Advanced Engineering

Statistics
Jay Liu
Dept. Chemical Engineering
PKNU
Statistical Process Control (A.K.A Process Monitoring)

• What we will cover

Reading: Textbook Ch. ? ~ ?

2012-06-27 Adv. Eng. Stat., Jay Liu© 1

What is process monitoring (or SPC) ?
We know that quality is not optional; customers move onto suppliers
that provide quality products
Quality is not a cost-benefit trade-off either
Car sales in North America: the steady rise of the Asian manufacturers
Key point: apply monitoring at every step in the manufacturing line/system
Low variability early on; don’t wait to the end
Definition
A collection of methods for controlling the quality of a product by
collecting and interpreting data to determine the capability and current
performance of a process.
SPC methods make a distinction between what is called common
cause variation and special cause variation.

2012-06-27 Adv. Eng. Stat., Jay Liu© 2

Terminology
Common cause variation (sometimes called inherent variation)
is always present. It normally arises from several sources, each of
which usually makes a relatively small contribution. Common cause is
typically quantified using measures such as the sample standard
deviation s or the range R.
Processes exhibiting only common cause variation are said to be in
statistical control, even if they not be meeting specifications. Such
processes are stable, and hence predictable (within appropriate limits
identified by confidence intervals). The magnitude of common cause
variation determines the system capability.

2012-06-27 Adv. Eng. Stat., Jay Liu© 3

Terminology
In contrast, special cause variation is sporadic, sometimes
upsetting a process when it occurs. Special cause variation can be
distinguished from common cause variation by the size or pattern of
change that occurs in process behaviour. Detection of special cause
variation is often subjective, with guidance from objective techniques.
Because special cause variation is abnormal variation, it may be
harmful or beneficial.
Detecting and acting upon special cause variation is a responsibility of
everyone in an organization, from operators to management. Sytems
exhibiting special cause variation which is not acted upon are not in
statistical control. System capability has no meaning for such
systems (i.e. it is important to ensure that a process is stable before
evaluating its capability).

2012-06-27 Adv. Eng. Stat., Jay Liu© 4

For Your Information
Words from Quality Engineering
Reduction of common cause variation requires fundamental changes in an
operation, requiring management authorization (i.e. fine tuning will have
little effect).
Process Improvement comes about through the identification of special
cause variation and then its deliberate elimination or persistence.

2012-06-27 Adv. Eng. Stat., Jay Liu© 5

Relationship to automatic feedback control
Similar to automatic (feedback) control
continually applied
check for deviations (error)
Different to automatic (feedback) control
adjustments are infrequent
usually manual
adjust due to special causes
aim is to make (permanent) adjustments to avoid that
variability from ever occurring again

2012-06-27 Adv. Eng. Stat., Jay Liu© 6

Control Charts
Used to display and detect this unusual variability
it is most often a time-series plot, or sequence
a target value may be shown
one or more limit lines are shown
displayed in real-time, or pretty close to real-time

SPC (Shewhart) Chart

7
6
Controlled V ariable

5
4
3
2
1
0
1

11

13

15

17

19

21

23

25

27

29
Time

2012-06-27 Adv. Eng. Stat., Jay Liu© 7

Features of Control Charts

Is the process
operating
normally?

SPC (Shewhart) Chart

7
6
Controlled V ariable

5
4
3
2
1
0
1

11

13

15

17

19

21

23

25

27

29
Time

2012-06-27 Adv. Eng. Stat., Jay Liu© 8

 SPC (Shewhart) Chart

7
6 A typical Operating Region
Controlled V ariable

5
4
Normal Operating Region
3
2
1 A typical Operating Region
0
1

11

13

15

17

19

21

23

25

27

29
Time

2012-06-27 Adv. Eng. Stat., Jay Liu© 9

 SPC (Shewhart) Chart

7
Upper Confidence Limit Variable Being
6
Controlled V ariable

Monitored
5
4
3
2
1 Target
Lower Confidence Limit
0
1

11

13

15

17

19

21

23

25

27

29
Time

2012-06-27 Adv. Eng. Stat., Jay Liu© 10

 Suggests something
has changed .
Acts as(Shewhart)
SPC a warning Chart
for production
personnel
7
6
Controlled Variable

5
4
3
2
1
0
1

11

13

15

17

19

21

23

25

27

29
Time

2012-06-27 Adv. Eng. Stat., Jay Liu© 11

Type of control charts

Data
types

Defect rate
Numeric defect count
Attributes …
variables

C chart/ NP chart/
X bar, R X bar, s
U chart P chart

See related minitab command

2012-06-27 Adv. Eng. Stat., Jay Liu© 12

What do we want to see?
Location
Close to target or not?

SPC (Shewhart) Chart

7
6
Controlled V ariable

5
4
3
2
1
0
1

11

13

15

17

19

21

23

25

27

29
Time

2012-06-27 Adv. Eng. Stat., Jay Liu© 13

What do we want to see?
Too much fluctuation?

Var y=0.47
Var y=2.2382

2012-06-27 Adv. Eng. Stat., Jay Liu© 14

X -R / X -s charts

Also known as Shewhart chart

Named for Walter Shewhart from Bell Telephone and Western Electric
In general, we want to know whether the quality meets the target and
variation is also within certain range.
Sample mean ( X ) for monitoring target
Range R or sample standard deviation s for monitoring variability
※Why R instead of s?
In 1920’s when control charts were first introduced, calculations were
carried out by hand, and so the sample range R was strongly preferred to
the sample standard deviation as an estimate of dispersion because it was
so much easier to calculate.

2012-06-27 Adv. Eng. Stat., Jay Liu© 15

X-bar Chart
To “build a control chart” is to determine values for the three lines on
the chart
Centre line
Upper control limit
Lower control limit
The centre line value for an X-bar chart may be
the target value for the performance characteristic of interest
Or the overall sample mean of X values from recent samples of the
measured characteristic, where

Total of X values for al l samples

X
number of samples

2012-06-27 Adv. Eng. Stat., Jay Liu© 16

X-bar Charts - Control Limits
Upper and lower control limits for an X-bar chart are determined from
the pdf of the individual sample means , which is N(,2/n),
Where  denotes the population standard deviation of individual X
measurements
And n denotes the size of each sample
Essentially, the limits represent 100(1-)% confidence interval for the
mean. The UCL and LCL are determined from the following.
x
X  z / 2 x x 
n
x
 X  z / 2
n

2012-06-27 Adv. Eng. Stat., Jay Liu© 17

X-bar Charts - Control Limits
Of course, we generally don’t know the true value of the variance and
we have to use an estimate in its place. In this case, the confidence
limits (and consequently, the control limits) are determined using the t
distribution instead of the normal distribution.
We can estimate the variance of the mean values as follows.
Total of s values for all individual samples
s  sX 
Number of samples

DOF==k(n-1)
Then we compute the control limits as:

X  t . / 2s x
sx
 X  t . / 2
n

2012-06-27 Adv. Eng. Stat., Jay Liu© 18

X-bar Charts - Control Limits
In the past, all calculations were done by hand and were often done by
personnel with very little background in statistics. Therefore, simple
formulas were developed for confidence limits. Some of these are still
in use today. One of the more common methodologies for X-bar and s
charts is given below.

2012-06-27 Adv. Eng. Stat., Jay Liu© 19

X-bar Chart - Control Limits
The centre line value for theX chart in this case is again

Total of X values for al l samples

X
number of samples

and the lower and upper control limits for theX chart are again
determined from the pdf of the individual sample means X , which is
N(, /n) . Using s / c 4 as an estimate of , limits that contain
2

approximately 99.73% of the possible values X of , assuming that the

population mean  is equal to the centre line value
X , are
X s
X 3  X 3
n c4 n
For easier use these limits are expressed as:
Lower control limit = X  A3s

Upper control limit = X  A3s

2012-06-27 Adv. Eng. Stat., Jay Liu© 20

X-bar Chart - Control Limits

where
3
A3 
c4 n

is a constant whose values are shown in the following table:

n 2 3 4 5 6 7 8
A3 2.659 1.954 1.628 1.427 1.287 1.182 1.099

2012-06-27 Adv. Eng. Stat., Jay Liu© 21

S Charts - Control Limits

The centre line value for the s chart is

Total of s values for all individual samples

s  sX 
Number of samples

The lower and upper control limits are intended to contain 99.73 % of
the possible value of s, assuming that the population standard deviation  is
s
equal to the centre line value .
The pdf of the standard deviation s is such that
E (s) = c 4
and
Var (s) =1  c 4 
2 2

where c4 depends on the number of data used to calculate s.

2012-06-27 Adv. Eng. Stat., Jay Liu© 22

S Charts - Control Limits
Even though s is not normally distributed, both the lower and upper
control limits are set as
 1  c2 
3 (standard deviation of s ) = s  3 4 s

c4
 
For easier use these limits are expressed as:
Lower control limit = B 3 s
Upper control limit = B 4 s

where
 1  c2 
B 3  1  3 4 
 c4 
 
and
 1  c2 
B 4  1  3 4 
 c4 
 

2012-06-27 Adv. Eng. Stat., Jay Liu© 23

Constants used to Build s Charts

n 2 3 4 5 6 7 8
B3 0 0 0 0 0.030 0.118 0.185
B4 3.267 2.568 2.266 2.089 1.970 1.882 1.815
c4 0.793 0.886 0.921 0.940 0.952 0.959 0.965

2012-06-27 Adv. Eng. Stat., Jay Liu© 24

Example
Using following measurement data, construct an Xbar-R chart.

Stat > control chart > variables chart for subgroups > Xbar-R

Sample group x1 x2 x3 x4 x5

1 77 82 81 80 78

2 80 77 79 79 78

3 82 79 81 81 82

4 81 77 82 79 79

5 79 82 82 82 79

6 82 82 78 81 80

7 77 82 80 78 82

8 84 81 79 79 81

9 80 77 78 77 81

10 84 80 80 78 79

2012-06-27 Adv. Eng. Stat., Jay Liu© 25

Overview of SPC
General model for control charts
Let W be a sample statistic that measures some quality characteristics. W
and W are mean and standard deviation of W. then the center line, upper
control limit and the lower control limit become

Where k is the distance of the control limits from the center line, expressed
in standard deviation units.
Ex. For monitoring X ,  X  3 X

2012-06-27 Adv. Eng. Stat., Jay Liu© 26

Overview of SPC
General approach for SPC
Phase I: building and testing from off-line data
Data should be collected from “in-control” state
very iterative process
you will spend most of your time here
Phase II: using the control chart
on new, unseen data
implemented with computer hardware and software
usually for real-time display for operators

Line stop, or operator alarm depending on the errors

Control limits are reviewed periodically.

 Continual process improvement.
2012-06-27 Adv. Eng. Stat., Jay Liu© 27
Performance of SPC
Correctness (or falseness)
Error probability
Type I error () : x typical of normal operation, but falls outside UCL
or LCL limits. (ex. diagnose a normal person as a patient)
 = 0.0027 when using 3 x limits

Synonyms: false alarm, false positive, overkill

Type II error(b): x is not in control but falls within UCL and LCL limits
(ex. Diagnose a patient as being with no disease)
Synonyms: false negative, missing rate

※ Nothing makes a control chart more useless to operators than frequent

false alarms.

2012-06-27 Adv. Eng. Stat., Jay Liu© 28

Performance of SPC (Cont.)
Promptness
average number of sequential samples we expect before seeing a point
outside limits.
ARL = 1/ (i.e., the probability that any point exceeds the control limits)
Ex. For an Xbar chart with 3-sigma limit,  = 0.0027.
ARL = 1/a = 1/0.0027 ∼ 370
That is, even if the process remains in control, an out-of-control signal will be
generated every 370 points, on the average.

2012-06-27 Adv. Eng. Stat., Jay Liu© 29

X -R / X -s charts _ Modification

Basic Shewhart chart is not too sensitive to process shifts.

 Western Electric Rules to enhance sensitivity.
2 out of 3 consecutive values of X-bar on the same side of the centre line
and more than 2 standard deviations from the centre line
4 out of 5 consecutive values of X-bar on the same side of the centre line
and more than 1 standard deviation from the centre line
8 consecutive points are the same side of the centre line
7 or more consecutive values is a consistently rising or falling pattern
a recurring cyclic pattern
abnormal clustering close to the centre line (signals a decrease in variation
in the process)
clustering of values close to both control limits (suggests X-bar is following
two distributions instead of one).

2012-06-27 Adv. Eng. Stat., Jay Liu© 30

CUSUM chart
Cumulative Sum (CUSUM) Chart
Shewhart chart takes a long time to detect shift in the mean, away from
target, T.
A Cumulative Sum Chart monitors Si, the cumulative sum (“cusum”) of
departures of sample mean values of measurements x, up to and including
sample i , from their target value, T.

Si    X j  T 
i

j 1

This definition of Si includes the case of samples of size 1. Note that each
value Si includes all of the data collected up to that point.
※ In Shewhart charts, only the current sample value is used as a
basis for decision.

2012-06-27 Adv. Eng. Stat., Jay Liu© 31

CUSUM chart (Cont.)
The following statistics are maintained separately:
SUi = max [0 , SUi-1 + X i - (T + K) ]
SLi = max [0 , SLi-1 - X i + (T - K) ]

If the value of either of these two statistics falls below zero, it is reset to 0.
SU0 = 0 and SL0 = 0
K is normally set to D/2, where D is the magnitude of the shift in population
mean level away from the target value that is to be detected.

2012-06-27 Adv. Eng. Stat., Jay Liu© 32

CUSUM chart (Cont.)
As each new data value is acquired, both SUi and SLi are compared to a
decision limit H
Recommended values of K and H (by Montgomery and Runger)
If we define H  h X and K  k X where  X is the standard deviation of
the samples, Using h = 4 or 5 and k = ½ will give good performance.
See related options in Minitab

2012-06-27 Adv. Eng. Stat., Jay Liu© 33

Example
Suppose you work at a car assembly plant in a department that
assembles engines. In an operating engine, parts of the crankshaft
move up and down a certain distance from an ideal baseline position.
AtoBDist is the distance (in mm) from the actual (A) position of a point
on the crankshaft to the baseline (B) position. To ensure production
quality, take five measurements each working day, from September 28
through October 15, and then ten per day from the 18th through the
25th. (open cranksh.mtw in minitab)
Construct Shewhart chart
Stat > control chart > variables chart for subgroups > Xbar-R
Construct CUSUM chart
Stat > Control Charts > Time-weighted Charts > CUSUM.

2012-06-27 Adv. Eng. Stat., Jay Liu© 34

EWMA (Exponentially Weighted Moving Average) chart
With a Shewhart chart, decisions about departures of interest from
a target value are made on the basis of only the current value of the
measured performance characteristic. That is, zero weight is given to
previous measured values. In contrast, the cusum chart assigns
equal weight to the current measured value and all previous
values of the performance characteristic.
The exponentially weighted moving average chart (EWMA
chart) provides a compromise between these two approaches.
heavier weights for recent observations
small weights old observations

2012-06-27 Adv. Eng. Stat., Jay Liu© 35

EWMA chart (Cont.)
The EWMA chart monitors Et, the exponentially weighted moving
average of all measured values of the performance characteristic, up to
and including time t. The exponentially weighted moving average is
defined as
Et   xt  (1   ) Et 1
which can also be expressed as
Et  Et 1   ( xt  Et 1 )
By custom, Eo is usually set to the target value. This definition for Et is
equivalent to

Et   xt   (1   ) xt 1   (1   )2 xt 2 

2012-06-27 Adv. Eng. Stat., Jay Liu© 36

EWMA chart (Cont.)
Comparison in terms of weight

Control limits for an EWMA chart are


target value  3s
2
where s is an estimate of the standard deviation of the charted
characteristic (e.g. s/ n for sample means).

2012-06-27 Adv. Eng. Stat., Jay Liu© 37

Other control charts
Control charts for count data
np CHART: used for monitoring the occurrence of defective product or
process behaviour when the sample size (number of items inspected at each
sampling interval) is constant, and each item inspected is declared either
acceptable or unacceptable.
p Chart: a tool for monitoring the occurrence of defective product or
process behaviour when the sample size (number of items inspected at each
sampling interval) may vary (because of fluctuating production levels or
other reasons), and each item inspected is declared either acceptable or
unacceptable.
Stat > control charts > attributes charts

2012-06-27 Adv. Eng. Stat., Jay Liu© 38

Examples
Construct P chart using the following data

date Sample size defective date Sample size defective

9/5 12 2 - 46 7

9/6 17 3 - 43 5

9/7 25 4 - 43 0

9/8 30 4 - 43 7

9/9 44 3 - 40 5

9/10 24 4 - 50 3

- 18 2 - 22 3

- 13 1 - 24 5

- 26 4 - 36 6

- 36 6 - 45 8

- 40 2 - 33 3

2012-06-27 Adv. Eng. Stat., Jay Liu© 39

Process capability
Suppose you need to choose a raw material supplier among company A
and company B. You received a database containing quality of a raw
material from each company and plotted them with spec. limits (LSL
and USL) that you product requests. Which one would you choose?

A社 B社

How to quantify this capability?

2012-06-27 Adv. Eng. Stat., Jay Liu© 40

Process capability (Cont.)
Cp (or PCR, process capability ratio)
USL  LSL
CP 
6
Cpk (or PCRk) for one-sided limit  and : calculated from data

 USL     LSL 
C pk  min  , 
 3 3 

In general, Cp (or Cpk) = 1.33 is minimum requirement

※Stat > quality tools > capability analysis
※Note: Cpk and Cp are only useful for a process which is stable

2012-06-27 Adv. Eng. Stat., Jay Liu© 41

Monitoring in industrial practice
Widely used in industry, at all levels
Management: monitor plants, geographic region, countries (e.g. hourly
sales by region)
Dashboards, ERP, BI, KPI
Challenges for you:
Getting the data out
Real-time use of the data (value of data decays exponentially)
Training is time consuming

2012-06-27 Adv. Eng. Stat., Jay Liu© 42

General workflow
1. Identify variable(s) to monitor.
2. Retrieve historical data (computer systems, or lab data, or paper
records)
3. Import data and just plot it.
Any time trends, outliers, spikes, missing data gaps?
4. Locate regions of stable, common-cause operation.
Remove spikes and outliers
5. Estimate limits by eye
6. Calculate control limits (UCL, LCL), using formula
7. Test your chart on new, unused data.
Testing data: should contain both common and special cause operation
8. How does your chart work?
Quantify the type I and II error.  Adjust the limits; repeat

2012-06-27 Adv. Eng. Stat., Jay Liu© 43

General workflow
1. Run chart on your desktop computer for a couple of days
Confirm unusual events with operators; would they have reacted to it? False
alarm?
Refine your limits
2. Not an expert system - will not diagnose problems:
use your head; look at patterns; knowledge of other process events
3. Demonstrate to your colleagues and manager
But go with dollar values
4. Installation and operator training will take time
5. Listen to your operators
make plots interactive - click on unusual point, it drills-down to give more
context

2012-06-27 Adv. Eng. Stat., Jay Liu© 44

Process improvement using the control charts

2012-06-27 Adv. Eng. Stat., Jay Liu© 45

Industrial case study: Dofasco
ArcelorMittal in Hamilton (formerly called Dofasco) has used
multivariate process monitoring tools since 1990’s
Over 100 applications used daily
Most well known is their casting monitoring application, Caster SOS
(Stable Operation Supervisor)
It is a multivariate monitoring system.

We will briefly review multivariate statistics later!

2012-06-27 Adv. Eng. Stat., Jay Liu© 46

Dofasco case study: slabs of steel

2012-06-27 Adv. Eng. Stat., Jay Liu© 47

Dofasco case study: casting

2012-06-27 Adv. Eng. Stat., Jay Liu© 48

Dofasco case study: breakout

2012-06-27 Adv. Eng. Stat., Jay Liu© 49

Dofasco case study: monitoring for breakouts
Screenshot of caster SOS

Warning limits and the action limits.

Lots of other operator-relevant information
2012-06-27 Adv. Eng. Stat., Jay Liu© 50
Dofasco case study: economics of monitoring
Implemented system in 1997; multiple upgrades since then
Economic savings: more than $ 1 million/year
each breakout costs around $200,000 to $500,000
process shutdowns and/or equipment damage

2012-06-27 Adv. Eng. Stat., Jay Liu© 51