Sie sind auf Seite 1von 3

Technical Support Document

Creating the Binomial Plot on the Binomial Process Capability Analysis Graph By Hand
To assess quality, technicians pull twenty parts from a lot and count the number of defective parts. They repeat this procedure every two hours, for a total of 30 samples, and then record the following values in column C1 (Defectives): 12 10 15 11 10 14 15 13 11 14 11 13 13 12 11 11 15 10 11 15 12 13 12 13 14 12 13 13 10 12 To calculate the expected defectives in each lot: 1. Tally the column of observations. a. Choose Edit > Command Line Editor. b. Type: TALLY C1; STORE C2 C3. c. Click Submit Commands. d. Label column C2 Observed Defectives and column C3 Frequency. 2. Create an indicator column. a. Label column C4 Indicator. b. The indicator for the first distinct value is 1. In row 1 of C4, type 1. c. The indicator for the second distinct value is equal to the first indicator plus the frequency of the first value: 1 (the first indicator) + 4 (the frequency of the first indicator) = 5. In row 2 of C4, type 5. d. The indicator for the third distinct value is equal to the second indicator plus the frequency of the second value: 5 (the second indicator) + 6 (the frequency of the second indicator) = 11. In row 3 of C4, type 11. e. Continue this pattern for the remaining rows in C4. The indicator for a particular row is equal to the previous indicator plus the frequency of the previous value. Note: The completed worksheet appears below. 3. Calculate the rank of each distinct value based on the frequency and the indicator (i). Note: Rank a. b. c. d.

2 * Indicator Frequency 1
2

Choose Calc > Calculator. In Store result in variable, enter Rank. In Expression, enter ((2 * C4) + C3 - 1) / 2 Click OK.
Rank .375 N .25

4. Calculate the expected CDF for each distinct value. Note: ECDF a. b. c. d.

Choose Calc > Calculator. In Store result in variable, enter ECDF. In Expression, enter (C5-.375)/(N(C1) + .25) Click OK.

Knowledgebase ID 1759: http://www.minitab.com/support/answers/answer.aspx?ID=1759

Page 1

Technical Support Document

Creating the Binomial Plot on the Binomial Process Capability Analysis Graph By Hand
5. Calculate the average defective proportion. a. Choose Calc > Calculator. b. In Store result in variable, enter AverageProportion. c. In Expression, enter mean(C1/20). d. Click OK. 6. Calculate the expected defective values from the inverse binomial CDF. a. Choose Calc > Probability Distributions > Binomial. b. Choose Inverse cumulative probability. c. In Number of trials, enter the sample size: 20. d. In Event probability, enter the value from the AverageProportion column: .6183. e. In Input column, enter ECDF. f. In Optional storage, enter an empty column: C8. g. Label column C8, Expected Defectives. You should now have the following data:

Display the Binomial Plot The expected defectives are on the y-axis. The observed defectives (the raw data) are on the x-axis. 1. 2. 3. 4. 5. 6. 7. Choose Graph > Scatterplot > Simple. Under Y variables, enter Expected Defectives. Under X variables, enter Observed Defectives. Click Labels. Under Title, enter Binomial Plot. Click OK in each dialog box. Choose Editor > Add > Calculated Line. In Y column, enter Observed Defectives. In X column, enter Observed Defectives. Click OK. Double-click on the line. Choose Custom. Under Color, choose Blue. Click OK.

Knowledgebase ID 1759: http://www.minitab.com/support/answers/answer.aspx?ID=1759

Page 2

Technical Support Document

Creating the Binomial Plot on the Binomial Process Capability Analysis Graph By Hand
Binomial Plot
16 15
Expected Defectives

14 13 12 11 10 9 10 11 12 13 Observed Defectives 14 15

Note: To create this graph using Capability Analysis: 1. Choose Stat > Quality Tools > Capability Analysis > Binomial. 2. In Defects, enter Defectives. 3. Under Sample size, after Constant size, enter 20. Click OK.
Binomial Process Capability Analysis of Defectives
P C har t U C L=0.9442 _ P =0.6183
Expected Defectives

Binomial P lot 16 14 12 10 10 12 14 O bser ved Defectives Dist of % Defective S ummary S tats Tar 8 6 4 2 0

1.0
P r opor tion

0.8 0.6 0.4 LC L=0.2925 1 4 7 10 13 16 19 Sample 22 25 28

C umulative % Defective

65.0
% Defective

(using 95.0% confidence) % Defectiv e: Low er C I: U pper C I: Target: P P M Def: Low er C I: U pper C I: P rocess Z: Low er C I: U pper C I: 61.83 57.81 65.74 0.00 618333 578123 657372 -0.3011 -0.4053 -0.1971

62.5 60.0 57.5 55.0 5 10 15 20 Sample 25 30

10 20 30 40 50 60 70

Note: If you run Stat > Quality Tools > Capability Analysis > Binomial with a constant sample size a binomial plot is displayed. If you run it with sample sizes that are not equal, a rate of defectives chart is displayed in place of the binomial plot.
Knowledgebase ID 1759: http://www.minitab.com/support/answers/answer.aspx?ID=1759 Page 3

Das könnte Ihnen auch gefallen