Basic of Statistical Process

Control

Module 3 – Quality Control

 TMA@2008
Learning Outcomes
 After completing this module, you will be able to:
 Understand chance and assignable causes of
variability in a process
 Explain the basic statistic method for modelling
process variation
 Explain the basic of process control and the
development of control chart
 Understand the basic of statistical process control

Module 3: Basic of Statistical Process Control 2

 TMA@2008
Introduction
No two products are ever
identical
Process to produce the
product is inherently vary
Cause of Variability:
1. Random (common
cause)
2. Non-Random
(assignable)
Reducing variability in
process will improved
quality of product
Module 3: Basic of Statistical Process Control 3
 TMA@2008
Process Variability – Common
Cause
 A common cause is a natural cause of variation in the
system.
 No pattern (tidak ada pola)

 Inherent in process; unavoidable (melekat pada proses)

 Can be quantified with summary statistics that are

consistent over time (dapat disimpulkan melalui
kesimpulan statistik secara konsisten dari waktu ke
waktu)
 CANNOT be reduced by adjusting the existing process,
only by changing it (Tidak dapat dilakukan penyesuaian
melalui proses yang ada kecuali dengan penggantian)
Module 3: Basic of Statistical Process Control 4
 TMA@2008
Prediction

Time

The process is stable or in-control

Common causes examples:
Machine vibration (getaran mesin)
Temperature fluctuations (perubahan suhu)
Slight variation in raw materials (sedikit variasi dalam bahan baku)
Human variation in setting control dials (variasi manusia dlm pengaturan
pengendalian kecepatan)

Module 3: Basic of Statistical Process Control 5

 TMA@2008
Process Variability –
Assignable Cause
 Assignable cause or special cause is a response
to some inconsistency in process operation
 Non-random variation (unusual)/ tidak biasa terjadi
 May exhibit a pattern (dapat terlihat bentuk polanya)
 Causes summary statistics that are not consistent
over time (tidak konsisten dari waktu ke waktu)
 CAN be reduced by adjusting the existing process
(dapat dikurangi dengan penyesuaian proses yang
ada.

Module 3: Basic of Statistical Process Control 6

 TMA@2008
? ?

Prediction

Time

The process is unstable or out-of-control

Assignable causes examples:
Batch of defective raw material (Cacat material)
Faulty set-up (Kegagalan set up)
Human error
Incorrect recipe

Module 3: Basic of Statistical Process Control 7

 TMA@2008
Modelling Process Variation(1)
 Process variation can be modelled by statistical
method
 Descriptive Statistics include
 The mean: measure of central tendency
 The Range: difference between largest/smallest
observations in a set of data
 Standard Deviation: measures the amount of data
dispersion around mean
 Data distribution shape: normal or bell shaped or
skewed

Module 3: Basic of Statistical Process Control 8

 TMA@2008
Modelling Process Variation(2)
Normal
bell-shaped
Add up about 30 of most things
and you start to be “normal”
Curve
Normal distributions are divide up
into 3 standard deviations on
each side of the mean

Once your that, you

know a lot about
what is going on

Module 3: Basic of Statistical Process Control 9

 TMA@2008
Modelling Process Variation(3)
 Process characteristics to be controlled can be
measured by statistical inference:
 Parameter Estimation
 Hypothesis Testing
 Review again the concept !

Module 3: Basic of Statistical Process Control 10

 TMA@2008
Statistical Method and Control
Chart
 For the purpose of quality
control a control chart is
developed
 A typical control chart has
control limits set at values
such that if the process is
in control, nearly all
points will lie between the
upper control limit (UCL)
and the lower control limit
(LCL).

Module 3: Basic of Statistical Process Control 11

 TMA@2008
Illustration
 We have a process that we assume the true
process mean is  = 74 and the process
standard deviation is  = 0.01. Samples of size
5 are taken giving a standard deviation of the
sample average as:

 0.01
x    0.0045
n 5

Module 3: Basic of Statistical Process Control 12

 TMA@2008
 Control limits can be set at 3 standard deviations
from the mean.
 This results in “3-Sigma Control Limits”
UCL = 74 + 3(0.0045) = 74.0135
CL= 74
LCL = 74 - 3(0.0045) = 73.9865

Module 3: Basic of Statistical Process Control 13

 TMA@2008
 Choosing the control limits is equivalent to
setting up the critical region for testing
hypothesis
H0:  = 75
H1:   75

Module 3: Basic of Statistical Process Control 14

 TMA@2008
Module 3: Basic of Statistical Process Control 15
 TMA@2008
Conclusion from Illustration
A process that is operating
A process that is
with only chance causes of
operating in the
variation present is said to
presence of
be in statistical control
assignable causes is
said to be out of
control

The utilization of control chart is part of what is

called Statistical Process Control (SPC)

Module 3: Basic of Statistical Process Control 16

 TMA@2008
What is Statistical Process
Control (SPC)
 Statistical process control is a collection of tools
that when used together can result in process
stability and variability reduction
 The eventual goal of SPC is reduction or
elimination of variability in the process by
identification of assignable causes.

Module 3: Basic of Statistical Process Control 17

 TMA@2008
SPC Tools
The seven major tools are

1) Histogram or Stem and Leaf plot

2) Check Sheet
3) Pareto Chart
4) Cause and Effect Diagram
5) Defect Concentration Diagram
6) Scatter Diagram
7) Control Chart

Module 3: Basic of Statistical Process Control 18

 TMA@2008
Module 3: Basic of Statistical Process Control 19
 TMA@2008
Basic Principles(1)

Module 3: Basic of Statistical Process Control 20

 TMA@2008
Basic Principles(2)

Module 3: Basic of Statistical Process Control 21

 TMA@2008
Basic Principles(3):
Important uses of the control chart

• Most processes do not operate in a state of statistical

control
1
• Consequently, the routine and attentive use of control
charts will identify assignable causes. If these causes can
be eliminated from the process, variability will be reduced
2 and the process will be improved

• The control chart only detects assignable causes.

Management, operator, and engineering action will be
3 necessary to eliminate the assignable causes.

Module 3: Basic of Statistical Process Control 22

 TMA@2008
Basic Principles(4):
Popularity of control charts

Control charts are Control charts

Control charts are
a proven technique prevent
effective in defect
for improving unnecessary
prevention
productivity process adjustment

Control charts
Control charts
provide information
provide diagnostic
about process
information
capability

Module 3: Basic of Statistical Process Control 23

 TMA@2008
Basic Principles(5):
Types of the Control Chart

Variables Control Charts

• These charts are applied to data that follow
a continuous distribution (measurement
data).
Attributes Control Charts
• These charts are applied to data that follow
a discrete distribution.

Module 3: Basic of Statistical Process Control 24

 TMA@2008
Choice of Control Limits
 Control limits tell us where the measurements in
a stable process should fall
Lower Upper
Control Control
Limit Limit

Module 3: Basic of Statistical Process Control 25

 TMA@2008
Creating a Control Chart
Upper Control Limit

Center Line

Lower Control Limit

Turn the distribution on its side

Module 3: Basic of Statistical Process Control 26
 TMA@2008
 The use of 3-sigma limits generally gives good
results in practice.
 If the distribution of the quality characteristic is
reasonably well approximated by the normal
distribution, then the use of 3-sigma limits is
applicable.
 These limits are often referred to as action
limits.

Module 3: Basic of Statistical Process Control 27

 TMA@2008
 Warning limits (if used) are typically
set at 2 standard deviations from the
mean.
 If one or more points fall between the
warning limits and the control limits, or
close to the warning limits the process
may not be operating properly.
 Good thing: warning limits often
increase the sensitivity of the control
chart.
 Bad thing: warning limits could result
in an increased risk of false alarms.

Module 3: Basic of Statistical Process Control 28

 TMA@2008
Sampling Size and Sampling
Frequency
 In designing a control chart, both the sample
size to be selected and the frequency of
selection must be specified.
 Larger samples make it easier to detect small
shifts in the process.
 Current practice tends to favor smaller, more
frequent samples.

Module 3: Basic of Statistical Process Control 29

 TMA@2008
Rational Subgroups
 Subgroups or samples should be selected so
that if assignable causes are present, the
chance for differences between subgroups will
be maximized, while the chance for differences
due to these assignable causes within a
subgroup will be minimized.
 Two Approaches:
 Snapshot Approach (Shewhart Approach)
 Random Approach

Module 3: Basic of Statistical Process Control 30

 TMA@2008
Module 3: Basic of Statistical Process Control 31
 TMA@2008
Guidelines for choosing size of subgroup:

One for destructive or chemical batch processes

Two to three when sampling is expensive

Four (or more) for best statistical modeling

Five for computational ease

Ten (or more) for increased sensitivity

Module 3: Basic of Statistical Process Control 32

 TMA@2008
Analysis of Patterns on
Control Charts

Module 3: Basic of Statistical Process Control 33

 TMA@2008
Module 3: Basic of Statistical Process Control 34
 TMA@2008
 Look for “runs” - this is a
sequence of observations of
the same type (all above the
center line, or all below the
center line)
 Runs of say 8 observations
or more could indicate an
out-of-control situation:
 Run up: a series of
observations are increasing
 Run down: a series of
observations are decreasing

Module 3: Basic of Statistical Process Control 35

 TMA@2008
 Cyclic behaviour: may indicate a problem
with the process such as operator fatigue; heat
or stress build-up; etc

Module 3: Basic of Statistical Process Control 36

 TMA@2008
Phase I and Phase II of Control
Chart Application
 Phase I is a retrospective analysis of process
data to construct trial control limits
 Charts are effective at detecting large, sustained shifts
in process parameters, outliers, measurement errors,
data entry errors, etc.
 Facilitates identification and removal of assignable
causes
 In phase II, the control chart is used to monitor
the process
 Process is assumed to be reasonably stable
 Emphasis is on process monitoring, not on bringing
an unruly process into control
Module 3: Basic of Statistical Process Control 37
 TMA@2008
Process Characterization
Ideal State Threshold State
• Process in control • Mean value not centered
• 100% acceptable product • Variability exceed specification
limits

State of Chaos Brink of Chaos

• Process out of control • Process out of control
• Uncontrolled special cause • Product quality wanders
• Off spec product

Module 3: Basic of Statistical Process Control 38

 TMA@2008
What do you do when in
chaos state ?

1. Control special cause variability BEFORE

you try to adjust the mean
2. If you adjust the mean first, your work may
be undone at any time by an uncontrolled
special cause
Module 3: Basic of Statistical Process Control 39
 TMA@2008
Reality of Real Processes
 Real processes, left unattended, will tend to a
state of chaos
 Statistical process control monitors all
processes, even those that have achieved the
Ideal State, so that changes in process
operation can be detected and corrected before
they lead to off-spec product.

Module 3: Basic of Statistical Process Control 40