To measure or to guess
An Introduction into
Capability Analysis
&
Measurement System Analysis
Albert Eisele
MSA-Introduction-ASQ
What we see / Total Variation
Process A
2
Observed
(Total)
Process B
Total Variability
(Observed variability)
Which process is better?
Albert Eisele
MSA-Introduction-ASQ
Pg 2
Influence of the Measurement System
Process A
2
Measurement
System
2
Observed
(Total)
Process B
Measurement
Variability
Total Variability
(What we see)
Which process is better?
Albert Eisele
MSA-Introduction-ASQ
Pg 3
The whole picture!
Process A
Measuring
2
Measurement
2
Part
/
Real Process
System
2
Observed
(Total)
Process B
Guessing
Part Variability
(What we produce)
Measurement
Variability
Total Variability
(What we see)
Which process is better?
Albert Eisele
MSA-Introduction-ASQ
Pg 4
We always have an error.
What is the error due to our measurement
system? and
Is our measurement error within the agreed
boundaries?
There are 2 specific sources leading to an error:
Accuracy
Resolution
Bias
Linearity
Stability
Precision
Repeatability
Reproducibility
Albert Eisele
MSA-Introduction-ASQ
Accurate
Precision:
Accuracy:
Accuracy
The spread or Variance
of measured values
when measuring the
values
Difference between the
observed average value
of the measurements
and the master value
Albert Eisele
Precise
MSA-Introduction-ASQ
Accuracy:
RESOLUTION
Alias: smallest readable unit,
measurement resolution,
detection limit.
Resolution or discrimination is
Capability to detect the smallest
significant/tolerable changes in
order to detect present variation!
10 to 1 rule of thumb:
Increments in the measurement
system should be one9
one9tenth of
the product specification or
process variation!
Albert Eisele
MSA-Introduction-ASQ
Accuracy:
BIAS
Bias is the difference between
the true value (the reference
value) and the observed
average of measurements of
the same characteristic on the
same part.
Bias is the measure of the
systematic error of the
measurement system.
system
Bias can be minimized by the
use of calibration procedures!
Albert Eisele
MSA-Introduction-ASQ
Accuracy:
LINEARITY
The difference of bias
throughout the expected
operating (measurement) range
of the equipment is called
linearity.
Linearity can be thought of as a
change of bias with respect to
size!
Does my gauge have the same
accuracy for all sizes of objects
being measured?
Albert Eisele
MSA-Introduction-ASQ
Accuracy:
STABILITY
Stability (or drift) is the total
variation in the measurements
obtained with a measurement
system on the same master or
parts when measuring a single
characteristic over an extended
time period.
That is, stability is the change in
bias over time!
Albert Eisele
MSA-Introduction-ASQ
10
Precision:
Reference Value
REPEATABILITY
Repeatability is traditionally
referred to as the within
appraiser variability.
Repeatability is the variation in
measurements obtained with
one measurement instrument
by one appraiser while
measuring the identical
characteristics on the same
part!
Repeatability is the within9
within9
system9
system9variation!
Albert Eisele
MSA-Introduction-ASQ
11
Precision: REPRODUCIBILITY
Reproducibility is traditionally
referred to as the between
appraiser variability.
Reference Value
Reproducibility is typically defined
as the variation in the average of
the measurements made by
different appraisers using the
same measuring instrument
when measuring the identical
characteristics on the same part!
Repeatability is the between9
between9
systems9
systems9variation!
Albert Eisele
MSA-Introduction-ASQ
12
Summary:
Albert Eisele
Components of variation
MSA-Introduction-ASQ
13
Statistical Output: Requirements
Gage R&R
Decision
Comments
Under 10
percent
Generally considered to
be an acceptable
measurement system.
Recommended, especially useful when trying to
sort or classify parts or when tightened process
control is required.
10 percent to
30 percent
May be acceptable for
some applications.
Decision should be based upon, for example,
importance of application measurement, cost of
measurement device, cost of rework or repair.
Should be approved by the customer.
Over 30
Percent
Considered to be
unacceptable.
Every effort should be made to improve the
measurement system.
This condition may be addressed by the use of
an appropriate measurement strategy; for
example, using the average result of several
readings of the same part characteristic in order
to reduce final measurement variation.
Measurement Systems Analysis (MSA)
AIAG Reference Manual, Fourth Edition, 2010 page 78
Chrysler Corporation, Ford Motor Company, General Motors Corporation.
Albert Eisele
MSA-Introduction-ASQ
14
Statistical Output: Requirements
Measurement System Standard Deviation
(R&R) as a percentage of Total Observed
Process Standard Deviation.
Includes both repeatability and reproducibility.
9 Less than 10% is ideal
9 10930% depends on the situation
This number represents the number of non9
overlapping confidence intervals that will span
the range of product variation. For example, if
the number of distinct categories was three,
the Measurement System could only put
things into small9medium9large buckets.
Albert Eisele
MSA-Introduction-ASQ
15
Statistical Output: Example
Albert Eisele
MSA-Introduction-ASQ
16
Practical Exercise:
Capability Analysis
Albert Eisele
MSA-Introduction-ASQ
17
Practical Exercise: Capability Analysis
Tolerance:
Measurement:
10,0 +/9
+/9 1 mm
Sample of 50 values
OK or NOK?
Albert Eisele
MSA-Introduction-ASQ
18
Practical Exercise: Capability Analysis
Tolerance:
Measurement:
10,0 +/9
+/9 1 mm
Sample of 50 values
Min
Max
OK?
Albert Eisele
MSA-Introduction-ASQ
19
Practical Exercise: Capabilty Analysis
OK or NOK?
Analysis using Minitab
Albert Eisele
MSA-Introduction-ASQ
20
Practical Exercise: Capability Analysis
Tolerance:
10 +/9
+/9 1m
Min: 9,1144
Max: 10,469
Seems to be ok?
Albert Eisele
MSA-Introduction-ASQ
21
Practical Exercise: Capability Analysis
Tolerance:
10 +/9
+/9 1m
StDev:
StDev: 0,3245
Min:
9,1144
Max:
10,469
Whats about the Standard Deviation?
Albert Eisele
MSA-Introduction-ASQ
22
Practical Exercise: Capability Analysis
Recall the properties of the Standard Normal Distribution:
99.7%
95.5%
68.3%
-6
-5
-4
-3
-2
-1
+1
+2
+3
+4
+5
+6
68.27 % of the data will fall within +/- 1 Standard Deviation
95.45 % of the data will fall within +/- 2 Standard Deviations
99.73 % of the data will fall within +/- 3 Standard Deviations
99.9937 % of the data will fall within +/- 4 Standard Deviations
99.999943 % of the data will fall within +/- 5 Standard Deviations
99.9999998 % of the data will fall within +/- 6 Standard Deviations
Albert Eisele
23
MSA-Introduction-ASQ
23
Practical Exercise: Capability Analysis
Process Performance over time:
Cpk = 0,61 Ppm = 33.555 = 3,36% nok
Result:
Result: 66 nok Units/day
Units/day (based on 1850)
Albert Eisele
MSA-Introduction-ASQ
24
Capability Analysis & Sample Size
Albert Eisele
MSA-Introduction-ASQ
25
Capability Analysis & Sample Size
Rule of Thumb based on our example
Sample
Size
Tolerance
Mean
5
10
30
50
+/- 1 mm
+/- 1 mm
+/- 1 mm
+/- 1 mm
9,71 +/- 0,5 mm
9,71 +/- 0,3 mm
9,70 +/- 0,15 mm
9,77 +/- 0,1 mm
Albert Eisele
Confidence
Interval
for Mean
MSA-Introduction-ASQ
Blocked Available
Tolerance Tolerance
[%]
[%]
50%
30%
15%
10%
50%
70%
85%
90%
26
Q&A
Albert Eisele
MSA-Introduction-ASQ
27
