Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 12

Six Sigma

Green Belt

Pareto Analysis


-6 -4 -2 0 2 4 6
Sigma Quality Management

1
Six Sigma
Vilfredo Frederico Pareto & Joseph Juran Green Belt

Our Heroes!
2
Six Sigma
Pareto Principle Green Belt

 A large fraction of power plant failures are due to boiler tube


problems,

 Seventy percent of assembly defects on irrigation sprinklers are due


to problems inserting two components,

 Over 75% of a printing company’s sales are to just three customers.

 Wasted days in a hospital are due mainly to problems transferring


patients to Skilled Nursing Facilities,

 Delays receiving spare parts are most often associated with one
vendor,

 Over 90% of Mutual Fund transaction errors fall into four categories,

 Sixty-five percent of employee injuries are back strains and sprains,

3
Six Sigma
Pareto Analysis Process Green Belt

1. What’s the Effect or Problem?


2. How Should the Effect be Measured?
3. How Can the Effect be Stratified?
• What
• When
• Where
• Who
4. What Does the Data Tell You?

 “Zero-Type” Problems

4
Six Sigma
The Pareto Chart Green Belt

Pareto Chart - Typographical Errors


Total Count = 135 Percent
Frequency
100
126 98%
87% 93% 90
112
79% 80
98
70
64%
84 60
70 50
56 40
48 - 36%
42 38 30

28 20
20
14 12 10
8 6 3
0 0
Punctuation Wrong Word Missed Word Wrong Font
Misspelling Duplicate Word Missed Sentence

5
Six Sigma
Multiple Stratifications Green Belt
Pareto Chart - Typographical Errors

Total Count = 135 Percent


Frequency
100
126 98%
87% 93% 90
112 79% 80
98
70
64%
84 60
70 50
56 40
48 - 36%
42 38 30

28 20
20
14 12 10
8 6 3
0 0
Punctuation Wrong Word Missed Word Wrong Font
Misspelling Duplicate Word Missed Sentence

Pareto Chart - Punctuation Errors


Frequency Total Count = 48 Percent
100
45
90
40 77%
83% 80
35 31 - 65% 70
30 60
25 50
20 40
15 30
10 8 20
6
5 3 10
0 0
No Comma before AND No Quotes
No Period Other
6
Six Sigma
Contingency Table Analysis Green Belt

2 x 2 Contingency Table - 2 Rows, 2 Columns


Flu Vaccine Type
Shanghai Malaysian Total
Contracted Flu 673 816 1489
Did Not Contract Flu 2880 2194 5074
Total 3553 3010 6563

2 x 5 Contingency Table - 2 Rows, 5 Columns


Hospital Units
2E 2W 3N 3W 4E Total
Infections 5 3 6 4 7 25
No Infection 124 212 186 134 303 959
Total 129 215 192 138 310 984

4 x 3 Contingency Table - 4 Rows, 3 Columns


Post-Test Performance
Bad Average Good Total
0 QI Courses 20 20 10 50
1 QI Courses 10 30 15 55
2 QI Courses 5 30 15 50
3 QI Courses 5 20 20 45
Total 40 100 60 200

7
Six Sigma
Notation Green Belt

n COLUMNS
A1 A2 ... An Total
B1 X11 X12 ... X1n X1.
m ROWS

B2 X21 X22 ... X2n X2.


B3 X31 X32 ... X3n X3.
. . . ... . .
Bm Xm1 Xm2 ... Xmn Xm.
Total X.1 X.2 ... X.n X..

8
Six Sigma
Contingency Analysis - Steps Green Belt

1. Establish the Hypothesis:


• Null Hypothesis (Ho) - There is no relationship (i.e. independence
exists) between Attributes A and B.
• Alternative Hypothesis (Ha) - There is a relationship (i.e.
dependence exists) between Attributes A and B.

2. Choose the Significance Level of the Test ().

3. Plan the Test:


k
(Oi  Ei ) 2
a) The Test Statistic is:  
2
0
i 1 Ei
where :
k number of cells in the table (m  n)
Oi  observed count for cell i
Ei  expected count for cell i
(assuming H o is true)
9
Six Sigma
Contingency Analysis - Steps Green Belt

3. Plan the Test:


b) Determine the Rejection Region: Appendix A provides a table of
the 2 distribution. Find the table value for (m - 1)(n - 1) degrees of
freedom at the level of significance.

For example, for a 4 row, 3 column contingency table, m = 4, n = 3


and the 2 value for 6 {(4 - 1) x (3 - 1) = 3 x 2 = 6} degrees of freedom
at the 0.05 level of significance would be obtained from the look up
table (this value is 12.59). The flu example is a 2 x 2 table, which
therefore has {(2 - 1) x (2 - 1)} = 1 degree of freedom. From the
table, then, the critical value (at the 0.05 level of significance) is 3.84.

10
Six Sigma
Contingency Analysis - Steps Green Belt

4. Collect the Data and calculate the Test Statistic. Data should then be
collected and sorted into the cells and the expected counts table
prepared. Now the test statistic can be calculated:
k
(Oi  Ei ) 2
 
2
0
i 1 Ei
(673  806.1) 2 (816  682.9) 2 (2880  2746.9) 2 (2194  2327.1) 2
 
2
0   
806.1 682.9 2746.9 2327.1
 02  21.98  25.94  6.45  7.61
 02  61.98

5. Draw the Conclusion. The last step is to compare the calculated


value of the test statistic to the table value obtained from the chi-
squared table. If the calculated value is greater than the table value,
then it falls into the rejection region. The null hypothesis would then
be rejected in favor of the alternative hypothesis.

11
Six Sigma
Chi-Squared Table Green Belt
Degrees a = 0.2 a = 0.1 a = 0.05 a = 0.01
of
Freedom
1 1.64 2.71 3.84 6.64
2 3.22 4.61 5.99 9.21
3 4.64 6.25 7.82 11.34
4 5.99 7.78 9.49 13.28
5 7.29 9.24 11.07 15.09
6 8.56 10.65 12.59 16.81
7 9.80 12.02 14.07 18.48
8 11.03 13.36 15.51 20.90
9 12.24 14.68 16.92 21.67
10 13.44 15.98 18.31 23.21
11 14.63 17.28 19.68 24.73
12 15.81 18.55 21.03 26.22
13 16.99 19.81 22.36 27.69
14 18.15 21.06 23.69 29.14
15 19.31 22.31 25.00 30.58
16 20.47 23.54 26.30 32.00
17 21.62 24.77 27.59 33.41
18 22.76 25.99 28.89 34.81
19 23.90 27.20 30.14 36.19
20 25.04 28.41 31.41 37.57
25 30.68 34.38 37.65 44.31
30 36.25 40.26 43.77 50.89
12

You might also like