Vous êtes sur la page 1sur 29

IOE 466 W08

Gage and Measurement System Analysis

1
Topics

 Measurement Systems Analysis


 Gage R&R for Variable data

 Attribute Gage R&R

 Case Study

2
Measurements Systems Analysis
 Purpose:
 Determine how much variability is due to the gage

or instrument
 Isolate the components of variability of the

measurement system
 Assess whether the instrument or gage is capable

(suitable for intended application)

3
Components of a Measurement System

 Equipment or Gage
 Type of Gage:
 Attribute: go-no go, vision systems (part present or not present)
 Variable: calipers, probe, tape measure, coordinate measurement machines, checking
fixture with inspection device

 Discrimination of Measurement – General Rules:


 At least 1/10 of tolerance (tol = 1 mm, measure to at least 0.1)
 Or, at least 1/10 of 6*process standard deviation (6s)

 Operator & Operating Instructions


 Part locating or orientation scheme
 gage must be able to consistently locate the part being measured.

4
Gage R&R

 Total variability decomposition

s2total  s2product  sgage


7
2

s2measuremen t _ error  sgage


2
 s2repeatability  s2reproducibility

inherent precision of gage different operators or conditions


Gage R&R
 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.

 Common Applications (parts x trials x operators):


 5 or 10 parts
 2 or 3 trials
 2 or 3 operators
 Example: 5x3x2  Two operators will measure each of 5 parts three
times.

6
Gage Capability Criteria

 Precision to tolerance ratio or P/T ration


P 6sˆ gage
  0.1
T USL  LSL

 gage error as a percentage of the product variability


sˆ gage
100%
sˆ product
7
Example 7-7 P354

s2total  s2product  sgage


2

R
sˆ gage 
d2

• X-bar chart represents variability between different product units


• R chart represents the gage measurement variability:
Gage R&R : Example 7-7

 Be careful! Don’t interpret this like you would a process control


chart.

Xbar-R Chart of M1, ..., M2  X-bar: out of control


30 1
points, show that
1

measurement system
Sample M ean

1
1
25
U C L=24.06
_
_
X=22.28 can discriminate
LC L=20.49
20

1
1 1
1
between units of
1 1
1 3 5 7 9 11
Sample
13 15 17 19
products
3 U C L=3.104
Sample Range

2  R-bar: in-control, show


1
_
R=0.95 that operators are
0 LC L=0 consistent.
1 3 5 7 9 11 13 15 17 19
Sample

9
7

Example 7-7: continued


Suppose that instead of having only 1 operator measure the parts,
you make 3 operators measure each part twice.
7

Example 7-7: Gage R&R

(1) average of all ranges


1 1 R 1.15
R  (R 1  R 2  R 3 )  (1  1.25  1.2)  1.15 sˆ repeatability    1.02
3 3 d 2 |n  2 1.128
Rx 0.32
R x  x max  x min  22.60  22.28  0.32 sˆ reproducibility    0.19
d 2 |n 3 1.693
x max  max( x1 , x 2 , x 3 )
(2) Difference among operators
x min  min( x1 , x 2 , x 3 )

(3) Each operator’s average

11
Gage and Measurement System Capability
 Variation Decomposition

s2total  s2product  sgage


2
 s2measuremen t  sgage
2
 s2repeatability  s2reproducibility

r m n

 ( x
k 1 i 1 j 1
kij  x )2
s total
2

rmn  1
r m n

 x kij RX
x
k 1 i 1 j 1 sˆ reproducibility 
R d2
rmn s
ˆ repeatability 
d2
r R X  x max  x min ;
R
Use R chart for estimation
r: # of operators k x max  max( x1 , x 2, , x r )
m: # of samples
R k 1

m
r x min  min( x1 , x 2, , x r )
n: # of repeated measurements
R ki m m n
xkij :
i: sample index
Rk  i 1
m x ki  x
i 1 j1
xij

R ki  max j ( x kij )  min j ( x kij )


xk  i 1

j: repeated measurement index m mn
k: operator index
Gage and Measurement System Capability (Cont’s)

 Gage capability: precision-to-tolerance ratio (P/T ratio)


 Generally, an adequate gage capability: P/T0.1
P 6sˆ gage

T USL  LSL
 gage variability-to-product variability ratio
 independent of specification limits
sˆ gage
100%
sˆ product

s 2total
s2product  s2total  sgage
2

s 2repeatability sgage
2
 s2repeatability  s2reproducibility s2product
sgage
2

s 2reproducibility
Gage R&R for Attribute Variables

 Some quality inspection systems rely on human


judgment – “good/bad” or “best/good/poor”
 Examples
 Fabric color matching

 Contact Lens appraisal

 Delamination (printing)

 How can we test whether the measurement system is


working accurately?

14
Gage R&R for Attribute Variables

 Gage R&R Study set up steps


 Select 20-30 product samples (include mix of

“good” and “bad” parts)


 Identify # of parts, # of inspectors and # of trials

 Have a master appraiser (expert) rate each part

 Inspectors rate each part an ‘x’ number of trials, at

random, without knowing the master results

15
Gage R&R for Attribute Variables
Then:

# of measuremen t matches within tr ials


Operator Repeatabil ity n 
number of parts inspected

 Operator Repeatabil ity n


Overall System Repeatabil ity  i 1
n

# of matches with standard


Individual Effectiven ess 
number of parts inspected

# of times all operators agree with standard


Overall System Effectiven ess 
number of parts inspected
16
Gage R&R for Attribute Variables

 General Guideline: 90% effectiveness is acceptable

 Next steps:
 Identify best measurement system procedure

 Document standardized work

 Train all operators in new system

 Periodically check gage R&R of system

17
Gage R&R for Attribute Variables: Example

A hospital is trying to evaluate the consistency of their doctors in rating


mammograms. Each mammogram is rated according to the following
scale:

1 – No cancer (best)
2 – Benign cancer
3 – Possible malignancy
4 – Malignancy (worst)

A sample of 15 mammograms is collected, and three randomly selected


doctors within that specialty are selected. Each doctor rates each
mammogram three times at random. In the study, these ratings will also
be compared to a standard (ratings provided by a panel of senior
doctors).
18
Gage R&R for Attribute Variables: Example

1 No cancer
2 Benign cancer
3 Possible malignancy
4 Malignancy

Mammogram Standard Doctor 1 Doctor 2 Doctor 3


1 4 4 4 4 4 4 4 4 4 3
2 4 4 4 4 4 4 4 4 4 4
3 2 2 2 2 2 2 2 2 2 2
4 3 3 3 3 3 3 3 2 3 2
5 2 2 2 2 2 2 2 1 2 2
6 1 1 1 1 2 2 1 2 1 2
7 3 3 3 3 3 3 3 2 2 3
8 4 4 4 4 4 4 3 4 4 4
9 4 4 4 4 4 4 4 3 3 4
10 2 2 1 1 2 2 2 2 2 2
11 2 2 2 2 2 2 2 2 2 1
12 4 4 4 4 4 4 4 4 4 4
13 1 2 2 2 1 1 1 1 2 1
14 1 1 1 1 1 1 1 1 1 2
15 3 3 3 3 3 3 2 4 4 4

19
Gage R&R for Attribute Variables: Example

 Results

Individual
Repeatability
Effectiveness
Doctor 1 93.3% 93.3%
Doctor 2 80.0% 93.3%
Doctor 3 40.0% 80.0%

 System Repeatability = 71.1%


 Overall Effectiveness = 87.7%

20
Case Study:
Improving Data Reliability for Valve Bodies

 Need to adequately measure bore diameter data.


Excessive variation is causing rejects from process.
 Suspected that data for water valve bodies not reliable
 Critical measurement is the bore diameter, with a
specification of 1.334 +/- .002”

Bore diameter

21
Problem Definition

 Need to adequately measure bore diameter data.


 Excessive variation is causing rejects from process –
need to ensure diameter is measured properly
because of small tolerance for error.
 Currently utilizing a dial caliper method

 To find the current state of the process:


 10 x 3 x 3 Gage R&R experiment

22
Current State: Gage R&R results

23
Current State: Gage R&R results
 Appraiser variation takes up 58% of tolerance width
 Equipment variation takes up 69% of total variation

24
Current State: Cause and Effect Diagram

Cause-and-Effect Diagram
Measurements Material Personnel

Lack of training

V ariability betw een


operators
Variability in
bore
diameter
data

Improper use of Dial caliper not precise


caliper

Lack of standardised Dial caliper not accurate


w ork

Env ironment Methods Machines

25
Improvement alternatives

 Use different type of gage


 Plug-gages

 Internal calipers

 Self centering electronic bore gauge

 Gage R&R done for top two alternatives, internal


calipers and electronic bore gauge.

26
Self centering bore gauge: Gage R&R results

27
Self centering bore gauge: Gage R&R results
 Appraiser variation takes up 2.7% of tolerance width
 Equipment variation takes up 5.2% of total variation

28
Results

 Switch from Dial Caliper to Self Centering Bore Gage

 Reduced % of R&R compared to total variance


from 90.2% to 6.2%.

 Expected reduction in errors reported is 75%

29

Vous aimerez peut-être aussi