Академический Документы
Профессиональный Документы
Культура Документы
Operation
OK
Inspect
First Time
Correct
NOT
OK
Rework
Hidden Factory
Scrap
Bill Rodebaugh
Objectives
Operation
OK
Inspect
First Time
Correct
NOT
OK
Rework
Hidden Factory
Scrap
slide 3
Lab Work
OK
Inspect
Production
NOT
OK
Re-test
Hidden Factory
Waste
slide 5
Objectives
Measurement Variation
Long-term
Short-term
Variation
Variation due
Variation due
Process Variation
Process Variation
w/i sample
to gage
to operators
Repeatability
Calibration
Stability
Linearity
LSL
Frequency
15
USL
10
0
30
40
50
60
70
80
90
100
110
Process
15
LSL
USL
10
Frequency
Observed process
variation With measurement error
0
30
40
50
60
70
Observ ed
slide 8
80
90
100
110
slide 9
515
. * MS
P/T
Tolerance
Usually
Usually expressed
expressed
as
percent
as percent
Note: 5.15 standard deviations accounts for 99% of Measurement System (MS) variation.
The use of 5.15 is an industry standard.
slide 10
MS
x 100
Usually
Usually expressed
expressed
as
percent
as percent
slide 11
Objectives
High Impact Six Sigma project was chartered to improve an important quality variable,
CTQ1
The measurement of CTQ1 was originally not questioned, but the team decided to study
the effectiveness of this measurement
Initial project improvements have somewhat equalized performance across sites. Small
level improvements are masked by the measurement effectiveness of CTQ1
slide 13
Site 2 Lab
Site 3 Lab
Site 4 Lab
6 analyses/site/sample
2 samples taken from each site
2*4 Samples should be representative
Each site analyzes other sites sample.
Each plant does 48 analyses
6*8*4=196 analyses
slide 14
Gage name:
Date of study:
Reported by:
Tolerance:
Misc:
Z-14 MSA
JULY 2002
All Labs
110
Surface Area
Components of Variation
Response By Sample
120
%Contribution
%Study Var
%Tolerance
100
Percent
890
80
840
60
790
40
20
740
0
Gage R&R
Repeat
Reprod
Sample
Part-to-Part
R Chart by Operator
Sample Range
100
V1
V2
V3 W1
Response By Operator
W2
890
W3
840
UCL=52.45
50
790
R=16.05
0
LCL=0
740
Oper
V1
V2
V3 W1
Operator*Sample Interaction
W2
UCL=851.5
850
Mean=821.3
800
Operator
900
W3
LCL=791.1
Average
900
Sample Mean
850
800
750
Sample
slide 15
CB1
CB2
CB3
LC1
LC2
LC3
V1
V2
V3
W1
W2
DF
Sample
SS
MS
14221
2031.62
5.0079
0.00010
Operator
11
53474
4861.27
11.9829
0.00000
Operator*Sample
77
31238
405.68
1.4907
0.03177
Repeatability
96
26125
272.14
191
125058
Total
%Contribution
Source
VarComp
617.39
90.11
Repeatability
272.14
39.72
Reproducibility
345.25
50.39
278.47
40.65
66.77
9.75
67.75
9.89
Operator
Operator*Sample
Part-To-Part
(of VarComp)
slide 16
Sample, Operator,
& Interaction are
Significant
*Conf Int not calculated with Minitab, Based upon R&R Std Dev
slide 17
DotplotsResults
of C16 by C17
CTQ1 MSA Study
(Minitab Output)
(group means are indicated by lines)
C16
840
790
slide 18
Site 3
WO SA
Site 2
VF SA
Site 1
LC SA
C17
CB SA
740
Site 4
DF
SS
MS
37514
12505
26.86
0.000
Error
188
87518
466
Total
191
125032
Site
Mean
StDev
Site 1
48
824.57
15.38
Site 2
48
819.42
22.11
Site 3
48
800.98
20.75
Site 4
48
840.13
26.58
-+---------+---------+---------+----(---*---)
(---*---)
(---*---)
(---*---)
-+---------+---------+---------+-----
Pooled StDev =
21.58
795
810
825
840
Per
790
40
740
CTQ1
MSA Study Results (Minitab Output)
0
20
Gage R&R
Repeat
Reprod
Part-to-Part
X-bar
R ofbyAll
Samples for All Sites
R Chart
Operator
Sample Range
100
V1
V2
V3 W1
W2
UCL=52.45
R=16.05
0
LCL=0
0
Discrimination
Index840is 0,
however
can
790
probably see
740
differences
of 5
Oper
CB1 CB2 C
V1
V2
V3 W1
O
W2
900
W3
UCL=851.5
850
Mean=821.3
800
LCL=791.1
Most850of the
samples are
800
seen as
noise
Average
Sample Mean
900
750
Sample
slide 20
890
W3
50
Sample
Pe
50
Repeat
Reprod
Part-to-Part
Sample Range
X-bar R ofRAll
Samples
for Site 4
Chart
by WO OP
70
60
50
40
30
20
10
0
W1
W2
W3
UCL=60.99
R=18.67
LCL=0
0
Xbar Chart by WO OP
Sample Mean
900
W1
W2
W3
UCL=875.2
850
Mean=840.1
LCL=805.0
800
0
Sampl
LC1
LC2
LC3
850
UCL=853.1
840
Mean=819.4
LCL=785.7
1000
1
1
1
900
800
4
2
22 4
6
MSA Study
820
Results with
810
800 Mean = 819.4
830
790
Individual Value
LC OP*Sa
Average
Sample Mean
Xbar Chart by LC OP
Sample
11
22
22
55
6 6
662 62
2 22
UCL=899.2
Mean=832.5
2
5
LCL=765.8
700
Subgroup
100
g Range
150
1
100
200
300
400
1 1
Selected
Samples
are Representative
1
1
11
1
11
1
UCL=81.95
slide 22
2002 Historical
Process
Results with
Mean = 832.5
CTQ1 MSA
Study Results Process Linkage
I and MR Chart for TSA (t)
Site 2 Example
810
760
Gage R&R
Sample Range
Individual Value
1000
Reprod
Sample
Part-to-Part
R Chart by LC
1 1OP
1
1
1
100
900
LC1
4
50
800
222 4
6
LC3
22
22
UCL=58.54
1
0
100
860
850
100
840
830
820
50
810
800
790
0
780
LC1
1
11
LC2
11
1
22
810
760
300
LC OP
400
LC1
LC2
LC3
LC OP*Sample Interaction
LC3
850
UCL=853.1
2
2
Mean=832.5
2
Mean=819.4
2
2
222
UCL=899.2
R=17.92
LCL=0
200
55860
6 6
662 62
2 22
Xbar Chart by LC
1 OP
150
By LC OP
1LC2
700
Subgroup
MovingMean
Range
Sample
Repeat
2 LCL=785.7
2
840
Average
Perc
50
830
810
800
790
Sample
UCL=81.95
820
R=25.08
LCL=0
1
2002 Historical LC O
L
Process
L
L
Results with
Range = 25.08
Calc
for
pt7 to8 pt
4
5
6
How can Site 2 see small, real change with a Measurement System with
70+% GR&R?
Use Power and Sample Size Calculator with and without impact
of MS variation. Lack of clarity in process improvement work,
results in missed opportunity for improvement and continued
use of non-optimal parameters
slide 24
2-Sample t Test
Alpha = 0.05
Alpha = 0.05
Sigma = 22.23
Sample
Target
Actual
Difference
Size
Power
Power
2117
0.9000
530
Sigma = 6.67
Sample
Target
Actual
Difference
Size
Power
Power
0.9000
192
0.9000
0.9011
0.9000
0.9002
49
0.9000
0.9036
236
0.9000
0.9002
22
0.9000
0.9015
133
0.9000
0.9001
13
0.9000
0.9074
10
86
0.9000
0.9020
10
0.9000
0.9188
12
60
0.9000
0.9023
12
0.9000
0.9361
14
44
0.9000
0.9007
14
0.9000
0.9156
16
34
0.9000
0.9018
16
0.9000
0.9091
18
27
0.9000
0.9017
18
0.9000
0.9555
20
22
0.9000
0.9016
20
0.9000
0.9095
Previously, it would be pointless to make any process changes within the 22 point
range. Would you really see the change?
As the Six Sigma team pushes the CTQ1 value higher, DOEs and other
tools will have greater benefit
slide 26
Objectives
Six Sigma Black Belts and Green Belts Perform MSAs during Project Work
Lab Managers and Technicians are Part of Six Sigma Teams
Measurement Systems are Improved as Six Sigma Projects are Completed
slide 28
slide 29
Final Thoughts
slide 30