Date post: | 15-Dec-2015 |
Category: |
Documents |
Upload: | marlee-kealey |
View: | 213 times |
Download: | 0 times |
1111
VIII Measure - Capability and Measurement
2222
Success depends upon the ability to measure performance.
Rule #1: A process is only as good as the ability to reliably measure.
Rule #2: A process is only as good as the ability to repeat.
Gordy Skattum, CQE
Initial Thoughts
VIII-2
3333
It is impossible for us to improve our processes if our gaging system cannot discriminate between parts or if we cannot repeat our measurement values.
Every day we ask “Show me the data” - yet we rarely ask is the data accurate and how do you know?
Initial Thoughts
VIII-2
4444
Product Control◦ Detection, conformance to a design
specification
Process Control◦ Prevention, real-time control,
assessing a feature to its natural process variation
Purpose of Measurement
What is a measurement system used for?
What is a measurement system used for?
VIII-2
5555
AIAG – Automotive Industry Action Group◦ Collaboration between “Big 3” to create
one set of guidelines for all suppliers MSA Reference Manual –
Measurement System Analysis◦ Introduction to MSA◦ Covers normally occurring measurement
situations◦ Developed to meet specific needs of the
automotive industry Gage Repeatability and
Reproducibility◦ A study to understand the within-system
and between-system variation in a measurement system
◦ A comparison of standard deviation
AIAG/MSA/Gage R&R
VIII-2
6666
Discrimination ◦ In selecting or analyzing a
measurement system, we are concerned about the system’s discrimination, or the capability of the system to detect and faithfully indicate even small changes of the measured characteristic - also known as resolution.
◦ The smallest readable unit
Measurement Terms
100ths graduation decimal rule
VIII-4
7777
Accuracy vs. PrecisionPrecise (low variation)
Yes No
Yes
Acc
urat
e (o
n ta
rget
)N
o
VIII-4
8888
Bias◦ The difference between the
observed average of measurements and the reference value. The reference value, also known as the accepted reference value or master value, is a value that serves as an agreed-upon reference for the measured values. Bias is measured as “accuracy” or as “accuracy shift.”
Choices for addressing bias error◦ Calibrate the gage; adjust, correct,
or apply an offset ◦ Change the system (instrument,
condition, masters, …)
MSA Terms
valuereading trueaverageBias VIII-4
9999
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.
◦ The change in bias over time.
MSA Terms
Stability
Time 2
Time 1
VIII-4
10101010
Linearity◦ Linearity is the bias over the
operating range of a measurement system. This, along with bias, is checked as part of the calibration procedure.
Choices for addressing linearity◦ Calibrate, adjust the gage or build
offset table◦ Change the system (condition,
masters, …)
MSA Terms
VIII-4
11111111
Repeatability (EV)◦ Repeatability is the variation in
measurements obtained with one measurement instrument when used several times by one appraiser while measuring the identical characteristics on the same part. Includes all within-system variation.
MSA Terms
Repeatability
VIII-4
12121212
Reproducibility (AV)◦ Reproducibility is 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. Includes all between-system variation.
MSA Terms
Operator A
Operator B
Operator C
Repeatability
VIII-4
13131313
The sensitivity of a measurement system to detect process variation
Rules for determining effective resolution
1. Count the number of “0” plot points on the process range control chart. If >25%, then the gage lacks effective resolution
2. Count the gage discrimination levels between the UCL and the LCL of the process average control chart. If <5 levels, gage lacks effective resolution
Effective Resolution
VIII-4
14141414
Effective Resolution
Number of zeros?
Levels between UCL-LCL?
VIII-4
Individuals Chart
UCL=5.2633
LCL=-0.9433
CEN=2.16
-2
0
2
4
6
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
Moving R Chart
UCL=3.8115
LCL=0.0
CEN=1.1667
-1
0
1
2
3
4
5
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
15151515
A capability measure Typically we compare
◦ Gage R&R to tolerance◦ Gage R&R to process variation
Three levels of results◦ 0-10% ◦ 10-30% ◦ +30%
Gage R&R Levels
VIII-5
16161616
If you have a poor measurement system…
Difficult/impossible to make process improvements
Causes quality / cost / delivery / responsiveness problems
False alarm signals, increases process variation, loss of process stability
Improperly calculated control limits
Can make your processes worse!
Impact on SPC and 6s
VIII-5
17171717
Guidelines for Determining a Gage R&R
Since the purpose of the analysis of a measurement system is to understand the systems variation, the use of graphical tools is very important. Personal investigations have unveiled many powerful gage analysis software packages. • SPCXL (Sigma Zone)• Minitab
• All will deliver identical results*• SPCXL is easy to use and inexpensive. • Minitab is a complete statistical analysis
package which requires a lot of training.
VIII-7
18181818
Guidelines (cont.)
The four standard methods for analyzing measurement systems are:
• Range method (short form)• Average and Range (long form)• ANOVA (Analysis of Variance)• Attribute gage study (both short
and long form methods)
Each method has its advantages and disadvantages as well as limitations.
Refer to the AIAG MSA Reference Manual (Measurement Systems Analysis) for additional information.
VIII-7
23232323
Gage R&R Sheet Available on Website
GAGE R&R STUDYEQUIPMENT:DATE:
OPERATOR A NAME: OPERATOR B NAME: OPERATOR C NAME:
SAMPLE (n) 1ST TRIAL -B 2ND TRIAL -C 3RD TRIAL -D RANGE 1ST TRIAL - F 2ND TRIAL - G 3RD TRIAL - H RANGE 1ST TRIAL -J 2ND TRIAL - K 3RD TRIAL -L RANGE
1
2
3
4
5
6
7
8
9
10
TOTALS
SUM OF B,C,D SUM OF F,G,H SUM OF J ,K,L
AVE. X BAR A Rbar A AVE. X BAR B Rbar B AVE. X BAR C Rbar C
Rbar A
Rbar B
Rbar C # TRIALS D4 (AVE. R BAR) * (D4) = UCLr MAX. X BAR
SUM 2 3.27 MIN. X BAR
AVE. R BAR 3 2.58 ALL RANGE VALUES OVER UCL ARE RECALCULATED X BAR DIFF.
TO REDUCE TO AVE. RANGE VALUE
REPEATABILITY-EQUIPMENT VARIATION (EV) ADDITIONAL INFORMATION
EV = AVE. R BAR * k1 NO. TRIALS (m) 2 3
EV = n k1 4.56 3.05
EV =
m
REPRODUCIBILITY- APPRAISER VARIATION (AV) OPERATORS 2 3
AV = SQRT OF [(X BAR DIFF.)*(k2)]̂ 2-[(EV 2̂)/(n*m)] k2 k2 3.65 2.7
AV = n = NUMBER OF PARTS
m = NUMBER OF TRIALS
REPEATABILITY AND REPRODUCIBILITY
R&R = SQRT OF[(EV)̂ 2+(AV)̂ 2] USL LSL % GRRtol
R&R = GRRtol at 5.15
differnce= GRRtol at 6
(constant values rounded)
Tolerance
www.jimakers.com/downloads/Basic Quality Tools.xlsx
VIII-8
24242424
Analysis of Results
Following are examples of gage analysis charts one would find using SPCXL.
MSA XbarR Method Results
Source Variance Standard Deviation % ContributionTotal Measurement (Gage) 0.00182449 0.042714008 5.75% Repeatability 0.00087222 0.029533373 2.75% Reproducibility 0.00095227 0.030858813 3.00%Product (Part-to-Part) 0.02991377 0.172955975 94.25%Total 0.03173826 0.178152338 100.00%
USL 1LSL 0.6Precision to Tolerance Ratio 0.64071012Precision to Total Ratio 0.23976114Resolution 5.7
BIAS ANALYSISReference Bias
Not Available
MSA Data Template
Date:Part Type:
USL:LSL:
Operator 1 Operator 2 Operator 3Part # Reference Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3
1 0.65 0.6 0.62 0.55 0.55 0.5 0.5 0.55 0.52 1 1 0.95 1.05 0.95 0.95 1.05 1 13 0.85 0.8 0.8 0.8 0.75 0.75 0.8 0.8 0.854 0.85 0.95 0.9 0.8 0.75 0.85 0.8 0.8 0.85 0.55 0.45 0.5 0.4 0.4 0.45 0.45 0.5 0.456 1 1 1 1 1.05 1 1 1.05 17 0.95 0.95 0.95 0.95 0.9 0.9 0.95 0.95 0.98 0.85 0.8 0.85 0.75 0.7 0.75 0.8 0.8 0.89 1 1 1 1 0.95 0.95 1.05 1.05 110 0.6 0.7 0.65 0.55 0.5 0.55 0.85 0.8 0.85
3/4/2004
1.00.6
For Attribute data enter A for Accept and R for Reject
Description:
VIII-8
25252525
Results (cont)MSA- Range Chart
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Part Number
Pa
rt R
an
ge
Operator 1
Operator 2
Operator 3
UCL = .129
Center = .05
LCL = .
MSA- Xbar Chart
0
0.2
0.4
0.6
0.8
1
1.2
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Part Number
Pa
rt A
ve
rag
e
Operator 1
Operator 2
Operator 3
UCL = .857
Center = .806
LCL = .754
VIII-8
26262626
The total amount of variance in the gage and in the process
Total Variance
tsmeasuremen repeated # from
study gage from
size sample from
chart control from
,
2
2
222
d
R
d
R
where
gage
total
processgagetotal
VIII-14
27272727
Most processes are designed to meet the customer specification
Because we are using all of our tolerance, we’re forced to keep the process exactly centered.
If the process shifts at all, nonconforming parts will be produced
What Happens When “Shift Happens”?
TargetUpper LimitLower Limit
VIII-19
28282828
Using 75% or less of a tolerance will allow processes to shift slightly without producing any defects.
The goal is to improve your process in order to use the least amount of tolerance possible◦ Reduce the opportunity to produce
defects◦ Reduce the cost of the process
Getting Started
VIII-19
29292929
Defines the width of the process distribution
Cp is calculated by dividing the tolerance zone width by the width of the +/- 3 sigma distribution
This Cp number (or index) tells how many times the distribution will fit into the tolerance zone
A Cp of at least 1.33 is desired
Potential Process Capability Index (Cp)
63
LSLUSLToleranceCp
*
* Which standard deviation do I use?
VIII-25
30303030
If a process uses 50% of a tolerance zone, the Cp value would be 2.0
If a process uses 100% of the tolerance zone, the Cp value would be 1.0
If a process uses 200% of the tolerance zone, the Cp value would be 0.5
What it Looks Like…
VIII-25
31313131
Process capability as a percentage of tolerance
The inverse of the calculations for Cp
Divide the width of the +/- 3 sigma distribution by the width of the tolerance zone
A CR of no more than .75 is desired
Capability Ratio (CR)
LSLUSLToleranceCR
63 *
* Which standard deviation do I use?
VIII-25
32323232
If a processes Cp = 1.0the CR = 100%
If a processes Cp = 2.0the CR = 50%
If a processes Cp = .5the CR = 200%
Calculating CR
neat
VIII-25
33333333
Takes into account not only the spread of the distribution, but also the location of it as well
A Cpk of at least 1.33 is desired Calculating Cpk:
Actual Process Capability (Cpk)
Cpk = Cp - a “Penalty” for off-
center distributions!
3
,3
minLSLMeanMeanUSL
Cpk*
* Which standard deviation do I use?
*
VIII-26
34343434
If a process uses 100% of a tolerance zone, Cp = 1.0
If the distribution is not centered, the Cpk <1.0
What it Looks Like...
Cpk = 1.0 Cpk <1.0
VIII-26
35353535
If a process uses 1/2 of the tolerance zone, the Cpk = 2.0
If the process is not centered, the Cpk value would be <2.0
What it Looks Like (cont.)
Cpk = 2.0 Cpk <2.0
LSL USLTarget TargetLSL USL
this stuff is so awesome
VIII-26
36363636
CR =USL - LSL
6 s
PROCESS CAPABILITY “Cp”, “CR” & “Cpk”
LowSpeed Limit
HighSpeed Limit
65 7570
MEAN
Cp =USL - LSL
6s
Cpk =USL - MEAN
3s
Cpk =MEAN - LSL
3s
1s
Min
*
*
*
VIII-26
37373737
Where Do I Improve?
3rd Location – Cpk Center process Cpk>1.33?
1st Shape – control chartStabilize processAm I in control?
2nd Spread – Cp Reduce variation Cp>1.33?
Control then capabilityVIII-26
38383838
We can determine next steps to improve the process by comparing the Cp and Cpk numbers. For example:
High Cp, high Cpk…Process is centered (accurate) and capable (precise). No improvements are needed.
High Cp, low Cpk…Process is capable (precise) but not centered (accurate). Improvements should shift the process mean to match the target.
Low Cp, low Cpk…Process is not centered (accurate), and variation must be reduced to be precise.
What Cp and Cpk Can Tell Us
VIII-26
39393939
Capability Indices Exercise
USL = 1.505LSL = 1.500s = .00045CR = Cp =
USL = .507LSL = .506s = .00006CR = Cp =
USL = 2800 PPHLSL = 2700 PPHXbar = 2750 PPHs = 12.5PPHCR = Cp = Cpk =
USL = 750 MhzLSL = 735 MhzXbar = 740 Mhzs = 1.333MhzCR = Cp = Cpk =
USL = 1.503LSL = 1.500Xbar = 1.501s = .00045CR = Cp = Cpk =
USL = .251LSL = .250Xbar = .250s = .00015CR = Cp = Cpk =
VIII-26