Msa

MEASUREMENT
SYSTEM ANALYSIS
BY-DIVYANSHU MISHRA
ROHIT ARORA
TOPICS
Introduction
Terminology in Measurement System
Gage R and R Study
Measurement System
Measurement System is the collection of

instruments or gages, methods and
personnel used to quantify a unit of
measure to the feature characteristic
being measured i.e. the complete process
used to obtain measurements.
Importance of Measurement
System?
Most of the decisions in the quality are
made on the basis of data which we get
from measurements of output of a
process.
Output of a Measuremen
Data
process t System
If it contains
error then it
will be
reflected in
data
Variation in Data
Whenever we measure anything, the

variation that we observe can be
segmented into the components as shown
in figure:
What is MSA?
Measurement system analysis(MSA) is an

experimental and mathematical method of
determining how much the variation within
the measurementprocess contributes to
overall process variability
Other Basic Concepts
Measurements Accuracy-Measurements are said to be
accurate if their tendency is to centre around the actual
value of the entity being measured.
Measurements precision-Measurements are precise if they
differ from each other by a small amount
Terminology in Measurement System
TERMS RELATED TO ACCURACY
Bias-Difference between the observed average of measurements
and the true value
Linearity-The change in bias over the normal operating range.
Instability-The change in bias over time
TERMS RELATED TO PRECISION

Repeatability-Variation in measurements obtained with one
measuring instrument when used several times by an appraiser
while measuring the identical characteristic on the same part
Reproducibility-Variation in the average of the measurements
made by different appraisers using the same gage when
measuring a characteristic on one part
Terminology in Measurement
System
Gage R and R Studies
Measurement variability can be broken down as
Varmeasurement error = Varrepeatability + Varreproducibility.
This study establish repeatability and

reproducibility components of measurement
variability.
Measurement Capability indices- P/T and %GR&R

are calculated by this study
How to conduct Gage R & R
studies
In conducting a Gage R&R study, we need to
identify # parts, # trials per part, and # operators.
We also need tolerance width for each feature.
Tolerance Width = USL LSL
USL ~ Upper Spec Limit and LSL ~ Lower Spec Limit.
Commonly Used
5 or 20 parts
2 or 3 trials
2 or 3 operators
(parts x trials x operators):
Example: 5x3x2 Two operators will measure each
of 5 parts three times.
11
Calculations of repeatability and
reproducibility
The standard deviation due to
repeatability is given as
repeatability = RR / d2
RR = (RR1 + RR2 +.+ RRp ) /p
where p is no of operators
d2 calculated from chart and depend on
number of trials
Calculations of repeatability and
reproducibility
The standard deviation due to
reproducibility is given as
reproducibilty= RxRR / d2
RxRR = XRmax XRmin
XRmax = max(XR1 , XR2 .. , XRp )
XRmin = min(XR1 , XR 2 .. , XR p )
d2 depends on number of operators.
Calculations of example
RR=(RR1 +RR2 +RR3 )/3= (1+1.25+1.2)/3 =1.15

RxRR = XRmax XRmin = 22.60-22.28 =0.32
reproducibilty= RxRR / d2 = 0.32/1.693=0.19
repeatability = Rbar / d2 = 1.15/ 1.128 = 1.02
ms= 0.192 + 1.022 = 1.08
Measurement Capability
Index(GR&R)
%R&R= ms / observed process variation
Addresses what percentage of observed

process variation is taken up by
measurement error.
As a target industries look for %GR&R less

than 30%
Measurement Capability
Index(P/T)
Precision to Tolerance Ratio
P/T=5.15*ms / Tolerance
ms includes both repeatability and reproducibility.
Tolerance= USL-LSL
It addresses what percent of the tolerance is
taken by total measurement error ms .
Best case:10% Acceptable:30%
The use of 5.15 is industry standard
How to improve gage
capability?
Repeatability
Gage maintenance
Gage redesign to better fit application
Reproducibility
Operator training
More clearly define measurement scale available
to the operator
Gage R and R Studies with
ANOVA
It also considers interaction b/w part and
operator. So reproducibility is further broken down
as variation due to operator and variation due to
part and operator interaction.
Sum of Squares Concept
Sample variance=(((xi xR)^2)/n-1)
By applying some arithmetic operation , we
get
Sample variance= (n*xi2 (x)^2/n(n-1))
=(xi2 ((x)^2/n))/n-1)
Numerator(xi2 ((x)^2/n)) is known as sum
of squares
Denominator n-1 is degrees of freedom
In ANOVA (x)^2/n is known as correction
factor
We will solve this problem using
ANOVA
STEP 1
Find correction factor.

C.F.= Grand Sum/(parts x trials x
operators)
Grand Sum=all trials
In this example
Grand Sum= 21+20+ 24+..
+19+17=2687
C.F.=(2687)^2/120=60166.4083
STEP 2
Find Total sum of squares

SST= (All trials)^2 Correction factor
In this example
(All trials)^2 = (21^2)+(20^2)+ ..
+(17^2)=61441
SST =61441-60166.4083=1274.5917
STEP 3
Find sum of square for parts
SSA =((Find row total)^2))/(NO.OF OPERATORS*NO.OF
trials ) C.F.
In this example
((Row Total)^2)/6=( (SUM OF ALL trials FOR PART
1)^2+.+ (SUM OF ALL trials FOR PART 20)^2 )/6
=(1212 + 1422 ++ 1092 )/ 6=61311
SSA = 61311-60166.4083 =1144.5917
STEP 4
Find sum of square for operators
SSB =((Find the column sum)^2))/(NO.OF
PARTS*NO.OF trials ) C.F.
In this example
((SUM OF ALL trials FOR OPERATOR X)^2)/40=( (SUM
OF ALL trials FOR OPERATOR A)^2+.+ (SUM OF
ALL trials FOR OPERATOR C)^2 )/40
=(8922 + 8912 + 9042 )/40=60169.025
SSB =60169.025-60166.4083=2.6167
STEP 5
Find total of trials of each part and thus get a new table
PART OPERATO OPERATO OPERATO

RA RB RC
1 41 40 40
2 47 48 47
3 41 40 42
4 54 54 55
5 37 37 39
6 44 45 45
7 43 46 42
STEP 6
Find Total sum of squares due to new table

SSTI= ((All trials)^2)/2 Correction factor
In this example
((CELLS^2 ))/2= ((41^2)+(40^2)+ ..
+(36^2))/2=61381.5
SSTI =61381.5-60166.4083=1215.0917
STEP 7
Find sum of square for part*operator

interaction
SSAB= SSTI (SSA + SSB)
In this example
SSAB =1215.0917-(1144.5917+2.6167)
=67.8833
STEP 8
Find sum of square for repeatability

SSE =SST - SSTI
In this example
SSE =1274.5917-1215.0917
=59.5
FINAL ANOVA TABLE
SOURCES OF DEGREES SUM OF MEAN SUM OF
VARIATION OF SQUARES SQUARE
FREEDOM
PART 19 1144.5917 60.241
OPERATOR 2 2.6167 1.30835
PART*OPERATOR 38 67.8833 1.785
INTERACTION
REPEATABILITY 60 59.5 0.991666
TOTAL 119 1274.5917
Interpreting Result
% contribution of Variance component(up
to 9 percent is acceptable)
SOURCES OF Variance % Variance
VARIATION Component component
PART 60.241 93.649
OPERATOR 1.30835 2.033
PART*OPERATOR 1.785 2.774
INTERACTION
REPEATABILITY 0.991666 1.646
Total Gage R and R 4.08501 6.35
TOTAL Variation 64.32601 100
Interpreting Result
% std variation(up to 30 % is acceptable)
Sources of Variation Std Dev %Std Dev

Part 7.761 96.770
Operator 1.143 14.25
Part*operator 1.336 16.658
interaction
Repeatability 0.99582 12.416
Total Gage R and R 2.0211 25.200
Total variation 8.020 100
THANK YOU

Msa

Uploaded by

Copyright:

Available Formats

Msa

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Msa

Uploaded by

Copyright:

Available Formats

MEASUREMENT

Measurement System is the collection of

Whenever we measure anything, the

Measurement system analysis(MSA) is an

TERMS RELATED TO PRECISION

This study establish repeatability and

Measurement Capability indices- P/T and %GR&R

RR=(RR1 +RR2 +RR3 )/3= (1+1.25+1.2)/3 =1.15

Addresses what percentage of observed

As a target industries look for %GR&R less

Find correction factor.

Find Total sum of squares

PART OPERATO OPERATO OPERATO

Find Total sum of squares due to new table

Find sum of square for part*operator

Find sum of square for repeatability

% std variation(up to 30 % is acceptable)

Sources of Variation Std Dev %Std Dev

You might also like