26-01-2024 Quality Solution 1

Measurements Systems Analysis - Agenda

1. Is our data accurate?

• Repeatability & Reproducibility
• Accuracy & Precision
• Measurements System Variation
• Bias, Linearity, Stability, Repeatability, Reproducibility,
Calibration, Gauge R&R
2. Variable Gauge R&R
• Parts, Operators, Variation
• Is the gauge good?
• Workshop
3. Attribute Gauge R&R
• Risk Analysis Method
4. Appendix
• Analysis of Variance (ANOVA)

26-01-2024 Quality Solution 2

 What is a Measurement System?
What is a Measurement System?

• Measurement system is the collection of

instruments or gages, standards, operations,
methods, fixtures, software, personnel,
environment and assumptions used to quantify
a unit of measure or fix assessment to the
feature characteristic being measured; the
complete process used to obtain
measurement.

26-01-2024 Quality Solution

3
What is a Measurement System?
A measurement system is a process

Input Operation Output

Key Process Output Variable

KPOV (deliverable) = a decision on a product,
process, or service
– via a number assigned to a characteristic of a product, process, or
service

The first step in assessing a system is to understand this process and

determine if it will satisfy our requirements

26-01-2024 Quality Solution

4
 What is a Measurement System?
(cont’d)

• From this definition, it follows that a measurement process may be

viewed as a manufacturing process that produces decisions via
numbers (data) for its output.

• Viewing a measurement system this way is useful because it allows

us to bring to bear all the concepts, philosophy and tools that have
already demonstrated their usefulness in the area of statistical
process control.

26-01-2024 Quality Solution

5
What is MSA?
 Measurement System Analysis (MSA)
– MSA primarily deals with analyzing the effect of the
measurement system on the measured value
– The objective of MSA is to assess the quality of the
measurement system.
 We test the system to determine its statistical properties,
and use them in comparison with accepted standards, our
needs and customer requirements.

26-01-2024 Quality Solution

6
Overall Objective of MSA

• Understanding the Measurement System – as a process

• Uncertainty of Measurement
• The range within which the true value of a characteristic is
estimated to lie.
• System properties can be expressed as
• statistical distribution of a series of measurements,
• standard deviations,
• probability,
• percentages,
• error as the difference between actual value and the reference
value,
• as points on a control chart or diagram,
• etc.

26-01-2024 Quality Solution

7
 Is our Data Accurate?
• Gauge R & R is a means of assessing the repeatability and
reproducibility of our measurement systems.

• Gauge R & R studies are carried out in order to discover

how much of the process variation is due to the
measurement device and measurement methods.

Dimension
26-01-2024 Quality Solution 8
 Why MSA required?

Defect prevention
Prediction
Statistics SPC
Data
Inspection

MSA is mandatory prior to SPC

26-01-2024 Quality Solution 9

10
 Measurement Systems Variation

Measurement process components and their interactions contribute

variation in outcome of data

Standards Material People

Measured
Values
Variation

Environment Method Equipment

(Machine)

26-01-2024 Quality Solution

10
 Effects of Measurement Error

Measurement
System Bias —
Averages Determined through
“Bias Study”

 total   product   measurement

Measurement System
Variability Variability —
Determined through
“R&R Study”
2 2 2
 total   product   measurement

26-01-2024 Quality Solution

11
 PURPOSE OF MSA

•Quantify Variation present in the

Measurement system
•Compare the Consistency of
inspectors
•Provide methods for Validation of
Measurement System

26-01-2024 Quality Solution 12

13
 Measurement Systems

Process to assign a number or decision to a

characteristic
• Process
• Appraiser
• Checking Method
• Instrument
• Environment
• E.g., OD checking, Concentricity checking

26-01-2024 Quality Solution 13

14
 ISO-TS16949 Requirements

• Cover all measurement systems covered in Control Plan

• Perform Appropriate statistical study
• Use MSA ( or Customer approved) Manual
• Obtain Customer approval for use of methods other than
MSA manual

26-01-2024 Quality Solution 14

15
 ISO-TS16949 Requirements
Types of Measurement System should be based on
• L.C
• Range
• Construction
• Parameter
• Manufacturer

26-01-2024 Quality Solution 15

16
 ISO-TS16949 Requirements
Vernier
Vernier
L.C – 0.02 mm L.C – 0.02 mm
Range – 0 – 200mm Range – 0 – 200mm
Construction- Ordinary Construction- Ordinary
Parameter- O.D Parameter- Width

Is it Same or different type of MS ?

26-01-2024 Quality Solution 16

 ISO-TS16949 Requirements

Bore Dial Gage Air gage

L.C – 0.001 mm L.C – 0.001 mm
Range – 18-35 mm Range – ---
Construction- Ordinary Construction- Dial
Parameter- Bore Parameter- Bore

26-01-2024 Quality Solution 17

 ACCURACY & PRECISION

ACCURACY PRECISION
• Closeness to reference • Ability of MS to repeat
or master value the same reading
• Required where two or • Required where MS is
more MS measuring a repeatedly used to
same characteristic assess and adjust the
• Same parameters are process
checked at Suppliers end • In process inspection as
or at Customer end per control plan

26-01-2024 Quality Solution 18

 Measurement Accuracy & Precision

Measurement is Measurement is
accurate but not precise but not
precise accurate

Measurement is accurate
and precise
26-01-2024 Quality Solution 19
 ACCURACY & PRECISION
ACCURACY PRECISION
Captured by Captured by

• BIAS • REPEATABILITY
• LINEARITY • REPRODUCIBILITY
• STABILITY

26-01-2024 Quality Solution 20

 Discrimination

• Ability to detect the small changes in process

• Traditional rule
• 1/10th of Tolerance
• 1/3rd Considering the cost/ criticality
• MSA/ SPC requirement
• 1/10th of Process variation

26-01-2024 Quality Solution 21

22
 Measurement System Variation
Bias

Accuracy Linearity

Measurement
System Stability
Variation
Repeatability
Precision
Reproducibility

26-01-2024 Quality Solution 22

 Bias
Bias

True Observed
Value Average

Bias is the difference between the observed

average of the measurements and the true value.

26-01-2024 Quality Solution 23

 Calculating Bias

n max  xi   min  xi 
 xi  repeatability  d 2*
i 1
X 
*
where d 2 is taken from Appendix C
n with g = 1 and m = n

bias = observed average measurement – reference value


b  r
n 
 2 
 
 
Bias  σ b t , 1   zero  Bias  σ b t , 1 
2 

where d2 , d 2* and  are found in MSA Appendix C

bias with g = 1 and m = n and
t t ,1 is found using the standard t tables.
b 2

26-01-2024 Quality Solution 24

 The  level which is used depends on the level of sensitivity
which is needed to evaluate/control the process and is
associated with the loss function (sensitivity curve) of the
product/process. Customer agreement should be obtained if an 
level other than the default value of .05 (95% confidence) is used.

26-01-2024 Quality Solution 25

Gage Bias Work Instruction
1. Obtain accepted reference value using a master or
measuring equipment of higher level, such as layout
equipment to create a “golden unit” or “master part” to
use as a reference value.
2. Measure same part by same appraiser minimum of 10
times using gage under evaluation

26-01-2024 Quality Solution

26
 Gauge Bias Work Instruction
3. Analysis:
– plot the histogram of all the readings
• Determine if any special causes are present
– calculate
• average of all the readings
• bias = average of all the reading - reference value
• sd(bias) = standard deviation of all the readings divided by
the square root of the sample size
• The t-statistic and/or confidence bounds.
– determine if the bias is statistically zero; i.e.
• The t-statistic is less than the critical value
• Zero is contained within the confidence bounds

26-01-2024 Quality Solution

27
 Bias Calculation- Example
Reference
Bias
Value = 6.00
1 5.8 -0.2
2 5.7 -0.3
3 5.9 -0.1
T 4 5.9 -0.1
R 5 6.0 0.0
I 6 6.1 0.1
A 7 6.0 0.0
L 8 6.1 0.1
S 9 6.4 0.4
10 6.3 0.3
11 6.0 0.0
12 6.1 0.1
13 6.2 0.2
14 5.6 -0.4
15 6.0 0.0

26-01-2024 Quality Solution 28

 Bias Calculation- Example

3
Frequency

5.6 5.7 5.8 5.9 6.0 6.1 6.2 6.3 6.4

Measured value

26-01-2024 Quality Solution 29

Bias Calculation- Example – Range Method

max  xi   min  xi 
Reference
Bias
Value = 6.00
1 5.8 -0.2 
2 5.7 -0.3  repeatability d 2*
3 5.9 -0.1
T 4 5.9 -0.1
R 5 6.0 0.0
6.4  5.6
I 6 6.1 0.1
  .2254
A
L
7
8
6.0
6.1
0.0
0.1
3.55
S 9 6.4 0.4
10 6.3 0.3

b  r
.2254
  .0582
11 6.0 0.0
12 6.1 0.1

n = 15 13 6.2 0.2 n 15
14 5.6 -0.4
15 6.0 0.0 bias .0067
t   .1151
b .0582
n
 xi 90.1
i 1
X    6.0067
n 15

bias = observed average measurement –

reference value = 6.0067 – 6.00 = .0067
26-01-2024 Quality Solution 30
Bias Calculation- Example – Range Method

Given: m = n = 15 g=1 t  .1151

d2* = 3.55 d2 = 3.47 df = 10.8
t.975, 10.8 = 2.2060
 b  .0582

  
2 
   
Bias  σ b t , 1   zero  Bias  σ b t , 1 
2 


 3.47 *.05813   3.47 *.05813 

.0067    2.2060   0  .0067    2.2060 
 3.55   3.55 

.1186  0  .1320

26-01-2024 Quality Solution 31

 Linearity

Non-Linearity
Gauge is measuring lower
than true value at high end
Measured Value

Reference Value

Linearity is the difference in bias values over the

expected operating range of the measurement
gauge.
26-01-2024 Quality Solution 32
 Gage Linearity

• Gage linearity can be determined by conducting

bias studies through expected operating range
• Minimum of two bias studies should be
conducted, one at each end of operating range
• Middle of range should also be considered

26-01-2024 Quality Solution

33
 Gage Linearity Study
Gage Linearity Work Instruction
1. Select five to eight parts that can be measured at
different operating ranges of measurement system
2. Determine reference value for each part using
layout inspection or use a set of standards
3. Use one appraiser and the same instrument to
measure parts
4. Take 5 or more repeated measurements on each
part
5. Calculate each part’s bias
bias = observed average - reference value

26-01-2024 Quality Solution

34
 Gauge Linearity Study
(continued)

6. Create a scatter plot by putting the reference values

from smallest to largest on the x-axis, and the
corresponding bias values on the y-axis.
7. Calculate the 95% confidence intervals for the control
limits.
Note: you will find it helpful to graph all of the actual
groups of readings stacked vertically at the related
reference value points

26-01-2024 Quality Solution

35
 Gage Linearity Study
(continued)

8. To analyze the graph:

a. The bias = 0 line must fall entirely within the CI for
acceptable linearity
b. Watch for individual readings that do not follow
the pattern of the groups (e.g. outliers). Also watch
for other patterns indicating unusual variation or
abnormal behavior.
9. Remember that a measurement system with large
(i.e. unacceptable) repeatability can indicate an
statistically acceptable bias even if it practically
unacceptable.

26-01-2024 Quality Solution

36
 Gage Linearity Study
(continued)

10. If measurement system has a linearity problem,

use problem-solving methods to determine
modifications necessary to achieve acceptable
linearity

26-01-2024 Quality Solution

37
 Charting Linearity

LINEARITY STUDY DATA LINEARITY EXAMPLE

PA RT 1 2 3 4 5
Y = 0.736667 - 0.131667X R-Sq = 71.4
T
REFERENC E 2.00 4.00 6.00 8.00 10.00
VALUE
1 2.70 5.10 5.80 7.60 9.10 1
R 2 2.50 3.90 5.70 7.70 9.30
3 2.40 4.20 5.90 7.80 9.50
I 4 2.50 5.00 5.90 7.70 9.30
5 2.70 3.80 6.00 7.80 9.40
A 6 2.30 3.90 6.10 7.80 9.50
7 2.50 3.90 6.00 7.80 9.50
L 8 2.50 3.90 6.10 7.70 9.50
9 2.40 3.90 6.40 7.80 9.60 0
S 10
11
2.40
2.60
4.00
4.10
6.30
6.00
7.50
7.60
9.20
9.30
12 2.40 3.80 6.10 7.70 9.40

PART 1 2 3 4 5
REFERENCE 4.00 6.00 8.00 10.00
2.00
VALUE
1 0.7 1.1 -0.2 -0.4 -0.9 -1
B 2 0.5 -0.1 -0.3 -0.3 -0.7
3 0.4 0.2 -0.1 -0.2 -0.5 2 3 4 5 6 7 8 9 10
I 4 0.5 1 -0.1 -0.3 -0.7
reference values
5 0.7 -0.2 0.0 -0.2 -0.6
A 6 0.3 -0.1 0.1 -0.2 -0.5
Regression
7 0.5 -0.1 0.0 -0.2 -0.5
S 8 0.5 -0.1 0.1 -0.3 -0.5
95% Cl
9 0.4 -0.1 0.4 -0.2 -0.4
10 0.4 0.0 0.3 -0.5 -0.8 Bias Average
11 0.6 0.1 0.0 -0.4 -0.7
12
BIAS AVG
0.4
0.491667
-0.2
0.125
0.1
0.025
-0.3
-0.29167
-0.6
-0.61667
Linearity Study - Graphical Analysis

•Do not worry about calculations , you can use Minitab to help you
26-01-2024 •Consult with your Black Belt expert for further guidance
Quality Solution 38
 Stability
Stability

Time Time
1 2

Stability is the variation (differences) in the

average over extended periods of time using the
same gauge and appraiser to repeatedly
measure the same part
26-01-2024 Quality Solution 39
Stability
Stability (or drift) in a
measurement system
evaluated by measuring
the same master(s) or
part(s) on a single
characteristic over an
extended time period (a
time period is days, not
hours)

26-01-2024 Quality Solution 40

 Measurement System Stability

• Typically not as large a problem as GRR

• Useful to help determine calibration intervals
• Should track from test to test and chart (or at least
record actual readings and other pertinent data in the
gage record)

26-01-2024 Quality Solution

41
 Affects On Gage Stability

• Time – long idle periods or intermittent use

• Very large or very small number of measurements taken between
stability tests
• Environment or system changes, such as: humidity; air pressure

– potential confusion with statistical stability factors, such as

warm up effects, rate of wear, lack of maintenance, untrained
operators or lab technicians

26-01-2024 Quality Solution

42
 Cause of Gage Stability Error

• Infrequent or too frequent calibration

• Lack of air pressure regulator or filter
• Warm-up period for electronic or other gages
• Lack of maintenance
• Wear or damage not readily observable
• Oxidization (corrosion)

26-01-2024 Quality Solution

43
 Gauge Stability Study
Stability Analysis Instructions
1. Use a standard set of parts or reference/master
materials as sample
– Retain these as appropriate (life of product) in a
protected environment
– Label them with name and number for tracking
and further studies include samples at the low,
mid and high range.

26-01-2024 Quality Solution

44
 Analyzing Stability Charts
Xbar/R Chart for Stability
6.3 UCL=6.297
Sample Mean

6.2
6.1
6.0 6.021

5.9
5.8
LCL=5.746
5.7
Subgroup 0 10 20

1.0 UCL=1.010
Sample Range

0.5 0.4779

0.0 LCL=0

26-01-2024 Quality Solution

45
 Analyzing Stability Charts
If stability is not acceptable, the X-bar and possibly S or
R charts will show a shift or out-of-control condition
• out-of-control chart conditions indicate measurement
system not measuring consistently
• Check for:
• bias changed - determine cause of change and
correct
• if cause is wear - may be fixed by re-calibration

26-01-2024 Quality Solution

46
 Gauge Stability Study (continued)

2. (Recommended) Perform a Bias or Linearity study on the

sample. Use this information to establish the control
chart parameters
3. Measure part(s) three to five times (based on knowledge
of measurement system) at different times of the day.
Plot data on X-bar and R or X-bar and S chart
3. Evaluate per normal SPC requirements
Note: Maintain a chart for each part/master
4. Evaluate the within sample standard deviation
(repeatability) of measurements to determine suitability
for the application

26-01-2024 Quality Solution

47
 Repeatability

Repeatability

Repeatability is the variation between successive

measurements of the same part, same
characteristic, by the same person using the same
gauge.

26-01-2024 Quality Solution 48

 Reproducibility
Reproducibility

Operator Operator
1 2

Reproducibility is the difference in the average of

the measurements made by different people
using the same instrument when measuring the
identical characteristic on the same pieces.
26-01-2024 Quality Solution 49
DESIRED MEASUREMENT CHARACTERISTICS FOR CONTINUOUS VARIABLES

Adequate Resolution
There is enough resolution
in the measurement device
X so that the product can
X have many different values.
X X
X X X
X X X X
X X X X X
5.1 5.2 5.3 5.4 5.5

Good if 5 or more distinct

values are observed

26-01-2024 Quality Solution 50

51
 Constitution of total variation

• Total variation
• Manufacturing process variation
• Part to part variation
• Within part variation
• Measurement system variation
• Appraiser variation
• Equipment variation
• Appraiser part interaction

26-01-2024 Quality Solution 51

52
 Methods to capture variation

Elements of Range X bar & R ANOVA

Variation Method Method Method
TV
PV
WIV
AV
EV
App & part
interaction
 Repeatability & Reproducibility

Repeatability Reproducibility
• Same Equipment • Same equipment
• Same parts •Same Parts
•Several trials •Several trials
• One Appraiser •Different Appraiser

This Variation is represented by This Variation is represented by

Equipment Appraiser

26-01-2024 Quality Solution 53

 Range method

No. of Samples 5 nos. gives 80 % chance of

detecting a wrong MS.
10 nos gives 90% chance
No.of Appraisers Min. 2 Recommended 3

Formula Sigma = R bar / d2

Range method - Exercise

Appraiser A • Appraiser B Range

1. 15.03 • 15.05
0.02
2. 15.04 • 15.04
3. 15.02 • 15.03 0.00
4. 15.05 • 15.05 0.01
5. 15.02 • 15.02
0.00
0.00
Tol = 0.05, TV(SPC)= 0.120 (6*Process Std Deviation)

26-01-2024 Quality Solution 55

 Range method - Exercise
Average R = 0.006

Std.deviation = 0.006 / 1.128 = 0.0053

where n=2 d2 is1.128, n=3 d2 is 1.693

G R&R ( 1 sigma) = 0.0053

% GR&RTV = (0.0053)/ (0.120/6) * 100 = 26.5

% R&RTOL = (0.0053)/ ( 0.050/6) * 100= 66.5

Interpretation = MS IS NOT ACCEPTABLE

26-01-2024 Quality Solution 56
 Range method – Acceptance criteria

% R&R < 10 % OF TOL OR TV

% R&R 10 – 30 % OF TOL OR TV ACCEPTABLE SUBJECT TO

ANALYSIS AND JUSTIFICATION W.R.T COST OF REPAIR AND
CRITICALITY

26-01-2024 Quality Solution 57

 Range method – Applicability

Being the short study it is suitable for studying MS where it is time

consuming to measure one measurement e.g Concentricity, R/O

Being the Short study this can be used for Periodic verification of
R&R to detect any changes rather than average range method

26-01-2024 Quality Solution 58

 Average & Range Method

No.of Samples 10 nos. Randomly selected

representing actual process
variation
No.of Appraisers 3 From personnel actually
performing on a day to day
operation
No.of trials Min 2 . Preferably 3

26-01-2024 Quality Solution 59

Average & Range Method- steps for
conducting study
1. Identify the parts from 1 – 10

2.Mark the Place where to check to eliminate Within part variation

3.Communicate the purpose of the study to all appraisers

4.Conduct the study in a Random manner covering all parts for

required no. of trials through all appraisers

5. Monitor all appraisers all following the same method

26-01-2024 Quality Solution 60

Average & Range Method- steps for conducting study
6. Record the observations in a manner which is not seen by the
appraisers – FORMAT-

7. If any abnormal readings are observed ask the appraiser to

repeat the reading

8. Maintain the samples till the completion of analysis

9. Draw R chart and X bar chart and interpret

10. Calculate EV,AV,PV & TV and it Percentage against TOL or TV

and ndc

26-01-2024 Quality Solution 61

Xbar and
Gage R&R R Chart
(ANOVA) by Operator
for Measurement
Reported by :
G age name: Tolerance:
Date of study : M isc:

Xbar Chart by Operator

1 2 3
1.0
Sample Mean

_
_
UCL=0.8796
0.8 X=0.8075
LCL=0.7354
0.6

0.4

R Chart by Operator
1 2 3
0.12 UCL=0.1252
Sample Range

0.08
_
0.04 R=0.0383

0.00 LCL=0

26-01-2024 Quality Solution 62

63
 R chart interpretation

CONDITION INTERPRETATION
• One or more than one • His method is different
point of one appraiser from others
out of UCL R • Measurement System is
• One or more than one sensitive to appraisers
point of all appraisers skill
out of UCL R • Part is deformed or
• In one part all Damaged
appraisers points are
out of UCL R

26-01-2024 Quality Solution 63

 X chart interpretation
INTERPRETATION
CONDITION Measurement system is
More than 50 % of readings adequate enough to capture
are out of control limits process variation

26-01-2024 Quality Solution 64

 Example- Variable Gauge R&R

• Example

26-01-2024 Quality Solution 65

Example-Gasket Thickness Study

PT1 PT2 PT3 PT4 PT5 PT6 PT7 PT8 PT9 PT10 AP/TRIAL
0.65 1.00 0.85 0.85 0.55 1.00 0.95 0.85 1.00 0.60 A1
0.60 1.00 0.80 0.95 0.45 1.00 0.95 0.80 1.00 0.70 A2
0.55 1.05 0.80 0.80 0.40 1.00 0.95 0.75 1.00 0.55 B1
0.55 0.95 0.75 0.75 0.40 1.05 0.90 0.70 0.95 0.50 B2
0.50 1.05 0.80 0.80 0.45 1.00 0.95 0.80 1.05 0.85 C1
0.55 1.00 0.80 0.80 0.50 1.05 0.95 0.80 1.05 0.80 C2

Xbar & R Example

Specification: 0.6 - 1.0 mm

Process Variation: 1.6 mm

66
Gage R&R Study for Thickness – XBar/R Method

Source Variance StdDev 5.15*Sigma

Total Gage R&R 2.08E-03 0.045650 0.235099

Repeatability 1.15E-03 0.033983 0.175015
Reproducibility 9.29E-04 0.030481 0.156975
Part-to-Part 3.08E-02 0.175577 0.904219
Total Variation 3.29E-02 0.181414 0.934282

Source %Contribution %Study Var

Total Gage R&R 6.332 25.164

Repeatability 3.509 18.733
Reproducibility 2.823 16.802
Part-to-Part 93.668 96.782
Total Variation 100.000 100.000

Number of distinct categories = 5

67
 Calculation Explanation

• % Tolerance = the factor standard deviation divided by the

tolerance/6.
Repeatability = 100  repeatability standard deviation/(tolerance/6)

• % Process Variation = the factor standard deviation divided by the

process variation.
Repeatability = 100 x repeatability standard deviation/ process
variation.

• Number of Distinct Categories = part standard deviation divided by

the total measurement standard deviation times 1.41.

26-01-2024 Quality Solution

68
 Calculation Explanation

• 5.15 Sigma = 5.15  the factor standard deviation.

5.15 was developed empirically to approximate the gage population
distribution variation. (99% area for Normal forms)

• % Contribution = Percent contribution of each factor based upon the

variance.
Repeatability = 100  repeatability variance/ total variation variance.

• % Study Variation = the factor standard deviation divided by the total

variation standard deviation.
Repeatability = 100  repeatability standard deviation/ total variation
standard deviation.

26-01-2024 Quality Solution

69
 Is the Gauge Good?
% P/T Acceptability
(6R&R/Process Tolerance)

0 - 10% Very Good (Six Sigma Gauge)

May be Acceptable
10 – 30%
Probably Not Acceptable
>30%

The interpretation will also depend on the current

level of process variation
Note that these guidelines are as recommended in ”Measurement Systems Analysis “

26-01-2024 Quality Solution 70

 Causes for R&R

When repeatability is large compared to reproducibility

 Instrument needs maintenance
 Redesign gage for more rigidity
 Improve clamping or location of gauging
 Excessive within – part variation
( Conduct R & R with within part variation )

26-01-2024 Quality Solution 71

 Causes for R&R

When reproducibility is large compared to repeatability

 Appraisers need better gage use training
 Need better operational definition
 Incremental divisions on instrument are not readable
 Need fixture to provide consistency in gage use.

26-01-2024 Quality Solution 72

 X bar & R method - Applicability
• To capture Precision, Mandatory before SPC study

• Measurement Systems which are used repeatedly on line to

control the process based on the measurement data

• Where individual readings are affected by the precision of

the instrument

• Applicable for all MS used in In process where the MS can be

repeated

26-01-2024 Quality Solution 73

 IMPROVING A MEASUREMENT SYSTEM
•A measurement system consists of
• Measuring devices
• Procedures
• Definitions
• People

•To improve a measurement system, you need to

• Evaluate how well it works now (by asking “how much of the variation
we see in our data is due to the measurement system?”).
• Evaluate the results and develop improvement strategies.

26-01-2024 Quality Solution 74

74
 Rounding Errors

• Rounding is another component of measurement variation

which needs to be minimised
• It can be shown that to avoid rounding error getting in the
way of achieving six sigma quality, it is necessary to have a
minimum of 14 discrete values between the upper and
lower specification
• For one-side specifications, there need to be at least 7
discrete values between the process average and the
specification limit

26-01-2024 Quality Solution 75

75
 Rounding Errors - Interpolating

• If possible interpolate between graduation marks

• For example, thermometers are frequently marked to the

nearest degree but can be read to the nearest 0.2 degrees,
even if the last digit is not entirely accurate

• Interpolating frequently reduces and never increases the

measurement variation

26-01-2024 Quality Solution 76

76
 Destructive Gauge R&R
• Destructive gauge testing means that it is
impossible to carry out repeat tests!

• To complete an assessment of a destructive gauge

it is therefore necessary to assume homogeneity
within batches.

• If there is much more difference in parts between

batches than within batches, then a standard
variable Gauge R & R may be sufficient.

26-01-2024 Quality Solution 77

 Attribute Gauge Study

26-01-2024 Quality Solution 78

 Operational Definitions

• “An operational definition is one that people can do business with.

An operational definition of safe, round, reliable, or any other
quality [characteristic] must be communicable, with the same
meaning to vendor as to the purchaser, same meaning yesterday
and today to the production worker. Example:

1. “A specific test of a piece of material or an assembly

2. “A criterion (or criteria) for judgment
3. “Decision: yes or no, the object or the material did or
did not meet the criterion (or criteria)”

W. E. Deming, Out of the Crisis (1982, 1986), p. 277.

26-01-2024 Quality Solution

79
 Attribute Measure System Analysis
• A MSA study should be carried out on attribute gauges
referred in Control Plan

• Using attribute data, we would have a problem with the

measurement system if:

• Operators disagree with each others’ evaluation of a piece

• The same operator gains different results from a repeat

evaluation of the same piece

26-01-2024 Quality Solution 80

80
 Attribute Measurement System
• An attribute measurement system compares each part to a
standard and either accepts or rejects the part.

• The screen effectiveness is the ability of the attribute

measurement system to properly discriminate good from
bad.

• Screen effectiveness of 100% is desirable.

26-01-2024 Quality Solution 81

81
 Conducting Attribute MSA
1. Select a minimum of 30 parts(Recommended 50 Parts) from
the process. These parts should represent the full spectrum of
process variation (good parts, defective parts, borderline
parts).
2. An “expert” inspector performs an evaluation of each part,
classifying it as “Good” or “Not Good.”
3. Independently and in a random order, each of 2 or 3 operators
should assess the parts as “Good” or “Not Good.”
4. Calculate effectiveness scores.

26-01-2024 Quality Solution 82

82
 Nominal Data

• Kappa Coefficients are used.

• Kappa is an indicator of inter-rater agreement; i.e., the ratio of the
proportion of agreement (corrected for chance) divided by the
maximum number of times they could agree (corrected for chance).

K  Pr(agreement)  Pr(error)
1  Pr(error)

• Where Pr (agreement ) = Proportion(Agreement)

Pr (error) = Proportion (Error)
• A Kappa coefficient of 90% may be interpreted as 90% agreement.

26-01-2024 Quality Solution

83
 Kappa Coefficients
Based on an article by Altman, 1991 recommends the following
kappa interpretation scale. The definition of the interpretations,
such as “Poor”, must be agreed upon by the customer and the
supplier.
Kappa Value Interpretation
< 0.20 Poor
0.21-0.40 Fair
0.41-0.60 Moderate
0.61-0.80 Good
0.81-1.00 Very Good

• In addition, Gardner (1995) recommends that kappa exceed .70

before you proceed with additional data analyses.

26-01-2024 Quality Solution 84

 ATTRIBUTE GAGE CALCULATIONS FOR “COUNTS”

Part A-1 A-2 A -3 B -1 B-2 B-3 C-1 C-2 C-3 Reference

1 1 1 1 1 1 1 1 1 1 1
2 1 1 1 1 1 1 1 1 1 1
3 0 0 0 0 0 0 0 0 0 0
4 0 0 0 0 0 0 0 0 0 0
5 0 0 0 0 0 0 0 0 0 0
6 1 1 0 1 1 0 1 0 0 1
7 1 1 1 1 1 1 1 0 1 1
8 1 1 1 1 1 1 1 1 1 1
9 0 0 0 0 0 0 0 0 0 0
10 1 1 1 1 1 1 1 1 1 1

There are 34 times where

11 1 1 1 1 1 1 1 1 1 1
12
13
0
1
0
1
0
1
0
1
0
1
0
1
0
1
1
1
0
1
0
1
•
14
15
1
1
1
1
0
1
1
1
1
1
1
1
1
1
0
1
0
1
1
1
A-1 = 1 and B-1 = 1
16
17
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 (that is, of the 50 parts checked
there were 34 matches by A and B
18 1 1 1 1 1 1 1 1 1 1
19 1 1 1 1 1 1 1 1 1 1
20 1 1 1 1 1 1 1 1 1 1
21
22
1
0
1
0
0
1
1
0
0
1
1
0
0
1
1
1
0
0
1
0 on their FIRST check)
23 1 1 1 1 1 1 1 1 1 1
24 1 1 1 1 1 1 1 1 1 1
25 0 0 0 0 0 0 0 0 0 0
26 0 1 0 0 0 0 0 0 1 0
27 1 1 1 1 1 1 1 1 1 1
28 1 1 1 1 1 1 1 1 1 1
29 1 1 1 1 1 1 1 1 1 1
30 0 0 0 0 0 1 0 0 0 0
31 1 1 1 1 1 1 1 1 1 1
32 1 1 1 1 1 1 1 1 1 1
33 1 1 1 1 1 1 1 1 1 1
34 0 0 1 0 0 1 0 1 1 0
35 1 1 1 1 1 1 1 1 1 1
36 1 1 0 1 1 1 1 0 1 1
37 0 0 0 0 0 0 0 0 0 0
38 1 1 1 1 1 1 1 1 1 1
39 0 0 0 0 0 0 0 0 0 0
40 1 1 1 1 1 1 1 1 1 1
41 1 1 1 1 1 1 1 1 1 1
42 0 0 0 0 0 0 0 0 0 0
43 1 0 1 1 1 1 1 1 0 1
44 1 1 1 1 1 1 1 1 1 1
45 0 0 0 0 0 0 0 0 0 0
46 1 1 1 1 1 1 1 1 1 1
47 1 1 1 1 1 1 1 1 1 1
48 0 0 0 0 0 0 0 0 0 0
49 1 1 1 1 1 1 1 1 1 1
50 0 0 0 0 0 0 0 0 0 0

85
 ATTRIBUTE GAGE CALCULATIONS FOR “COUNTS”

Part A -1 A-2 A -3 B-1 B-2 B-3 C-1 C-2 C-3 Reference

1 1 1 1 1 1 1 1 1 1 1
2 1 1 1 1 1 1 1 1 1 1
3 0 0 0 0 0 0 0 0 0 0
4 0 0 0 0 0 0 0 0 0 0
5 0 0 0 0 0 0 0 0 0 0
6 1 1 0 1 1 0 1 0 0 1
7 1 1 1 1 1 1 1 0 1 1
8 1 1 1 1 1 1 1 1 1 1
9 0 0 0 0 0 0 0 0 0 0
10 1 1 1 1 1 1 1 1 1 1
11 1 1 1 1 1 1 1 1 1 1

There are 32 times where

12 0 0 0 0 0 0 0 1 0 0
13
14
1
1
1
1
1
0
1
1
1
1
1
1
1
1
1
0
1
0
1
1
•
15
16
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 A-2 = 1 and B-2 = 1
(that is of the 50 parts checked there
17 1 1 1 1 1 1 1 1 1 1
18 1 1 1 1 1 1 1 1 1 1
19 1 1 1 1 1 1 1 1 1 1
20
21
1
1
1
1
1
0
1
1
1
0
1
1
1
0
1
1
1
0
1
1 were 32 matches by A and B on their
SECOND check)
22 0 0 1 0 1 0 1 1 0 0
23 1 1 1 1 1 1 1 1 1 1
24 1 1 1 1 1 1 1 1 1 1
25 0 0 0 0 0 0 0 0 0 0
26 0 1 0 0 0 0 0 0 1 0
27 1 1 1 1 1 1 1 1 1 1
28 1 1 1 1 1 1 1 1 1 1
29 1 1 1 1 1 1 1 1 1 1
30 0 0 0 0 0 1 0 0 0 0
31 1 1 1 1 1 1 1 1 1 1
32 1 1 1 1 1 1 1 1 1 1
33 1 1 1 1 1 1 1 1 1 1
34 0 0 1 0 0 1 0 1 1 0
35 1 1 1 1 1 1 1 1 1 1
36 1 1 0 1 1 1 1 0 1 1
37 0 0 0 0 0 0 0 0 0 0
38 1 1 1 1 1 1 1 1 1 1
39 0 0 0 0 0 0 0 0 0 0
40 1 1 1 1 1 1 1 1 1 1
41 1 1 1 1 1 1 1 1 1 1
42 0 0 0 0 0 0 0 0 0 0
43 1 0 1 1 1 1 1 1 0 1
44 1 1 1 1 1 1 1 1 1 1
45 0 0 0 0 0 0 0 0 0 0
46 1 1 1 1 1 1 1 1 1 1
47 1 1 1 1 1 1 1 1 1 1
48 0 0 0 0 0 0 0 0 0 0
49 1 1 1 1 1 1 1 1 1 1
50 0 0 0 0 0 0 0 0 0 0

86
 ATTRIBUTE GAGE CALCULATIONS FOR “COUNTS”
Part A-1 A -2 A-3 B- 1 B-2 B-3 C-1 C-2 C-3 Reference
1 1 1 1 1 1 1 1 1 1 1
2 1 1 1 1 1 1 1 1 1 1
3 0 0 0 0 0 0 0 0 0 0
4 0 0 0 0 0 0 0 0 0 0
5 0 0 0 0 0 0 0 0 0 0
6 1 1 0 1 1 0 1 0 0 1
7 1 1 1 1 1 1 1 0 1 1
8 1 1 1 1 1 1 1 1 1 1
9 0 0 0 0 0 0 0 0 0 0
10 1 1 1 1 1 1 1 1 1 1
11 1 1 1 1 1 1 1 1 1 1
12
13
0
1
0
1
0
1
0
1
0
1
0
1
0
1
1
1
0
1
0
1 • There are 31 times where
14
15
16
1
1
1
1
1
1
0
1
1
1
1
1
1
1
1
1
1
1
1
0
1
1
0
1
1
1
1
1
A-3 = 1 and B-3 = 1
(that is of the 50 parts checked there
1 1
17 1 1 1 1 1 1 1 1 1 1
18 1 1 1 1 1 1 1 1 1 1
19
20
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1 were 31 matches by A and B on their
THIRD check)
21 1 1 0 1 0 1 0 1 0 1
22 0 0 1 0 1 0 1 1 0 0
23 1 1 1 1 1 1 1 1 1 1
24 1 1 1 1 1 1 1 1 1 1
25 0 0 0 0 0 0 0 0 0 0
26 0 1 0 0 0 0 0 0 1 0
27 1 1 1 1 1 1 1 1 1 1
28 1 1 1 1 1 1 1 1 1 1
29 1 1 1 1 1 1 1 1 1 1
30 0 0 0 0 0 1 0 0 0 0
31 1 1 1 1 1 1 1 1 1 1
32 1 1 1 1 1 1 1 1 1 1
33 1 1 1 1 1 1 1 1 1 1
34 0 0 1 0 0 1 0 1 1 0
35 1 1 1 1 1 1 1 1 1 1

• Total : where A-x =1 and B-x =1

36 1 1 0 1 1 1 1 0 1 1
37 0 0 0 0 0 0 0 0 0 0
38 1 1 1 1 1 1 1 1 1 1

= 34+32+31= 97
39 0 0 0 0 0 0 0 0 0 0
40 1 1 1 1 1 1 1 1 1 1
41 1 1 1 1 1 1 1 1 1 1
42 0 0 0 0 0 0 0 0 0 0
43 1 0 1 1 1 1 1 1 0 1
44 1 1 1 1 1 1 1 1 1 1
45 0 0 0 0 0 0 0 0 0 0
46 1 1 1 1 1 1 1 1 1 1
47 1 1 1 1 1 1 1 1 1 1
48 0 0 0 0 0 0 0 0 0 0
49 1 1 1 1 1 1 1 1 1 1
50 0 0 0 0 0 0 0 0 0 0

87
 Hypothesis Test Analyses
A * B Crosstabulation

B
.00 1.00 Total
A .00 Count 44 6 50
Expected Count 15.7 34.3 50.0
1.00 Count 3 97 100
Expected Count 31.3 68.7 100.0
Total Count 47 103 150
Expected Count 47.0 103.0 150.0

26-01-2024 Quality Solution

88
 Count & Expected Count Calculations

A * B Crosstabulation

B
.00 1.00 Total
A .00 Count 44 6 50
Expected Count 15.7 34.3 50.0
1.00 Count 3 97 100
Expected Count 31.3 68.7 100.0
Total Count 47 103 150
Expected Count 47.0 103.0 150.0

26-01-2024 Quality Solution 89

 "EXPECTED COUNTS"

The expected counts is calculated using the the following formula (based on Chi-
Square)

Expected Count = Column Total x [Row Total/Grand Total]

From the A*B Cross tabulation Table

97 from previous
Column Total = 103 slide
Row Total = 100 A * B Crosstabulation
The Grand Total = 150
B
.00 1.00 Total
A .00 Count 44 6 50
Expected Count 15.7 34.3 50.0
1.00 Count 3 97 100
Expected Count 31.3 68.7 100.0
Total Count 47 103 150
Expected Count
Hence 47.0 103.0 150.0

For A=1 and B=1

26-01-2024
the Expected Count = 103 x [100/150] =
Quality Solution
68.790
 A * B Crosstabulation

B
.00 1.00 Total
A .00 Count 44 6 50
Expected Count 15.7 34.3 50.0
1.00 Count 3 97 100
Expected Count 31.3 68.7 100.0
Total Count 47 103 150
Expected Count 47.0 103.0 150.0
po  pe
kappa 
1 pe
where po = Sum of the observed proportions in the diagonal cells (left to right direction)
pe = Sum of the expected proportions in the diagonal cells (left to right direction)
po  4497  0.94 pe 15.7  68.7  0.56
150 150
po  pe 0.94  0.56
kappa    0.86
1  pe 1  0.56
91
Miss Rate & False Alarm Rate For Appraiser “A”

• Miss Rate1: Calling a

“BAD” part GOOD.
= 3/48 = 6.3%

• False Alarm Rate2:

Calling a “GOOD”
part BAD.
= 5/102 = 4.9%
1 Type II error, Consumer’s risk
2 Type I error, Producer’s risk
92
Attribute Gauge R&R - Results

• The target effectiveness is always 100%

• Possible Corrective Actions include:

• Operator Training
• Clarification of Standards
• Simplification of Standards
• Conversion to Variable Data

26-01-2024 Quality Solution 93

93
 Measurement Systems Analysis - Summary
• Measurement errors can account for a large
proportion of the variation in our measures (y’s)
• We must evaluate our measurement systems
before assessing process stability or process
capability
• Errors in measurement systems can come from a
variety of sources
• Action should be taken to improve the capability of
our measurement systems if they are found to be
inadequate

26-01-2024 Quality Solution 94

 Appendix - ANOVA

26-01-2024 Quality Solution 95

 ANOVA Table - Construction
Construction of an Analysis of Variance (ANOVA)
table requires the following:

1. Identification of the Sources (Components) of Variation

2. Calculation of the Sum of Squares due to each Source of
Variation
3. Assignment of the appropriate Degrees of Freedom
4. Calculation of the Mean Squares
5. Calculation of the F-Ratio

26-01-2024 Quality Solution 96

 1. Components of Variation
Analysis of Variance (ANOVA) allows the decomposition of the
variability in the Gauge R&R study.

The components of variation in the Gauge R&R study are:

2Part = Variation due to the different parts

2Operator = Variation due to different operators
2Operator x Part = Variation due to the interaction between
operator and part
2Repeatability = Variation due to gauge repeatability
2Total = 2Part + 2Operator + 2Operator x Part + 2Repeatability

26-01-2024 Quality Solution 97

 2. Calculation of the Sum of Squares
The total sum of squares is calculated as follows:

SSTotal   y  y    y 
2  y2
2

n
Strictly speaking the sum of squares column is the sum of
squares around the mean, known as the corrected sum of
squares. We always use the corrected sum of squares when
estimating variation.
 y 0.65  0.60  1.00  1.00  .............  0.80  48.45
 y 2
 0 .65 2
 0 .60 2
 1 .00 2
 1 .00 2
 .......... .  0. 80 2
 41.3725

SSTotal y 

2 y
2

 41.3725 
48.45
2
 41.3725  39.1234  2.2491
n 60

26-01-2024 Quality Solution 98

 Calculation of the Sum of Squares
The sum of squares due to parts is calculated as follows:

 P   P   P   .........  P   y 
2 2 2 2 2

SS Part  1 2 3

10
np n

SS Part 
3.40   6.05  4.80   .........  4.00  48.45
2 2 2 2

2

6 60

SS Part  41.1821  39.1234  2.0587

Where:
P1, P2, P3…..P10 are the Sums for each Part
i.e the Sum of the 6 measurements made on each part.
np is the number of individual measurements of each part.

26-01-2024 Quality Solution 99

 Calculation of the Sum of Squares
The sum of squares due to operators is calculated as follows:

O   O   O   y 
2 2 2 2

SS Operator  1 2
 3
no n

SS Operator 
16.55   15.35   16.55  48.45 
2 2 2


2

20 60

SS Operator  39.1714  39.1234  0.0480

Where:
O1, O2, O3 are the Sums for each Operator
i.e the sum of the 20 measurements made by each operator.
no is the number of measurements made by each operator.

26-01-2024 Quality Solution 100

 Calculation of the Sum of Squares
The sum of squares due to the interaction between operators
and parts is calculated as follows:

O1 P1   O1 P2   ..........O3 P10   y 

2 2 2 2

SSOperatorPart    SSOperator  SS Part

nO P n

SSOperatorPart 
1.25  2.00  ........1.65 48.45
2 2 2

2
 0.0480  2.0587
2 60

SSOperatorPart  41.3338  39.1234  0.0480  2.0587  0.1037

Where:
O1P1, O1P2,…….O3P10 are the Sums for each Operator & Part Combination
i.e the sum of the 2 measurements made by each operator on each part.
nOxP is the number of measurements made by each operator on each part.

26-01-2024 Quality Solution 101

 Calculation of the Sum of Squares

The sum of squares due to repeatability is obtained by

subtraction:
SS Repeatability  SSTotal  SS Part  SS Operator  SS OperatorPart

SS Repeatability  2.2491  2.0587  0.0480  0.1037  0.0387

26-01-2024 Quality Solution 102

 Calculation of the Sum of Squares

Source of Variation Sum of Squares

Between Parts 2.0587

Between Operators 0.0480
Operator x Part 0.1037
Repeatability 0.0387

Total 2.2491

26-01-2024 Quality Solution 103

 3. Degrees of Freedom
Degrees of Freedom is a statistical concept relating
to the number of paired comparisons required to
distinguish between items.

For example, we need to find the tallest person out of

3 people. 2 comparisons would be required:

Person 1 v Person 2
Tallest v Person 3

We would then know who the tallest person is.

26-01-2024 Quality Solution 104

 Rules for Degrees of Freedom

The following rules apply to Degrees of Freedom:

DF for a Factor (Main Effect) = (Number of Levels) – 1

DF for interactions = Product of the DF of the Factors involved

DF for Repeatability = (Product of Factor Levels) x (Repeats – 1)

Total DF = (Number of Individual Results) - 1

26-01-2024 Quality Solution 105

 Degrees of Freedom
Degrees
of
Source of Variation Sum of Squares Freedom

Between Parts 2.0587 9

Between Operators 0.0480 2
Operator x Part 0.1037 18
Repeatability 0.0387 30

Total 2.2491 59

26-01-2024 Quality Solution 106

 4. Calculation of the Mean Squares
The Mean Square is calculated as follows:

Mean Square = (Sum of Squares) / (Degrees of Freedom)

Source of Variation Sum of Squares DF Mean Square

Between Parts 2.0587 9 0.2287

Between Operators 0.0480 2 0.0240
Operator x Part 0.1037 18 0.0058
Repeatability 0.0387 30 0.0013

Total 2.2491 59

26-01-2024 Quality Solution 107

 5.
Source of Variation
Calculation of the F-Ratio
Sum of Squares DF Mean Square F-Ratio

Between Parts 2.0587 9 0.2287 39.43

Between Operators 0.0480 2 0.0240 4.14
Operator x Part 0.1037 18 0.0058 4.46
Repeatability 0.0387 30 0.0013

Total 2.2491 59

The F-Ratio is used to test the significance of each source of

variation.
F-Ratio for Parts = (MSParts) / (MSOperator x Part)
F-Ratio for Operators = (MSOperators) / (MSOperator x Part)
F-Ratio for Operator x Part = (MSOperators x Parts) / (MSRepeatability)

26-01-2024 Quality Solution 108

 Estimating Components of Variation
The Mean Square column is expected to contain the following
components of variation. This expected mean square is only applicable
to this current study, where we have 3 operators, 10 parts and 2 repeat
measurements. For other studies, the number of the components will
change. (Fortunately, Minitab can do this for us!)

Source Mean Square Expected Mean Square

Parts 0.2287
6 2
Part
 2 2Operator  Part2Repeatability
Operators 0.0240
20 2
Operator
 22OperatorPart2Repeatability
Operator x Part0.0058
Repeatability
22OperatorPart 2Repeatability
2
0.0013 Repeatability
26-01-2024 Quality Solution 109
Source Estimating Components
Mean Square of Mean
Expected Variation
Square
Parts 0.2287
6 2Part  2Operator
2

 Part 2
Repeatabil ity
Operators 0.0240
20 2
Operator
 2Operator
2
 2
 Part Repeatability
Operator x Part 0.0058
2 2
Operator Part
 2Repeatability
Repeatability 0.0013
2Repeatability
Repeatabil
2
ity
 0 .0013
2
2 Operator Part2
Repeatability
 0.0058
2 Operator  Part 0.0058 0.00.13  0.0045
2

2 0.00225
Operator X Part
26-01-2024 Quality Solution 110
Source Mean Square
ExpectedofMean
Estimating Components Square
Variation
Parts 0.2287
Operators 0.0240
6 2  2 2  2
Operator x Part 0.0058
Part Operator  Part Repeatability
Repeatability 0.0013
20 2  2 2  2
Operator Operator  Part Repeatability

2 2  2
Operator  Part Repeatability

2
Repeatability

20 2  2 2  2  0 . 0240
Operator Operator  Part Repeatability

20  2  0 . 0240  2 2   2
Operator Operator  Part Repeatability

0.00240-2(0.00225)-0.0013  0.0091
 2 
Operator 20

26-01-2024 Quality Solution 111

Source Estimating Components
Mean Square of Variation
Expected Mean Square
Parts 0.2287 6 2Part  2 Operator
2
Part  2
Repeatability
Operators
0.0240 20 2
 2 Operator
Operator
2
Part  2
Repeatabili
Operator x Part ty
0.0058 2 2
Part
Operator
 2
Repeatability
Repeatability
0.0013  2
Repeatability

6 Operator OperatorPart Repeatability 0.02287

 
2 2 2
 2
2
6part  0 .02287 2

2 OperatorPart 2
Repeatability

2
Part 0.002287-2(0.00225)-0.0013  0.03715
6
26-01-2024 Quality Solution 112
 Estimating Components of Variation
  0.03715
2
Part

 2
Operator
 0.00091
 2
Part
Operator
 0.00225
2Repeatability  0.00130
2
Total  2Part  Operator
2
 Operator
2
Part  2
Repeatability
2
Total 0.037150.000910.002250.001300.04161
We have established estimates of each of the components of
variation!

26-01-2024 Quality Solution 113

 Variance Component Estimates
Part-to-Part
Variation
0.03715 Operator
Overall 0.00091
Variation
Reproducibility
0.04161
0.00316 Operator
Measurement by part
System Interaction
Variation 0.00225
0.00446
Repeatability
0.00130
Variances are additive!

26-01-2024 Quality Solution 114