R
- Cohen's Kappa statistic (Correct answer)
- Signal-to-noise ratio
- Cp index
Correct answer: Cohen's Kappa statistic
Cohen's Kappa statistic is the standard metric for assessing agreement in attribute (pass/fail) measurement systems. It accounts for the probability of agreement occurring by chance, making it more meaningful than simple percent agreement.
In a Gage R&R study, the 'reproducibility' component specifically measures variation due to: