Certified Six Sigma Black Belt Design of Experiments (DOE) Questions and Answers Flashcards
6 cards from real Certified Six Sigma Black Belt Exam practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 6 Certified Six Sigma Black Belt Design of Experiments (DOE) Questions and Answers flashcards as text
In a 2^k factorial design, what does the term 'confounding' refer to?
Answer: Aliasing a higher-order interaction with a block effect
Confounding deliberately aliases a high-order interaction effect with the block effect so the experiment can be run in fewer blocks.
A Six Sigma Black Belt runs a 2^4 full factorial with two replicates. How many total experimental runs are required?
Answer: 32
A 2^4 design has 16 runs, and two replicates yield 16 × 2 = 32 total runs.
Which of the following is the primary advantage of using a fractional factorial design over a full factorial?
Answer: It reduces the number of runs while still estimating main effects
Fractional factorials sacrifice information on higher-order interactions to significantly reduce the total number of experimental runs.
What is the resolution of a 2^(5-2) fractional factorial design with generators I=ABCD and I=BCE?
Answer: Resolution III
The generators produce a word-length of three (BCE), making it a Resolution III design where main effects are aliased with two-factor interactions.
In a Response Surface Methodology (RSM) study, what is the purpose of adding axial (star) points to a factorial design?
Answer: To allow estimation of quadratic (second-order) effects
Axial points extend the factorial design so that curved, second-order relationships between factors and the response can be modeled.
A Black Belt notices that the residual plot from a DOE shows a funnel shape with increasing spread. What assumption is most likely violated?
Answer: Constant variance (homoscedasticity)
A funnel-shaped residual plot indicates that variance increases with the predicted value, violating the constant variance assumption.