Measurement Systems Analysis

One root cause of variation is measurement.

While the QI Macros cannot teach you everything about Measurement Systems Analysis, here are the basic steps.
Comments: Many cells contain hidden comments. Hover your mouse of these cells to see the comment.
Gage R&R Average and Range Method and Anova Method (both in one worksheet)
1. Select 10 DIFFERENT parts and at least two "appraisers" and one gage
(If you don't use enough parts, part variation will be small and magnify the appraiser/equipment variation)
Rule: if you're measuring to 0.0001 range of parts should be 10 times the resolution (e.g., 0.4995 to 0.5005)
2. RANDOMLY, have appraisers measure each part at least twice but you can have up to five trials

3. Then enter the results of each measurement in Cells C3:L7, C11:L15, C19:L23
If there is a specification tolerance (+/-.05 = 0.10 tolerance) enter it in B43
If there is a reference value (independently verified and accurate target value for the parts) enter in O7, O15, O23
4. The template will automatically calculate all of the values based on your data
If NDC (F52) is less than 5, parts do not represent a big enough range for adequate analysis
Rows 44:45 will identify state of measurement system using Average and Range Method.
B48:B52 show Average and Range method results using part variation
C48:C52 show results using specification tolerance
Rows 66:71 show results using ANOVA method
5. Evaluate your measurement system based on %R&R and make adjustments as required.
Bias
Instructions
1. Measure a part 15 times using the gage (A2:A16)
2. Insert an actual, independently verified "reference value" in B2
2. Bias of gage and confidence intervales will be shown F7:G8
Linearity
Linearity evaluates a gage's ability to measure accurately across a wide range
Imagine measuring a baseball bat:
1. at the grip
2. in the middle
3. and at the widest part
Instructions
1. Measure 5 different parts 12 times using the gage (B3:F14)
2. Insert an actual, independently verified "reference values" for each of the parts in B16:F16
2. Bias of gage and confidence intervales will be shown F7:G8
Attribute Gage Method for Pass/Fail Gages
INSTRUCTIONS
1. Enter Spec Limits in N3:O3 (USL, LSL)
2. Select 10 parts and measure them accurately to get reference value
(These parts must have an actual reference value spanning unacceptable low to unacceptable high)
3. Have 3 appraisers accept/reject 10+ parts with at least three trials.
4. Enter pass/fail in columns B-J
5. Enter Reference value in column L
6. Evaluate effectiveness columns AJ-AV

Sheet
GageR&R
Range Method
Bias
Linearity

basic steps.

Watch GageR&R Video

How To Do Gage R&R

Attribute Gage
Analytic Attribute
Type 1 GRR

Ford Test Data

Destructive Testing
Purchase AIAG MSA Reference

Purchase GageR&R WhitePaper

How to Analyze Bias

How to Analyze Linearity

 2010 KnowWare International Inc

2253 S Oneida St Ste 3D
Denver, CO 80224

Licensed users:
May update this file
May copy this file
May share completed files with co-workers, suppliers and customers

Unlicensed users:
May receive and review completed templates created by licensed users
May NOT use this template to perform their own analysis without a license

To purchase a license:
http://www.qimacros.com/products.php
Email [email protected]
Call 888 468 1537 - (303) 756 9144
Fax a PO to 888 468 1536 - (303) 756 3107

Gage R&R
Part Number
http://www.aiag.org/
http://www.qimacros.com/free-lean-six-sigma-tips/aiag-msa-gage-r&r.html
Average & Range Method
1
2
3
4
5
6
7
8
9
10
Sum
Appraiser 1
Trial 1
-0.480
0.29
-0.56
Enter your data here-> Trial2
0.41
-0.68
Trial3
0.060
0.64
-0.58
Trial4
Xbar1
Reference
Trial 5
-0.08
-0.341 <-Insert Reference
Total
1.34
-1.82
Bias Value Here
Average-Appraiser
0.4467 1 -0.607 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
Rbar1
0.261
Range1
0.35
0.12 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
0.235
Appraiser 2
Trial 1
-1.970
0.08
-0.47
Enter your data here-> Trial2
0.25
-1.22
Trial3
-0.610
0.07
-0.68
Trial4
Xbar2
Reference
Trial 5
-0.328333
-0.341
Total
0.4
-2.37
Bias
Average-Appraiser
0.1333 2 -0.79 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
Rbar2
0.013
Range2
0.18
0.75 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
0.465
Appraiser
Trial 1
-3.690
0.04
-1.38
Enter your data here-> Trial2
-0.11
-1.13
Trial3
-1.110
-0.15
-0.96
Trial4
Xbar3
Reference
Trial 5
-0.615
-0.341
Total
-0.22
-3.47
Bias
Average-Appraiser
-0.0733 3 -1.157 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
Rbar3
-0.274
Range3
0.19
0.42 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
0.305
Range Average
0.3350
Constants
XDiff
0.5350
10 Trials 9Trials 8 Trials 7 Trials 6 Trials 5 Trials 4 Trials 3 Trials 2 Trials # Trials
3
UCL
0.8643
1.777 1.816 1.864
1.924 2.004
2.11
2.28
2.58
3.27
D4
2.58
UCL represents the limit of the individual
LCL
0.0000
0.223 0.184 0.136
0.076
0
0
0
0
0
D3
0
Circle those beyond the limit.
Repeatability(EV)
0.1979
0.577 0.729 1.023
1.88
A2
1.023
Identify and correct causes
0.308
0.337
0.373
0.419
0.483
Reproducibility(AV)
0.2679
0.3249 0.3367 0.3512 0.3698 0.3946 0.4299 0.4857 0.5908 0.8862
K1 0.590818
LCL = 0 D3=0 for up to 7 trials
Gage Capability(R&R)
0.3331
0.7071 0.5231
K2
0.5231
Spec Tolerance
2 Ops 3 Operators
XL2007 Bug causes this error me
2.02
3.0775
2.97 2.8472 2.7044 2.5344 2.3259 2.0588 1.6926 1.1284 d2
A formula in this worksheet contain
% Using % Using Gage system needs improvement
This bug was fixed in XL2010 and d
AIAG - Automotive IndustryTV
ActionTolerance
Group Formulas Operator may need to be better trained or gage is hard to read
EV (Equipment Variation) 0.1979
Equipment Variation (EV)
Components of Variation
%EV
24.9% 58.8% # Parts #Trials #Ops
% of Total Variation (TV)
90.8%
100.0%
AV: (Appraiser Variation)0.26794
2
3
3 Appraiser Variation(AV)
80.0%
%AV
33.7% 79.6%
% of Total Variation (TV)
60.0%
41.9%
R&R (Gage Capability) 0.3331
Repeatability and Reproducibility (R&R) 40.0%
33.7%
24.9%
%R&R
41.9% 98.9%
NDC 3
% of Total Variation (TV)
20.0%
PV (Part Variation)
0.7212 Not enough Part Variation for Study Part Variation (PV)
0.0%
%PV
90.8%
214%
% of Total Variation (TV)
%EV
%AV
%R&R
%PV
TV (Total Variation)
0.7945
Total Variation (TV)

Calculate GageR&R using Anova

With Interaction Without Interaction
P
P
Anova Source
df
SS
MS
F
F
Appraiser
2 0.8601 0.4301 10.307 0.088 11.525 0.001
Parts
1 4.6818 4.6818 112.2 0.009 125.46 0.000
Appraiser x Part
2 0.0217 0.0109
0.26
0.775
Gage w AP Interaction
12 0.5007 0.0417
Gage w/o AP Interaction
14 0.5224 0.0373
Total
17 6.0644
Without Interaction
NDC 3

Without Interaction
Repeatability
Appraiser
AppraiserxPart
R&R
Part

Estimate
of
Variance

0.03732
0.0655
0
0.10278
0.51605

Std. Dev

0.1932
0.2559
0
0.3206
0.7184

Total
%
Variation Estimate
Total
Contributi
Using
of
Variation
on
Tolerance Variance

EV 24.6%
AV 32.5%
INT
0.0%
R&R 40.8%
PV 91.3%
TV 4.7199

6% 57.4%
11% 76.0%
0%
0.0%
17% 95.2%
83% 213.4%
0.3367

######
0.0699
0.0103
######
######

With Interaction

Std. Dev

0.2043
0.2643
0.1015
0.3491
0.7204

NDC 3

Total
%
Variation
Total
Contributi
Using
Variation
on
Tolerance

EV 26.0%
AV 33.6%
INT 12.9%
R&R 44.4%
PV 91.6%
TV ######

7% 60.7%
11% 78.5%
2% 30.1%
20% 103.7%
84% 214.0%

Part Number

Part
Plot
3 by Appraiser
4
5 (Stacked)
6

10

0.6

0.4
0.2
0

Averagge

10

Average-Appraiser 1
Average-Appraiser 2

-0.2

Average-Appraiser 3

-0.4

UCL
XCL

-0.6

LCL

-0.8

Average or Reference Value

Error Appr 1
-1
-1.2
-1.4

0.6
0.4
0.2
0

Averagge

-0.2
-0.4
-0.6

-0.8
-1
-1.2
-1.4

0.1689 -0.851
0.1211 0.2911 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
0.2411 0.1711 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
0.4711 0.2711 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
Part
Error Appr 2 -0.0889 0.3811 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
0.0811 -0.369 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
-0.0989 0.1711 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
Part
Plot
#N/A
#N/A
#N/A
#N/Aby Appraiser
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
Error Appr -0.1289
3
-0.529 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
-0.2789 -0.279 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
-0.3189 -0.109 #N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
#N/A
1st Quartile
Upper Limit
Lower Limit
3rd Quartile
Appraiser 1
Appraiser 2
Appraiser 3
Part 1 - App 10.35
0.64
0.29
0.525
range count
range count
range count
2
0.075
0.25
0.07
0.165
#N/A
44 #N/A
44 #N/A
44
3
-0.13
0.04
-0.15 -0.035
#N/A
0 #N/A
0 #N/A
0
Part 2 - App 1-0.63
-0.56
-0.68
-0.57
#N/A
0 #N/A
0 #N/A
0
2
-0.95
-0.47
-1.22 -0.575
#N/A
0 #N/A
0 #N/A
0
3 -1.255
-0.96
-1.38 -1.045
#N/A
0 #N/A
0 #N/A
0
Part 3 - App 1
#N/A
0 #N/A
0 #N/A
0
2
#N/A
0 #N/A
0 #N/A
0
3
#N/A
0 #N/A
0 #N/A
0
Part 4- App 1
#N/A
0 #N/A
0 #N/A
0
2
#N/A
0 #N/A
0 #N/A
0
3
#N/A
0 #N/A
0 #N/A
0
Part 5- App 1
#N/A
0 #N/A
0 #N/A
0
2
#N/A
0 #N/A
0 #N/A
0
Appraiser 1
Appraiser
3
3
#N/A
0
#N/A
0
#N/A
0
Appraiser 2
Part 6 - App 1
#N/A
0 #N/A
0 #N/A
0

Series1
UCL

XCL
LCL

2
3
Part 7 - App 1
2
3
Part 8 - App 1
2
3
Part 9 - App 1
2
3
Part 10 - App 1
2
3

1
0.9
0.8

0.7

Range

0.6
0.5
0.4

#N/A
0
Range of Parts by Appraiser (Stacked)

#N/A

#N/A

Range1
Range2
Range3

UCL
Rave

0.3
0.2
0.1
0

10

Parts

Range of Parts by Appraiser

1
0.9

0.8
0.7

0.5

Series1
UCL

0.4

Rave

0.3
0.2

0.1

Appraiser 1

Appraiser 2

Appraiser 3

Sum

10

10

Sum

Sum

10

Range

0.6

Run Chart of All Parts

1

0.5

0
0

10

12

-0.5

-1

-1.5

Scatter Chart - Parts 1-4

1

0.5

-0.5

-1

-1.5
1

Part Number by Appraiser

Appraiser

Scatter Chart - Parts 5-8

1.2

0.8

0.6

0.4

0.2

0
5

Part Number by Appraiser

Scatter Chart - Parts 9-10

1.2

0.8

0.6

0.4

0.2

0
9

10
Part Number by Appraiser

Box and Whisker Chart

1

0.5
0
-0.5

-1
-1.5

Parts By Appraiser

Error Chart - Parts 1-4

0.6

0.4

0.2

-0.2

-0.4

-0.6
1

Part Number by Appraiser

Appraiser

Error Chart - Parts 5-8

1.2

0.8

0.6

0.4

0.2

0
5

Part Number by Appraiser

Error Chart - Parts 9-10

1.2

0.8

0.6

0.4

0.2

0
9

10

Part Number by Appraiser

Appraiser 1 - Normalized Error

50

45
40
35

Frequency

30

25
20
15

10
5
0

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

Appraiser 2 - Normalized Error

50
45

40

Frequency

35
30

25
20

15
10

5
0

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

Appraiser 3 - Normalized Error

50
45
40

35

Frequency

30
25
20

15
10
5
0

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

#N/A

X-Y Plot of Averages of Appr A

0.8
0.6
0.4
Appr A

0.2
0
-0.2
-0.4
-0.6
-0.8
0

10

15

20

25

Overall Average

30

35

40

45

X-Y Plot of Averages of Appr B

0.4
0.2
0

Appr B

-0.2

-0.4
-0.6
-0.8
-1

-1.2
-1.4
0

10

15

20
25
Overall Average

30

35

40

45

30

35

40

45

X-Y Plot of Averages of Appr C

0.2
0
-0.2

Appr C

-0.4
-0.6
-0.8

-1
-1.2
-1.4
-1.6
0

10

15

20
25
Overall Average

Comparison of Appr A-B

0.4
0.2
0

Appr B

-0.2
-0.4
-0.6
-0.8
-1
-1.2
-1.4
-0.8

-0.6

-0.4

-0.2

0.2

0.4

0.6

0.8

0.2

0.4

0.6

0.8

Appr A

Comparison of Appr A-C

0.2
0
-0.2

Appr C

-0.4
-0.6
-0.8
-1
-1.2
-1.4
-1.6
-0.8

-0.6

-0.4

-0.2

0
Appr A

Comparison of Appr B-C

0.2
0
-0.2

Appr C

-0.4
-0.6
-0.8
-1
-1.2
-1.4
-1.6
-1.4

-1.2

-1

-0.8

-0.6

-0.4
Appr B

-0.2

0.2

0.4

Instructions 1. Select 10 DIFFERENT parts and at least two "appraisers" and one gage
(If you don't use enough parts, part variation will be small and magnify the appraiser/equipment variation)
Rule: The 10 parts should cover the Spec Tolerance (e.g., 0.4995 to 0.5005), otherwise identical parts will cause high %R&R.
If NDC (Number of Distinct Categories, F52) is less than 5, there is not enough part variation to enable an accurate study.
2. RANDOMLY, have appraisers measure each part at least twice
3. Then enter the results of each measurement in Cells C3:L7, C11:L15, C19:L23
4. The template will automatically calculate all of the values based on your data
5. Evaluate your measurement system based on %R&R and make adjustments as required.
Gage Repeatability and Reproducibility
Measurement is one common cause of variation. Gage R&R helps improve measurement systems
If repeatability (EV - Can the same person using the same gauge measure the same thing consistently)
is larger than reproducibility, reasons include:
1. Gage instrument needs maintenance
2. Gage needs to be redesigned
3. Clamping or location needs to be improved
4. Excessive within-part variation
If reproducibility (AV-can two appraisers measure the same thing and get the same result)
is larger than repeatability, reasons include:
1. Operator needs to be better trained in how to use and read gage
2. Calibrations on gage are not clear
3. Fixture required to help operator use gage more consistently
Gage System Acceptability
%R&R<10%
Gage System Okay (Most variation caused by parts, not people or equipment)
%R&R<30%
May be acceptable based on importance of application and cost of gage or repair

CL represents the limit of the individual R's Verified using AIAG Measurement Systems Analysis 3rd Edition
cle those beyond the limit.
entify and correct causes
To learn how to use GageR&R, get the AIAG MSA reference.
L = 0 D3=0 for up to 7 trials
Order from:
http://www.aiag.org

http://www.qimacros.com/200303Six.html
L2007 Bug causes this error message (not a QI Macros bug):
formula in this worksheet contains one or more invalid references
his bug was fixed in XL2010 and does not exist in XL97-2003

Measurements
70.1000
70.5700
70.0500
71.1100
70.5500

Part Number
Part Name
Control Plan
USL
Target
LSL
Tolerance
Reference

12345
Name

Gage
Characteristic
Specification
Discrimination
Appraiser

80
70.5
60
20
70.55

Capability
Cg
Cgk
%Var - R
%Var R&B

Gage R&R
1.55 Acceptable
1.49 Acceptable
12.90%
13.39%

Average
Stdev
Bias
n
T
p

#1
Diameter
mm
0.001

70.476
0.430
0.074
5
0.385
0.720

Measurements
73.00
72.00
71.00
70.00
69.00
68.00
67.00
66.00
1

10 13 16 19 22 25 28 31 34 37 40 43 46 49

Measurements

Target

Ref - 0.1*Tol

Ref + 0.1*Tol

Range Method

Range Method
Part
Appraiser 1
Appraiser 2
Range
GR&R
Insert Process
Standard
Deviation
%GR&R

2
0.85
0.8
0.05

Insert Your Data Here

3
4
0.75
1
0.45
0.7
0.95
0.55
0.05
0.05
0.1

0.058824

0.0777
75.7%

Page 19

Average

0.5
0.6
0.1

0.71
0.72 d2
0.07
1.19

Bias

Reference
Value
Value
Bias
5.8
6.0
-0.2
5.7
-0.3
5.9
-0.1
5.9
-0.1
6.0
0.0
6.1
0.1
6.0
0.0
6.1
0.1
6.4
0.4
6.3
0.3
6.0
0.0
6.1
0.1
6.2
0.2
5.6
-0.4
6.0
0.0

Average/
n
Mean
Stdev
15
6.006667
0.21202
Bias
0.0067

Stdev(r)
Stdev(b)
0.2251409
0.0547433

tStatistic/
critical
d2*/d2
0.121781
3.55333
2.144787
3.47193

95% Confidence
Lower
Upper
-0.1107
0.1241
Bias Acceptable
Reference: AIAG MSA Version 4 pgs 87-92
Bias

Page 20

Bias

df

m
14 d2*
df
d2
t crit 95%

2
3
4
5
6
7
8
1.41421
1.91155
2.23887
2.48124
2.67253
2.82981
2.96288
1
2
2.9
3.8
4.7
5.5
6.3
1.12838
1.69257
2.05875
2.3593
2.53441
2.70436
2.8472
4.302653 3.182446 2.776445 2.570582 2.446912 2.364624 2.306004

Page 21

Bias

9
10
11
12
13
14
3.07794
3.17905
3.26909
3.35016
3.42378
3.49116
7
7.7
8.3
9
9.6
10.2
2.97003
3.07751
3.17287
3.25846
3.33598
3.40676
2.262157 2.228139 2.200985 2.178813 2.160369 2.144787

Page 22

15
16
17
3.55333
3.61071
3.66422
10.8
11.3
11.9
3.47193
3.53198
3.58788
2.13145 2.119905 2.109816

Bias

18
19
20
3.71424
3.76118
3.80537
12.4
12.9
13.4
3.64006
3.68896
3.735
2.100922 2.093024 2.085963

Page 23

Linearity

Part
Trials

Average
Reference
Values
Bias
Range
bias=
slope*ref+b
n=
Slope=
b=
Goodness of
fit

Bias
3

2.7
2.5
2.4
2.5
2.7
2.3
2.5
2.5
2.4
2.4
2.6
2.4
2.49

5.1
3.9
4.2
5
3.8
3.9
3.9
3.9
3.9
4
4.1
3.8
4.13

5.8
5.7
5.9
5.9
6
6.1
6
6.1
6.4
6.3
6
6.1
6.03

7.6
7.7
7.8
7.7
7.8
7.8
7.8
7.7
7.8
7.5
7.6
7.7
7.71

2.00
0.492
0.40

4.00
0.125
1.30

6.00
0.025
0.70

8.00
-0.292
0.30

-0.05333

-0.31667

1.5

0.473333

0.21

Insert Process
Variation
2.488105
Linearity
0.3276
%Linearity
13.17%

4.00
1.1
-0.1
0.2
1.0
-0.2
-0.1
-0.1
-0.1
-0.1
0.0
0.1
-0.2
0.125

8.00
-0.4
-0.3
-0.2
-0.3
-0.2
-0.2
-0.2
-0.3
-0.2
-0.5
-0.4
-0.3
-0.292

10.00
-0.9
-0.7
-0.5
-0.7
-0.6
-0.5
-0.5
-0.5
-0.4
-0.8
-0.7
-0.6
-0.617

10.00 0.358504 0.095171 -0.16816

-0.617 0.588162 0.324829 0.061496
0.50
Linearity

-0.4315
-0.20184

-0.69483
-0.46517

-0.58

1.0

60
-0.13167
0.736667
0.977907

6.00
-0.2
-0.3
-0.1
-0.1
0.0
0.1
0.0
0.1
0.4
0.3
0.0
0.1
0.025

9.1
9.3
9.5
9.3
9.4
9.5
9.5
9.5
9.6
9.2
9.3
9.4
9.38

2.00
0.7
0.5
0.4
0.5
0.7
0.3
0.5
0.5
0.4
0.4
0.6
0.4
0.492

0.5
Bias

1
2
3
4
5
6
7
8
9
10
11
12

0.0
0.00

2.00

4.00

6.00

-0.5

-1.0
Reference Values

Page 24

8.00

10.00

12.00

Linearity

-0.05
+ Confidence Limits
Linearity
Linearity evaluates a gage's ability
to measure accurately across a wide range
Imagine measuring a baseball bat:
1. at the grip
2. in the middle
3. and at the widest part

Page 25

Appraiser A
Part #

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

Trial 1

Trial 2

1
1
0

1
1
0

Appraiser B

Appraiser C

Reference Results
Trial 3 Trial 1 Trial 2 Trial 3 Trial 1 Trial 2 Trial 3 Reference Ref Value Code
1
1
1
1
1
1
1
1
0.476901
+
1
1
1
1
1
1
1
1
0.509015
+
0
0
0
0
0
0
0
0
0.573646
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-

LSL
USL %R&R Gray zone
0.45
0.545
25%
0.02375
INSTRUCTIONS
1. Enter Spec Limits Above
2. Have 3 appraisers accept/reject 10+ parts
3. Enter pass/fail in columns B-J
4. Enter Reference value in column L
5. Evaluate effectiveness columns AJ-AV

Cro

40
41
42
43
44
45
46
47
48
49
50

0
0
0
0
0
0
0
0
0
0
0
0 = Fail
1 = Pass

+ = within spec limits

- = outside spec limits
x = gray area around spec limits

eject 10+ parts

Crosstabulation
B
Total
A
0
1
0
3
0
3
Expected
1.0
2.0
3
1
0
6
6
Expected
2.0
4.0
6
Total
3
6
9
C
B

0
0
3
Expected
1.0
1
0
Expected
2.0
Total
3

0
0
3
Expected
1.0
1
0
Expected
2.0
Total
3

KappaA
A
B
1.00
C
1.00

0
1
Total

Total
1
0
2.0
6
4.0
6

C
A

Crosstabulation
REF
0
1
3
0
1.0
2.0
0
6
2.0
4.0
3
6
REF
0
3
1.0
0
2.0
3

B
3
3
6
6
9

0
1
Total

Total
1
0
2.0
6
4.0
6

B
1.00
1.00

C
1.00
1.00
-

0
1
Total

Ref

A
1.00

3
3
6
6
9

1
0
2.0
6
4.0
6

3
3
6
6
9
Total

1
0
2.0
6
4.0
6

C
1.00

Total

REF
0
3
1.0
0
2.0
3

C
3
3.0
6
6
9

Total

1.00

3
3
6
6
9

Total
# Match

Mixed
95% UCI
Score
95% LCI

% Appraiser
B
3
3
3
3

100%
100%
29%

100%
100%
29%

C
3
3

100%
100%
29%

System % Effective Score

3
3
100%
100%
29%

Total
# Match
95% UCI
Score
95% LCI

Effectiveness
A 100.0%
B 100.0%
C 100.0%
Acceptable
Marginal
Unacceptable

>= 90%
>=80%
<80%

Miss
Rate
0.0%
0.0%
0.0%

False
Alarm
0.0%
0.0%
0.0%

<=2%
<=5%
> 5%

<=5%
<=10%
>10%

% Score vs Attribute
A
B
C
3
3
3
3

0
100%
100%
29%

0
100%
100%
29%

3
3

0
100%
100%
29%

System % Effective Score vs Ref

3
3
100%
100%
29%

Xt
-0.016
-0.015
-0.014
-0.013
-0.012
-0.011
-0.0105
-0.01
-0.008

P'a
0
1
3
5
8
16
18
20
20

0.025
0.075
0.175
0.275
0.425
0.775
0.875
0.975
1

Slope(m)= 148.1918
b=
2.314112
R^2=
0.924564
F=
85.79383
df=
7
sum squares1.163923
regression=

Normal Probability Plot

y = 148.19x + 2.3141

-0.016

-0.014

-0.012

-0.01

Analytic Method Instructions (AIAG MSA Version 3, pp.135-140)

1. Set the spec tolerance (e.g., +/- 0.01)
2. Select 8 parts near the lower tolerance (-.01)
3. Measure each one 20 times using the attribute gage
4. Count the number of times each part is accepted
5. Smallest part must be accepted zero times (a=0)
6. The largest part must be accepted all a=20 tries
7. The other six parts must be accepted a=1-19 times (not zero or 20)
8. If any of the other six parts are accepted 0 or 20, measure another part until you get six between 1-19
9. Enter parts and number of times accepted into columns A and B as shown
For best results, take samples at a=0 and a=20, then work toward the middle
Repeat until criteria are met.
10. Plot on Normal Probability Paper or use Bias/Repeatability from this chart

1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
-0.008

Spec
Slope(m)=
Tolerance (+/-)
148.1918239
0.01
b=
2.314111635
P =.5 Cross
-0.012241645
Bias
0.002241645
Repeatability
0.006185676
t
11.34289451
Bias significantly different from zero