IENG 486 Lecture 14 - Operating Curve Characteristics
Download
Report
Transcript IENG 486 Lecture 14 - Operating Curve Characteristics
IENG 486 - Lecture 14
X-bar & s Charts:
Trial Limits & Standard Limits,
Control Chart Operating Characteristics
7/17/2015
IENG 486: Statistical Quality & Process
Control
1
Assignment
Reading:
CH 8
8.1 – 8.3.2
8.3.4
8.7.1
Homework:
CH 6 Textbook Problems:
7/17/2015
1; 6 a, b, d only – use Spreadsheet Template (Mat’ls pg)
11; 20 a only; 24; 30
IENG 486: Statistical Quality & Process Control
2
X-Bar & R-Charts
The X-Bar Chart
checks variability in
location between
samples
The R-Chart checks for
changes in sample
variation
UCL
UCL
x
R
LCL
LCL
Sample Number
X-Bar (Means) Control Chart
7/17/2015
Sample Number
R - (Range) Control Chart
IENG 486: Statistical Quality & Process Control
3
X-Bar & Sigma-Charts
Used
when sample size is greater than 10
X-Bar Control Limits:
Approximate 3 limits are
found from S & table
Sigma-Chart Control Limits:
Approximate, asymmetric 3
limits from S & table
UCL x A 3 S
UCL B4 S
CL x
CL S
LCL x A 3 S
LCL B3 S
7/17/2015
IENG 486: Statistical Quality & Process Control
4
X-Bar & Sigma-Charts
Limits can
also be generated from historical data:
X-Bar Control Limits:
Sigma-Chart Control Limits:
Approximate 3 limits are found
from known 0 & table
Approximate, asymmetric 3
limits from 0 & table
UCL μ Aσ
CL μ
UCL B 6
LCL μ Aσ
LCL B 5
7/17/2015
CL c 4
IENG 486: Statistical Quality & Process Control
5
Operating Characteristic
(OC) Curve
Ability of the x and R charts to detect shifts (sensitivity) is
described by OC curves
For x chart; say we know
Mean shifts from
m0
(in-control value) to
m1 = m0 +k (out-of-control value)
The probability of NOT detecting the shift on the first sample
after shift is
P LCL x UCL m m1 m0 k
LCL m1
UCL m1
P
z
n
n
7/17/2015
IENG 486: Statistical Quality & Process Control
6
Ex. Probability of NOT
Detecting Shift
A 3-sigma x chart is used to monitor a normally distributed
quality characteristic. The process std dev is 1.2 and the
sample size is 5. The process mean is in-control at 22.
Find the probability that a shift to 24.4 is not detected on the first
sample after the shift.
7/17/2015
IENG 486: Statistical Quality & Process Control
7
OC Curve for x Chart
Plot of vs. shift size (in std dev units) for various sample
sizes n
OC Curve for x-bar chart with 3-sigma limits
1.00
0.90
0.80
Beta
0.70
0.60
0.50
n=20
n=5
0.40
n=2
n=1
0.30
0.20
0.10
4.
8
4.
4
4.
0
3.
6
3.
2
2.
8
2.
4
2.
0
1.
6
1.
2
0.
8
0.
4
0
0.00
k
x chart not effective for small shift sizes, i.e., k 1.5
Performance gets better for larger n and L or larger shifts
7/17/2015
IENG 486: Statistical Quality & Process Control
8
OC curve for R Chart
Uses distribution of relative range r.v., i.e.,
W R
Suppose
0 - in-control std dev
1 - out-of-control std dev
OC curve for R chart plots vs. ratio of in-control to out-ofcontrol standard deviation for various sample sizes
That is, plot β vs. l 1/0
R chart not very effective for detecting shifts for small
sample sizes
(see Fig. 5-14 in text)
7/17/2015
IENG 486: Statistical Quality & Process Control
9
Probability of Detecting Shift
for Subsequent Samples
After the shift has occurred:
P(NOT detecting shift ON 1st sample)
0.07078
P(DETECTING shift ON 1st sample)
1 0.93
P(DETECTING shift ON 2nd sample)
1 0.066
P(DETECTING shift ON rth sample)
r 1 1
P(DETECTING shift BY 2nd sample)
1 1 0.93 0.066 0.996
P(DETECTING shift BY rth sample)
7/17/2015
i 1
1
i 1
r
IENG 486: Statistical Quality & Process Control
10
Average Run Length (ARL)
Expected number of samples taken before shift is detected is
called the Average Run Length (ARL)
ARL r
r 1
7/17/2015
r 1
1
1
1
IENG 486: Statistical Quality & Process Control
11
Performance of Any
Shewhart Control Chart
In-Control ARL:
Average number of points plotted on control chart before a false
alarm occurs
(ideally, should be large)
ARL0
1
Out-of-Control ARL:
Average number of points, after the process goes out-of-control,
before the control chart detects it
(ideally, should be small)
ARL1
7/17/2015
1
1
IENG 486: Statistical Quality & Process Control
12
ARL Curve for x Chart
3.0
2.9
2.7
2.6
2.4
2.3
2.1
2.0
1.8
1.7
n=1
1.5
1.4
n=2
1.2
1.1
0.8
n=4
0.6
0.5
0.3
0.2
n=20
0.9
20.00
18.00
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
0.0
ARL to detect shift
Plot of ARL1 vs. shift size (in sd units) for various sample
sizes n:
ARL for x-bar chart with 3-sigma limits
k (shift size)
Average Time to Signal, (ATS):
Number of time periods that occur until signal is generated on control
chart
ATS ARL h
h - time interval between samples
7/17/2015
IENG 486: Statistical Quality & Process Control
13
Next Assignment
Reading:
CH 7, CH 6
8.1 – 8.3.2,
8.3.4,
8.7.1
Get Tables 8.2 & 8.3 for your engineering notebook
Homework:
CH 8 Textbook Problems:
7/17/2015
9, 10, 25
IENG 486: Statistical Quality & Process Control
14