Presentation - Setting specifications, statistical ...
Setting Specifications
Statistical considerations
Enda Moran ? Senior Director, Development, Pfizer Melvyn Perry ? Manager, Statistics, Pfizer
Basic Statistics
Population distribution
1.5
(usually unknown).
Normal distribution
described by and .
1.0
0.5
True batch assay
Distribution of possible values
0.0
98
99
100
We infer the population from samples by calculating x and s.
1.5
Sample 1 1.0 Average
98.6
0.5
True batch assay
Distribution of possible values
1.5
Sample 2
True batch assay
1.0 Average 98.7
Distribution of possible values
0.5
1.5
Sample 3
True batch assay
Average 99.0
1.0
Distribution of possible values
0.5
0.0
0.0
0.0
98
99
100
98
99
100
98
99
100
2
Intervals
30 28 26 24 22 20 18 16
0
Population average
10
20
30
40
50
60
70
80
90
100
?100 samples of size 5 taken from a population with an average of 23.0 and a standard deviation of 2.0. ?The highlighted intervals do not include the population average (there are 6 of them). ?For a 95% confidence level expect 5 in 100 intervals to NOT include the population average. ?Usually we calculate just one interval and then act as if the population mean falls within this interval.
3
Intervals
Point Estimation The best estimate; eg MEAN
Interval Estimation A range which contains the true population parameter or a future observation to a certain degree of confidence.
Confidence Interval - The interval to estimate the true population parameter (e.g. the population
mean).
Prediction Interval - The interval containing the next single response.
Tolerance Interval - The interval which contains at least a given proportion of the population.
4
Formulae for Intervals
Intervals are defined as: x ? ks
Assuming a normal distribution
? Confidence (1-) interval
CI
=
x
?
1 n
0.5
t
1-
2
s
,n -1
? Prediction (1-) interval for m future observations
PI
=
x
?
1+
1 n
0.5
t
1-
,n 2m
s
-1
? Tolerance interval for confidence (1-) that proportion
(p) is covered
TI = x ?
( ) n
-
1
1+ 1 n 2
,n-1
z (21-2p )
s
5
Process Capability
Process capability is a measure of the risk of failing specification. The spread of the data are compared with the width of the specifications.
3s
3s
The distance from the mean to the nearest specification relative to half the process width (3s).
The index measures actual performance. Which may or may not be on target i.e., centred.
Ppk
= minUS3Ls-
x
,
x
- LSL
3s
LSL
x
USL
x - LSL
USL - x
6
Process Capability ? Ppk and Cpk
? Ppk should be used as this is the actual risk of failing specification. ? Cpk is the potential capability for the process when free of shifts and drifts.
Random data of mean 10 and SD 1, thus natural span 7 to 13. Added shifts to simulate trends around common average. With specs at 7 and 13 process capability should be unity.
Individual Value
15 14 13 12 11 10
9 8 7 6
14
I Chart of Shifted
1
7 10 13 16 19 22 25 28 Observation
UCL=13.589
_ X=10.092
LCL=6.595
P rocess D ata
LS L
7
T a rge t
*
USL
13
S ample M ean 10.0917
S ample N
30
S tD ev (Within) 1.16561
S tD ev (O v erall) 1.95585
Process Capability of Shifted
LSL
USL
W ithin Ov erall
P otential (Within) C apability C p 0.86 C P L 0.88 C P U 0.83 C pk 0.83
O v erall C apability
Pp PPL PPU P pk C pm
0.51 0.53 0.50 0.50
*
O bserv ed P erformance
P P M < LS L
0.00
P P M > U S L 100000.00
P P M Total 100000.00
6
8
E xp. Within P erformance P P M < LS L 3995.30 P P M > U S L 6296.59 P P M Total 10291.88
10
12
14
E xp. O v erall P erformance P P M < LS L 56966.34 P P M > U S L 68512.70 P P M Total 125479.03
When data is with trend Ppk less than Cpk due to method of calculation of std dev.
Ppk uses sample SD. Ppk less than 1 at 0.5.
Cpk uses average moving range SD (same as for control chart limits). Cpk is close to 1 at 0.83.
Individual Value
I Chart of Raw
13
UCL=13.038
12
11
_
10
X=10.092
9
8
7 14
7 10 13 16 19 22 25 28 Observation
LCL=7.145
Process Capability of Raw
LSL
P rocess Data
LS L
7
T a rge t
*
USL
13
S ample M ean 10.0917
S ample N
30
S tD ev (Within) 0.982187
S tD ev (O v erall) 1.14225
USL W ithin Ov erall
P otential (Within) C apability C p 1.02 C P L 1.05 C P U 0.99 C pk 0.99
O v erall C apability
Pp PPL PPU P pk C pm
0.88 0.90 0.85 0.85
*
7
8
9
10 11 12 13
O bserv ed P erformance P P M < LS L 0.00 P P M > U S L 0.00 P P M Total 0.00
E xp. Within P erformance P P M < LS L 822.51 P P M > U S L 1533.13 P P M Total 2355.65
E xp. O v erall P erformance P P M < LS L 3397.73 P P M > U S L 5446.84 P P M Total 8844.57
When data is without trend Ppk is same as Cpk. Only small differences are seen.
Cpk effectively 1 at 0.99.
Ppk close to 1 at 0.85.
7
Measurement Uncertainty
Bad parts almost always rejected
LSL
Good
USL
parts almost
measurement
always
passed
total
Bad parts
almost
always
rejected
?3measurement
?3measurement
The grey areas highlighted represent those parts of the curve with the
potential for wrong decisions, or mis-classification.
8
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related download
- lesson 8 bell curves and standard deviation
- what is a normal distribution
- combining means and standard deviations
- mean and standard deviation university of york
- distributions of random variables duke university
- the gaussian or normal pdf page 1 the gaussian or normal
- standard deviation estimator ncss
- normal distribution confidence intervals for the mean
- wisc iv frequently asked questions
- normal curve calculations valencia college
Related searches
- 2019 toyota highlander specifications edmunds
- ford v10 specifications motorhome
- vehicle manufacturer specifications using vin
- icd 705 technical specifications 2019
- fda pdf specifications guidance
- aluminum extrusion specifications standards
- briggs and stratton specifications chart
- pc specifications app
- asphalt paving specifications pdf
- specifications manager job description
- copper tubing specifications and sizes
- the joint commission specifications manual