A team of engineers conducted an experiment to improve the water
clearing ability of windshield wipers. Which three of the following are
noise factors in this experiment?
1: Rain, temperature, glass surface condition
Which one of the following is the LEAST desirable response variable from an experimental design point
of view?
1: Time to Failure
2: Go/no-go
A team of engineers studying three factors affecting engine performance conducted testing as shown in
the following table. Which one approach to experimentation did they use?
1: Fractional factorial experimentation
2: OFAT
3: Full factorial
A single factor experiment is conducted to investigate whether or not three different proposed
adhesives for mounting windshields affect the pull off force per inch. The experiment is replicated six
times for each of the three adhesives. The ANOVA table shown below summarizes the analysis of
variance on the data. Which one of the following is the most appropriate conclusion?
1: There is no evidence
A fractional factorial experiment is conducted. The residuals for that experiment versus run order are
plotted below. Which one of the following statements is most consistent with the plot?
1: Sudden change in conditions that biased later measurements HIGH
Which one of the following is a TRUE statement about One-Factor-at-a-Time type testing?
1: Orthogonal arrays are not part
Which one of the following is the correct interpretation for the plot and data shown above?
1: Strong interaction LIKELY between A and B
The main effects plots and interaction plots are shown for the terms believed to be significant. For the
factor A select the optimal setting to maximize the response.
1: High (+1)
The main effects plots and interaction plots are shown for the terms believed to be significant. For the
factor B select the optimal setting to maximize the response.
1: Does not affect response
2: Low (-1)
3: High (+1)
The main effects plots and interaction plots are shown for the terms believed to be significant. For the
factor C select the optimal setting to maximize the response.
1: Low (-1)
2: High (+1)
3: Does not affect response
Which one of the following specifies what is done with factors that are found to have minimal or no
effect on the response, that is, they are non-significant (inactive) in an experiment?
1: Productivity, efficiency, cost effectiveness
Which two of the following are reasons why it would not be feasible to run a full factorial experiment for
seven factors at two levels in most production environments?
1: Too many runs; too much time and effort
Which four of the following are benefits of using fractional factorials in experimentation?
1: All except “Increases confidence over full factorial runs…”
A team conducted an experiment on 15 two-level factors in 32 runs using a 2(15-10) fractional factorial
experiment. One team member remarked with despair, "We've made 32 runs and not one of them
meets our minimum desired response." The team statistician observed that with the design, assuming
all factors are significant, the probability that their 32-run experiment happened to include the optimal
design is very small. How small is this probability? Select one response.
1: 1/1024
A two-level fractional factorial experiment is performed to study five factors (A-E) and select two-factor
interactions. Main effects plots and interaction plots are as shown. Based on a normal effects plot, it is
concluded that A, B, C and the AC interaction are dominant (i.e., only those terms will be considered in
the predictive model). The average response for the entire experiment was 50, and the A, B, C and AC
effects were 10, 8, -6 and -12, respectively.
1: 42
2: 12
3: 68
Take 2
