Test your basic knowledge |

Six Sigma

Subjects : certifications, six-sigma, business-skills

Instructions:

Answer 50 questions in 15 minutes.
If you are not ready to take this test, you can study here.
Match each statement with the correct term.
Don't refresh. All questions and answers are randomly picked and ordered every time you load a test.

This is a study tool. The 3 wrong answers for each question are randomly chosen from answers to other questions. So, you might find at times the answers obvious, but you will see it re-enforces your understanding as you take the test each time.

1. The data points on an SPC p-chart of defective percent plot below the mean. Is this good or bad?

2. Residual(eij)
Controlling quality at the source.
= yij - y(bar)i --> value minus sum of that treatment (of row)
The target number of standard deviations from the mean for specifications.
10 if Almost impossible to detect

3. If a three-stage process has 90% yields at each stage - what is the overall yield?
0.9^3 or 72.9%
Incoming materials inspection - inspection and testing - maintaining test equipment - materials/services consumed
Before production starts
The gap between what is promised and what is delivered.

4. What is DPMO?
The ability of a process to consistently meet customer specifications.
Fitness for use.
The gap between what management thinks customers want and what customers really want.
Defects Per Million Opportunities

5. What are the Five Dimensions of Service Quality?
To mentor/coach (and sometimes train) black belts.
Controlling quality at the source.
Reliability - Assurance - Responsiveness - Tangibles - Empathy
Shows the temporal flow of activities in a project network.

6. Defect location check sheets
Rapid improvement process.
Variable.
Upper control limit.
Actual drawings - layouts - maps - etc which show where a defect occurs

7. Regression
No -- the variation also has to be random and "expected."
It usually isn't possible to hold all other variables constant - There is no way to account for the effect of joint variation of independent variables - such as interaction - There is no way to account for experimental error - including measurement v
Represents the behavior of a process
Fishbone - cause-and-effect diagram.

8. What are the four perspectives of the Balanced Scorecard?
=sqrt(Σ(x-x-bar)^2)/(n-1))
Hold all input variables constant except one. Observe the response as you vary the single input.
Financial - customer - internal process - innovation and learning.
The degree to which a product meets pre-established standards.

9. ANOVA

10. What does the abbreviation LCL stand for?
Lower control limit.
Prevention - appraisal - internal failure - external failure.
Technique used to relate through a model - one or more independent variables and a dependent variable (response)
Is the consequence of the failure.

11. When either attribute or variable measures could be used for SPC - why might attribute measurement be preferred?
99.73%
Fishbone - cause-and-effect diagram.
To manage the Six Sigma project.
Lower skill requirements - faster - less chance for error.

12. Test for constant variance
Technical quality is the "what" of the service; functional quality is the "how."
Plot the residuals vs. the fitted values (mean of each group - y(hat) = y (bar)i ) - Assumption of ____________not rejected if the dispersion of every group is approximately the same.
Mistake-proofing.
Large enough to find two of the attribute - on average.

13. What happens to control limits on an SPC chart when sample size is increased?
Yes - when there are opportunities to improve - when contribution margin is adequate - when achieving market share is important.
Prediction and estimation. based on an unknown x value - estimation is based on a known x value
Producer risk.
They move closer to the center line.

14. Process occurance ranking
Rank of 10 if there is >= 100 per 1000 piece Rank of 1 if there is <=.01 per 1000 pieces
Prevention - appraisal - internal failure - external failure.
An interaction is defined as a dependence relationship between the response and the levels of two or more variables
Shows the temporal flow of activities in a project network.

15. What are the 5 Ss?
Sort - Set in Order - Shine - Standardize - Sustain
Mistake-proofing.
68%
A scatterplot.

16. If you can only collect categorical data - what type of SPC charting can you do?
Attribute.
Deming.
= S x O x D S--> Severity O--> Occurrence D--> Detection Higher the number the worse it is
Hold all input variables constant except one. Observe the response as you vary the single input.

17. In the Service Quality Gap Model - what is Gap 2?
No -- the variation also has to be random and "expected."
The gap between what management thinks customers want and the process specifications.
Vary two or more variables simultaneously - Multiple measurements are obtained under the same experimental conditions
The Japanese national quality award.

18. What is Type 1 Error?
Actual drawings - layouts - maps - etc which show where a defect occurs
A random variable that summarizes the information from the sample(s).
Conformance to specifications.
Concluding there has been an effect/change when there has not.

19. What do you do when a process is out of control?
Look to your customer - not to other firms. "You can always find someone shorter - fatter and balder!"
How likely it is to have an F statistic of that value if there really is no relationship between the dependent variable and the independent variable(s).
Look for the assignable cause of non-random variation.
68%

20. In a regression - what does the standard error of the estimate tell you?
Rank of 10 if there is >= 100 per 1000 piece Rank of 1 if there is <=.01 per 1000 pieces
How much variance you expect around the prediction; two SEs gives you a 95% confidence interval.
Observed variation in response is caused by the input
Complaint adjustment - returned material - warranty charges - allowances - loss of future business.

21. What is Jidoka?
What the value of the dependent variable is when the independent variable is zero.
How likely it is to have an F statistic of that value if there really is no relationship between the dependent variable and the independent variable(s).
consists of plotting the residuals vs. the order of each experiment - If - as in the case of the constant variance a random horizontal band is obtained - the ______________ of the residuals is not rejected. (should be random above a below the line i
Controlling quality at the source.

22. What is FMEA?
Failure Modes and Effects Analysis
Observed variation in response is caused by the input
The process of ranking opportunities to determine which of many potential opportunities should be pursued first.
= S x O x D S--> Severity O--> Occurrence D--> Detection Higher the number the worse it is

23. Advantages of DOE
1) Obtain Residuals 2) Fill out table --> y coordinates: F = 100 (i-.5)/N x coordinates: Ordered Residual 3) Plot on Normal Probability Paper
1.Interactions can be detected and measured 2.Each value does the work of several values 3.Experimental error is quantified and used to determine the confidence the experimenter has in the conclusions
Actual drawings - layouts - maps - etc which show where a defect occurs
Shows the temporal flow of activities in a project network.

24. What is a dashboard?
3.4 (assuming 1.5 sigma shift in the mean)
Is the way in which the failure is manifested.
99.73%
A tool that shows performance along key dimensions all at once.

25. Defect check sheets
Different types of defects are listed (used for Pareto chart)
Reliability - Assurance - Responsiveness - Tangibles - Empathy
Brand image.
The gap between specifications and how service is performed.

26. Pareto Analysis
Rapid improvement process.
The process of ranking opportunities to determine which of many potential opportunities should be pursued first.
Causes that lead to a particular effect.
1) Obtain Residuals 2) Fill out table --> y coordinates: F = 100 (i-.5)/N x coordinates: Ordered Residual 3) Plot on Normal Probability Paper

27. Define product reliability.
The ability of a process to consistently meet customer specifications.
Mean time to failure.
It looks to minimize the probability of a failure - or to minimize its effects
Mean time to replacement.

28. What are the four regression assumptions?
Technique used to relate through a model - one or more independent variables and a dependent variable (response)
Linearity - normality - homoscedasticity - independence.
Initiated before or at design concept finalization
It looks to minimize the probability of a failure - or to minimize its effects

29. TWO-DIMENTIONAL SCATTER PLOT
No; if the process is centered - Cp will be the same; if not - Cpk is what you need to determine capabilty.
Inventory = throughput x flow time.
A simple graph between two variables - visualize the type - degree of strength and shape of the relationship between two variables
Walter A. Shewhart

30. What does a project prioritization matrix do?
Assigns scores to weighted criteria for each project under consideration.
The cycle time required to meet demand.
= yij - y(bar)i --> value minus sum of that treatment (of row)
Fractional factorial designs (orthogonal arrays)

31. Define performance quality.
Fractional factorial designs (orthogonal arrays)
The primary operating characteristics of a product.
Lower skill requirements - faster - less chance for error.
Used to create frequency distribution tally sheets

32. Manual Test for Nomality
95%
Waste.
1) Obtain Residuals 2) Fill out table --> y coordinates: F = 100 (i-.5)/N x coordinates: Ordered Residual 3) Plot on Normal Probability Paper
They move closer to the center line.

33. What does the abbreviation UCL stand for?
Define - measure - analyze - improve - control.
No -- if it is not the same - it can only be worse!
Upper control limit.
Causes that lead to a particular effect.

34. What is the 5S model?
Different types of defects are listed (used for Pareto chart)
System to reduce waste and optimize productivity through maintaining an orderly workplace and using visual cues to achieve more consistent operational results.
Concluding there has not been an effect/change when there has.
Kaoru Ishikawa.

35. What is the formula for the standard deviation of a proportion?
No; if the process is centered - Cp will be the same; if not - Cpk is what you need to determine capabilty.
= y = Beta(0) + Beta(1)X + epsilon y=dependent variable (response) x=independent variable (predictor of y) - epsilon=error component - Beta(not)=intersection. If data include zero - it represents the mean of the distribution of y when - x=0. It does
=sqrt((p*(1-p)/n)
Brand image.

36. What is a histogram?
Fishbone - cause-and-effect diagram.
Hold all input variables constant except one. Observe the response as you vary the single input.
Kaoru Ishikawa.
A chart that breaks the range of data into equal intervals and then shows the count of occurrences in each interval.

37. You have been plotting sample means on an x-bar chart and all points indicate normal - expected variation. Is the process in control?

38. List two appraisal costs.
Incoming materials inspection - inspection and testing - maintaining test equipment - materials/services consumed
No -- if it is not the same - it can only be worse!
Another way to assess the significance of the analyzed factors is through a graphic procedure developed by C. Daniel when only one replicate (a single observation) was obtained:
1.Interactions can be detected and measured 2.Each value does the work of several values 3.Experimental error is quantified and used to determine the confidence the experimenter has in the conclusions

39. When Crosby said - "Quality is free -" what dimension of quality was he referring to?
Conformance to specifications.
To mentor/coach (and sometimes train) black belts.
1) Identify the Range (Max Value - Min Value) 2) Determine # of bins 3) Determine the Width of each bin --> (Range / # Bins) 4) Put values in correct Bin
Deming.

40. Define yield.
The ratio of outputs to inputs; 1-scrap.
An experiment where one or more variables believed to have an effect on an experimental outcome are identified and manipulated according to a plan
= S x O x D S--> Severity O--> Occurrence D--> Detection Higher the number the worse it is
How likely it is to have an F statistic of that value if there really is no relationship between the dependent variable and the independent variable(s).

41. Presence of interaction effect
If the Response vs. Levels graph if the response lines are not parallel (crossing)
No -- if it is not the same - it can only be worse!
No -- the variation also has to be random and "expected."
Plot the residuals vs. the fitted values (mean of each group - y(hat) = y (bar)i ) - Assumption of ____________not rejected if the dispersion of every group is approximately the same.

42. Stratified defect check sheets
Hold all input variables constant except one. Observe the response as you vary the single input.
Upper control limit.
Stratify a particular defect type according to logical criteria
A tool that shows the physical flows through a space.

43. Affinity Diagrams
Data reduction to put a large number of qualitative inputs into a smaller number of major dimensions
Hold all input variables constant except one. Observe the response as you vary the single input.
Rank of 10 if there is >= 100 per 1000 piece Rank of 1 if there is <=.01 per 1000 pieces
A chart that breaks the range of data into equal intervals and then shows the count of occurrences in each interval.

44. Define serviceability.
Speed - courtesy - competence - and ease of repair.
Define - measure - analyze - improve - control.
What the value of the dependent variable is when the independent variable is zero.
Look for the assignable cause of non-random variation.

45. What is CTQ?
Outside.
Lean is waste reduction - Six Sigma is variation reduction.
Another way to assess the significance of the analyzed factors is through a graphic procedure developed by C. Daniel when only one replicate (a single observation) was obtained:
Critical to quality.

46. DOE
An experiment where one or more variables believed to have an effect on an experimental outcome are identified and manipulated according to a plan
It looks to minimize the probability of a failure - or to minimize its effects
Technique used to relate through a model - one or more independent variables and a dependent variable (response)
Not necessarily.

47. Simple Regression Analysis (one factor regression model)
Represents the behavior of a process
Concluding there has not been an effect/change when there has.
= y = Beta(0) + Beta(1)X + epsilon y=dependent variable (response) x=independent variable (predictor of y) - epsilon=error component - Beta(not)=intersection. If data include zero - it represents the mean of the distribution of y when - x=0. It does
Used to create frequency distribution tally sheets

48. Does it ever make sense to accept a job when a process is incapable?
Shifts - Trends - Repeating patterns - Correlation with known events
Yes - when there are opportunities to improve - when contribution margin is adequate - when achieving market share is important.
Because the mean and standard deviation are independent of each other.
Hold all input variables constant except one. Observe the response as you vary the single input.

49. If your process is capable at two sigma - what percentage of the output will you have to inspect to find defectives?
To monitor a process when measurement by attribute is used.
Observed variation in response is caused by the input
100%
1) Obtain Residuals 2) Fill out table --> y coordinates: F = 100 (i-.5)/N x coordinates: Ordered Residual 3) Plot on Normal Probability Paper

50. Disadvantages of DOE