Exercise 8.5.1

Model Answers to 08 ANOVA in Six Sigma Statistics using Minitab 17, Green Belt Edition.

Exercise 8.5.1    One Way ANOVA

Evelee golf balls monitors the golf ball weights produced by it’s 5 moulding machines. From capability tests they know that they are well below the max weight of 45.93 grams. They have hired you as a consultant and want to know if any of the moulding machines are producing golf balls at different mean weights. They feel that smallest difference that they would want to be able to detect between machines is 0.05g.

Analyse the data in File 08 ANOVA.xlsx worksheet Ex 8.5.1 and answer the questions shown below. The data was collected randomly and is recorded in time order.

  1. Is there a difference in populations between the golf ball weights produced by any of the moulding machines?
  2. Can the machines be grouped ?
  3. Have the requirements of the test that you have used been met?
  4. What was the Power of the test when you want to detect a difference of 0.05?
  5. Are there any issues associated with this level of  Power ?
  6. Does the Report Card generate any warnings?






Set-Up

Analysis

1. Click Assistant<< Hypothesis Tests

2. Click on One-Way ANOVA.





2.  Complete the menu as shown and click OK to execute the procedure.


Starting on the top left of the Summary Report, we can conclude that the sample data from the moulding machines came from populations where at least one pair of populations had differing mean weights.







The Mean Comparison Chart shows us how the moulding machines could be grouped.




The Grouping Information Table confirms that in terms of mean weights produced Moulds 1,2 & 3 can be considered identical and in one group. Also, Moulds 4& 5 can be considered identical and in a different group to Mould 1,2 & 3. It’s a bit confusing but Moulds 1,2,&3 are groups 3,4 &5.





On the Diagnostic Report, the control chart shows that there were no usual data points that could affect the validity of the test.

The histograms and control charts shows the distributions were not bi-modal.




 The Power Report shows that a Power of  between 60.6 and 89.7% % was achieved. There are no issues with this level of Power as a difference was detected.





The Report Card did not show any warnings.