Control Charts: 2WS02 Industrial Statistics A. Di Bucchianico
Control Charts: 2WS02 Industrial Statistics A. Di Bucchianico
Goals of this lecture
Further discussion of control charts:
– variable charts
• Shewhart charts
– rational subgrouping
– runs rules
– performance
• CUSUM charts
• EWMA charts
Statistically versus technically in control
“Statistically in
control”
• stable over time /
• predictable
“Technically in control”
• within specifications
Statistically in control vs technically in control
statistically controlled process:
– inhibits only natural random fluctuations (common causes)
– is stable
– is predictable
– may yield products out of specification
technically controlled process:
– presently yields products within specification
– need not be stable nor predictable
Shewhart control chart
graphical display of product characteristic which is important for
product quality
Upper
X-bar Chart for yield Control
14,4
Limit
14,2
Centre Line
X-bar
14
13,8
13,6
Lower
0 4 8 12 16 20 Control
Subgroup
Limit
Control charts
Basic principles
take samples and compute statistic
14,2
X-bar
13,6
0 4 8 12 16 20
Subgroup
Meaning of control limits
limits at 3 x standard deviation of plotted statistic
basic example:
P ( 3 X X 3 X )
P ( 3 X X 3 X )
LCL
P ( 3 Z 3) 0.9973
Example
diameters of piston rings
process mean: 74 mm
0.01
x 0.0045 mm
n 5
UCL 74 3( 0.0045) 74.0135 mm
LCL 74 3( 0.0045) 73.9865 mm
Individual versus mean
group means
individual
74,03 observations
Centre line
73,97
1 10
Range chart
• need to monitor both mean and variance
• traditionally use range to monitor variance
• chart may also be based on S or S2
• for normal distribution:
– E R = d2 E S (Hartley’s constant)
– tables exist
• preferred practice:
– first check range chart for violations of control limits
– then check mean chart
Design control chart
• sample size
• choice of measurement
Rational subgroups
how must samples be chosen?
within subgroup
variation
between subgroup
variation
Strategy 1
process mean
Strategy 2
•detects contrary to strategy 1 also temporary changes of process
mean
process mean
Phase I (Initial study): in control (1)
Phase I (Initial study): in control (2)
Phase I (Initial Study): not in-control
Trial versus control
•if process needs to be started and no relevant historic data is
available, then estimate µ and or R from data (trial or initial study)
•if points fall outside the control limits, then possibly revise control
limits after inspection. Look for patterns!
Control chart patterns (1)
6 CTR = 0.00
2 LCL = -10.00
-2
-6
-10
0 3 6 9 12 15 18
Observation
Cyclic pattern,
Control chart patterns (2)
6 CTR = 0.00
LCL = -10.00
2
-2
-6
-10
0 4 8 12 16 20
Observation
Trend,
course of pin
Control chart patterns (3)
6 CTR = 0.00
LCL = -10.00
2
-2
-6
-10
0 4 8 12 16 20
Observation
Shifted mean,
Adjusted height Dartec
Control chart patterns (4)
Phase II: Control to standard (1)
Phase II: Control to standard (2)
Runs and zone rules
•if observations fall within control limits, then process may still be
statistically out-of-control:
– patterns (runs, cyclic behaviour) may indicate special causes
– observations do not fill up space between control limits
•extra rules to speed up detection of special causes
•Western Electric Handbook rules:
– 1 point outside 3-limits
– 2 out of 3 consecutive points outside 2 -limits
– 4 out of 5 consecutive points outside 1 -limits
– 8 consecutive points on one side of centre line
•too many rules leads to too high false alarm rate
Warning limits
Detection: meter stick production
• mean 1000 mm, standard deviation 0.2 mm
Performance of control charts
expressed in terms of time to alarm (run length)
two types:
– in-control run length
– out-of-control run length
14,2
X-bar
14
13,8
13,6
0 4 8 12 16 20
Subgroup
Statistical control and control charts
•statistical control: observations
– are normally distributed with mean and variance 2
– are independent
•out of (statistical) control:
– change in probability distribution
•observation within control limits:
– process is considered to be in control
•observation beyond control limits:
– process is considered to be out-of-control
In-control run length
•process is in statistical control
Out-of-control run length
Metrics for run lengths
•run lengths are random variables
Run lengths for Shewhart Xbar- chart
in-control: p = 0.0027
UCL
LCL
0.9973 0.99730.99730.0027
Geometric distribution
Geometric Distribution
(X 0.001)
3 Event prob.
2.5 0.0027
2
1.5
probability
1
0.5
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
(X 1000)
Numerical values
Shewhart chart for mean (n=1)
P(|X|>
3) ARL SRL
3 0.5 2 1.4
Scale in Statgraphics
Are our calculations wrong???
200
150
100
50
A
0
-2 -1.5 -1 -0.5 0 0.5 1 1.5 2
Process mean
Sample size and run lengths
increase of sample size + corresponding control limits:
Numerical values
Shewhart chart for mean (n=5)
Runs rules and run lengths
• in-control run length: decreases (why?)
Performance Shewhart chart
•in-control run length OK
CUSUM Chart
plot cumulative sums of observation
change point
CUSUM tabular form
assume
– individual observations
i
C max 0, X i 0 K C
i 1
C
i max 0, 0 K Xi C
i 1
C C 0; alarm if max C , C
0 0
i
i H
Choice K and H
CUSUM V-mask form
UCL
CL
change point
LCL
Drawbacks V-mask
• only for two-sided schemes
Rational subgroups and CUSUM
• extension to samples:
– replace by /n
• contrary to Shewhart chart , CUSUM works best with
individuals
Combination
•CUSUM charts appropriate for small shifts (<1.5)
Headstart (Fast Initial Response)
•if process equals target value µ 0 is, then CUSUMs quickly return
to 0
•if process mean does not equal target value µ 0, then faster
alarm
CUSUM for variability
i
S max 0,Vi K / S
i 1
S
i max 0, K / V i S
i 1
S S 0; alarm if max S , S
0 0
i
i H
Exponentially Weighted Moving Average chart
EWMA Chart for Col_1
11.4
UCL = 11.00
11 CTR = 10.00
10.6 LCL = 9.00
EWMA
10.2
9.8
9.4
9
0 3 6 9 12 15
Observation
Why control charts for attribute data
•to view process/product across several characteristics
Control charts for attributes
p-chart
D
sample fraction of nonconforming:pˆ
n
p (1 p )
mean: p ˆ
2
variance pˆ
n
p-chart
m mn
i 1
i 1
p
i pˆ D
i
average of sample fractions: m m
p 1 p
UCL p 3
n
CL p
p 1 p
LCL p 3
n
Assumptions for p chart
c-chart
•Counts the number of non-conformities in sample.
c k
Prob(# nonconf. = k) = e c
k!
c-chart
Poisson distribution: mean=c and variance=c
UCL c 3 c
CL c
LCL c 3 c
Moving Range Chart
use when sample size is 1
indication of spread: moving range
Situations:
automated inspection of all units
low production rate
expensive measurements
repeated measurements differ only because of laboratory error
Moving Range Chart
calculation of moving range:
MRi xi xi 1
individual moving
measuremen range
ts
MR
UCL x 3 UCL D4 MR
d2
CL MR
CL x
LCL D3 MR
MR
LCL x 3
d2
d2, D3 and D4 are constants depending number of observations
Example:
Viscosity of Aircraft Primer Paint
8 33.27 0.41
x 33.52 MR 0.48
Viscosity of Aircraft Primer Paint
since a moving range is calculated of n=2 observations, d2=1.128,
D3=0 and D4=3.267
Viscosity of Aircraft Primer Paint
35
34.5
34
X 33.5
33
32.5
32
0 3 6 9 12 15
1.6
U
C
1.2
L
MR 0.8
0.4
0
0
3 6 9 12 15
Tool wear chart
known trend is removed (regression)
USL
UCL
reset
LCL
LSL
Pitfalls
bad measurement system
bad subgrouping
autocorrelation