Lean Six Sigma Green Belt Certification Lean Six Sigma Green Belt DMAIC: Analyze Phase Techniques 1 — Questions and Answers
Question 1: Which statistical technique is used in the Analyze phase to determine how much of the total variation in a process is attributable to measurement system error?
- Gauge R&R study (Correct answer)
- Control chart analysis
- Regression analysis
- Pareto chart
Correct answer: Gauge R&R study
Gauge Repeatability and Reproducibility (Gauge R&R) quantifies measurement system variation by separating it from actual process variation, ensuring that observed defects reflect true process issues rather than measurement error.
Question 2: In the Analyze phase, a process capability index (Cpk) of 0.85 indicates that the process is:
- Highly capable and centered within specification limits
- Not capable, as it falls below the minimum acceptable threshold of 1.0 (Correct answer)
- Capable but requires monitoring
- Perfectly centered with no defects
Correct answer: Not capable, as it falls below the minimum acceptable threshold of 1.0
A Cpk below 1.0 means the process cannot consistently produce output within specification limits. The industry standard minimum for a capable process is Cpk ≥ 1.0, with 1.33 or higher considered fully capable.
Question 3: What is the primary purpose of a multi-vari chart in the Analyze phase?
- To establish control limits for a process
- To prioritize defects by frequency and cumulative impact
- To visually identify and compare sources of variation such as positional, cyclical, and temporal (Correct answer)
- To calculate the sigma level of a process
Correct answer: To visually identify and compare sources of variation such as positional, cyclical, and temporal
A multi-vari chart displays variation from multiple sources simultaneously, allowing teams to identify which family of variation (positional, cyclical, or temporal) is the dominant contributor to process inconsistency.
Question 4: During an Analyze phase regression study, an R² value of 0.92 means that:
- The model has a 92% chance of being correct
- 92% of the variation in the output variable is explained by the input variable(s) in the model (Correct answer)
- There is a 92% probability the null hypothesis is false
- 8% of data points fall outside the confidence interval
Correct answer: 92% of the variation in the output variable is explained by the input variable(s) in the model
R² (coefficient of determination) represents the proportion of variance in the dependent variable that is predictable from the independent variable(s). An R² of 0.92 indicates a strong model fit.
Question 5: Which of the following best describes the concept of 'stratification' as used in the Analyze phase?
- Separating data into distinct groups or categories to identify patterns or differences between subgroups (Correct answer)
- Ranking defect types from most to least frequent
- Plotting data points over time to detect trends
- Calculating the mean and standard deviation of a dataset
Correct answer: Separating data into distinct groups or categories to identify patterns or differences between subgroups
Stratification involves breaking data into subgroups (e.g., by shift, machine, operator, or material lot) to reveal hidden patterns or differences that are masked when all data is combined, helping isolate root causes.
Question 6: In a 5 Whys analysis performed during the Analyze phase, what signals that you have likely reached the root cause?
- The answer points to a systemic issue such as a missing process, policy, or standard rather than a symptom (Correct answer)
- You have asked exactly five questions regardless of the answer
- The answer involves a person or operator error
- The defect frequency drops below 10% at that level
Correct answer: The answer points to a systemic issue such as a missing process, policy, or standard rather than a symptom
The goal of 5 Whys is to drill past symptoms to a systemic root cause — typically a missing standard, inadequate training, flawed process design, or absent control. Reaching a correctable system-level cause signals you have found the true root.
Which statistical technique is used in the Analyze phase to determine how much of the total variation in a process is attributable to measurement system error?