ASQ Six Sigma Black Belt 4 — Questions and Answers
Question 1: In a full factorial 2³ experiment with 3 factors at 2 levels each, how many experimental runs are required (without replication)?
- 6
- 8 (Correct answer)
- 9
- 12
Correct answer: 8
A 2³ full factorial design requires 2³ = 8 experimental runs to test all combinations of 3 factors at 2 levels.
Question 2: Which FMEA severity rating criterion would typically be assigned to a failure mode that could cause injury without warning?
- Rating 1-3 (minor)
- Rating 4-6 (moderate)
- Rating 7-8 (high)
- Rating 9-10 (very high/hazardous) (Correct answer)
Correct answer: Rating 9-10 (very high/hazardous)
A severity rating of 9-10 is reserved for failure modes that involve safety hazards or regulatory non-compliance, especially those occurring without warning.
Question 3: What is the key distinction between 'common cause' and 'special cause' variation?
- Common cause variation is larger in magnitude than special cause
- Special cause variation is inherent to the process design; common cause is from external events
- Common cause variation is random and systemic; special cause variation is assignable to a specific factor (Correct answer)
- Common causes can only be removed by operators; special causes require management action
Correct answer: Common cause variation is random and systemic; special cause variation is assignable to a specific factor
Common cause (random) variation is inherent to the process and requires system-level changes to reduce, while special cause variation stems from identifiable, assignable factors.
Question 4: A Black Belt wants to test whether a process median has changed after an improvement. The data is non-normal and small-sample. Which test is most appropriate?
- One-sample t-test
- Wilcoxon signed-rank test (Correct answer)
- Chi-square test
- Anderson-Darling test
Correct answer: Wilcoxon signed-rank test
The Wilcoxon signed-rank test is the non-parametric equivalent of the one-sample t-test and is used when data is non-normal or the sample is small.
Question 5: In a multiple linear regression model, what does multicollinearity refer to?
- The residuals are not normally distributed
- High correlation among two or more predictor variables (Correct answer)
- The response variable has a non-linear relationship with predictors
- Unequal variance of residuals across fitted values
Correct answer: High correlation among two or more predictor variables
Multicollinearity occurs when predictor variables are highly correlated with each other, inflating standard errors and making individual coefficient estimates unreliable.
Question 6: Which Six Sigma concept describes the practice of verifying that gains from a project are sustained 6-12 months after project closure?
- Benchmarking
- Control plan auditing
- Sustaining the gains / benefit realization (Correct answer)
- Kaizen follow-up
Correct answer: Sustaining the gains / benefit realization
Sustaining the gains (benefit realization) involves monitoring key metrics and control plan adherence long after the project is handed off to ensure improvements hold.
Question 7: When using a Pugh Concept Selection Matrix, what serves as the baseline against which all alternatives are compared?
- The highest-scoring concept
- The current process or datum (Correct answer)
- The customer's ideal product
- The lowest-cost solution
Correct answer: The current process or datum
In a Pugh matrix, the current design or a reference concept (datum) is used as the baseline; alternatives are rated as better (+), worse (-), or same (S) relative to it.
ASQ Six Sigma Black Belt 4
In a full factorial 2³ experiment with 3 factors at 2 levels each, how many experimental runs are required (without replication)?