Module III - Statistical Process Control

ID document: e.g.
PPT/L-TCIO-1
Chapter III –
Statistical Process
Control
Operational quality and Lean
management - S7 EENG 4
Title of Lesson
Title of Lesson 1
ID document: e.g. PPT/L-TCIO-1
Chapter organization
1 – Introduction / reminders about dispersions
2 – Common causes and Special causes
3 – Capabilities
4 – Principles of SPC
- Control charts
- Control charts with measures
- Attribute control charts => next lecture
- EWMA and CUSUM boards
Title of Lesson 2
Targets of today lecture :

● Understand the difference between common causes and special causes
● Make the difference between a gap between average and tolerance interval a gap
between dispersion and tolerance interval
● Understand what « capability » means for a mass production industry
● Understand how is used a control chart
● Know how to build Shewhart control charts
● Understand the advantage of Cusum and EWME charts compared to Shewhart

Chart
Title of Lesson 3
1 – Introduction / reminders
Title of Lesson 4
Reminder from the last lecture :

a population : N pieces a sample : n pieces
Mass production : too costly to check each piece

Statistics make possible to know what’s the whole batch
characteristics by sampling inspection Title of Lesson 5
Measures from the sample of 30 nuts
Thickness
despite nuts are

coming from the
same batch, they
don’t have the same
dimension …
Title of Lesson 6
2 – Common causes, special causes
Title of Lesson 7
What creates dispersion of dimensions ?

Let’s take an example : wood planer
t i n g ,
The objective is to get the expected thickness of the piece of wood and a smooth surface
e s e t d
s a m l w o o
Cutting tool
h t h e , f i n a m e
n w i t n u t s h
High speed
e s a Front roller
Eve r t e
Rear roller
h a y s t
rotation tool
a s f o e a l w t i o n
w e l l o t b o d u c
as Final wood
w i l l n s p r
e s s
thickness
m a s
i c k n a k e Piece of wood
th y o u m
i f
Title of Lesson 8
What creates dispersion of dimensions ?

-wear of the cutting tool -vibrations of the cutting tool
-initial thickness of the wood piece -temperature of the machine
-hardness of the wood piece -dust between table and piece of wood
Etc. Etc.
Causes we can eradicate Causes we will have to

or minimize a lot "live with".
Special causes Common causes
Title of Lesson 9
In industry for any process, causes of

dispersion are a lot
These 5 fundamental causes of dispersion, and therefore of non-quality, are generally
referred to as the 5Ms:
- Materials
- Manpower/people
- Machine/equipments
- Methods
- Mother nature (environment)
Title of Lesson 10
Common causes - Examples

● Examples of common causes :
- Gaps in the kinematic chain of the machine
- Defects in the spindle (=broche) of the machine
- Variation of raw material properties in the same batch
Title of Lesson 11
Special causes
● Special (or accidental) causes are identifiable causes, often irregular and
unstable, and thus difficult to predict. The appearance of a special cause
requires intervention on the process.
● Example of special causes:

- Tool misalignment
- Tool fatigue
- Bad lubrication
- Material non-conformity
- Badly adjusted machines
- Poorly qualified personnel
Title of Lesson 12
Special causes
● Special causes can be classified into two categories:
- Causes that affect the setting, the position of the monitored value (ex : tool
misalignment)
- Causes that affect the dispersion and therefore the capability of the
machine (ex : lubrication of the machine)
Title of Lesson 13
Common causes specificity

Reminder : Galton board experience
bell-shaped curve
Also called Gauss Curve
Any system, subjected to many factors, independent of each other,

and of equivalent order of magnitude of effect, generates a
distribution that follows a normal distribution.
All common causes together make a process following the Gauss curve
Title of Lesson 14
Reminder about Gauss curve (normal law)

- 68.26 % of the values are distributed
Inflexion point
around the mean between ± 1 standard
deviation

deviation

deviation
The last result is important because it

represents the dispersion
Title of Lesson 15
After eradication of special causes, process result is predictable
(tolerance interval)
No part out of the tolerance

interval should occur
Mean value of the production

is not exactly matching with
the distribution mean value
Variation due to common causes

of the process
Title of Lesson 16
3 – Capability
Title of Lesson 17
3 – Capability
A – Introduction
Title of Lesson 18
Requirement for finished products with low rate of

defective. Short experience :
Context : simulation of assembling of 10 products made of 10 components (random choice).
Overall good level as 90% of the components are complying the specifications
1 finished product
= 10 components
Red component : defective
Light brown : component

complying specifications
Title of Lesson 19
Requirement for finished products with low rate of

defective. Short experience :
Let’s bet :
simulation :
Result :
Conclusion :
Title of Lesson 20
Capability – Introduction
Now you are aware that :
Having a low rate of defective for finished products requires to be

very demanding on each component rate of defective
Even if few % of the production is out of the specification it is not

acceptable
Title of Lesson 21
Capability - Definition
Capability is the measure establishing the relationship between the actual

performance (dispersion) of a machine or process, and the required
performance (tolerance interval).
Title of Lesson 22
● The indicator Cp (Process Capability) compares the expected performance
of the process (tolerance interval) and the actual performance on it
(dispersion). We can therefore define the calculation of the capability by :
● TI : tolerance interval
● S: estimate of the standard deviation on a sufficient number of measurements
A process is capable if C > 1.33
Title of Lesson 23
Tolerance interval Tolerance interval
Non-capable process : Capable process :

Cp < 1 Cp > 1
Dispersion Dispersion
● On the left, a non-capable process. The dispersion is larger than the

tolerance interval.
● On the right, a capable process: the dispersion is less important than the
confidence interval.
Title of Lesson 24
Capability – Different cases

● 1st case :
Cp = 1
- Analysis: The curve modeling the total

population produced falls just within the
tolerance range.
- Action: No adjustment possible.
Title of Lesson 25

● 2nd case :
Cp < 1
- Analysis: The curve does not fit within the

tolerance range.
- Action: No adjustment is possible, and there are

rejects in the case of a perfect adjustment.
Title of Lesson 26

● 3rd case :
Cp > 1
- Analysis: The curve modeling the total

population produced is well within the tolerance
range.
- Action: Adjustments are possible, as well as a

slight drift of the process.
Title of Lesson 27

● 4th case :
Cp > 1
- Analysis: The fault is due to a major

misadjustment.
- Action: Do not exceed the drift (=dérive) and

adjust the machine correctly.
Title of Lesson 28
3 – Capability
- Short-term indicators
B – Cp, Cpk, Cm, Cmk, Pp, Ppk - Long-term indicators
- Performance rate
Title of Lesson 29
● The imposed capability assessment method is derived from the 2016
automotive industry standard IATF 16949 (formerly ISO/TS 16949). It defines
two types of indicators:
- Cp and Cm indicators: Short-term indicators that reflect the dispersion over

a very short time (a few days)
- Pp indicators: Long-term indicators that reflect the reality of delivered

productions (one week to months).
Title of Lesson 30
Short term capabilities
Title of Lesson 31
Short term capability Cm & Cp

They are short-term indicators that reflects the dispersion over a very short time.
𝑇𝑜𝑙𝑒𝑟𝑎𝑛𝑐𝑒 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙 𝑇𝐼
𝐶𝑚 = 𝐶𝑝 = =
𝑆ℎ𝑜𝑟𝑡 𝑡𝑖𝑚𝑒 𝑑𝑖𝑠𝑝𝑒𝑟𝑠𝑖𝑜𝑛 6𝜎𝑆𝑇
With the short term dispersion :
Title of Lesson 32
Short term capability Cm & Cp

To determine the standard deviation in the short term :
● About 50 consecutive parts manufactured by the process (or about 50

measurements in the case of studies of physical quantities) are taken.
● We can also take small samples at regular intervals without intervention on
the process during the production of the samples. It is thus necessary for this
formula to have :
n approximately = 50
These samples represent an instantaneous dispersion, taking into account only

the variability of the machine.
Title of Lesson 33
Adjustment indicator Cmk & Cpk

● This index characterizes the centering of the process. It is expressed by the
calculation :
𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑏𝑒𝑡𝑤𝑒𝑒𝑛 𝑡ℎ𝑒 𝑚𝑒𝑎𝑛 𝑎𝑛𝑑 𝑡ℎ𝑒 𝑛𝑒𝑎𝑟𝑒𝑠𝑡 𝑙𝑖𝑚𝑖𝑡 𝑜𝑓 𝑇𝐼

𝐶𝑚𝑘 = 𝐶𝑝𝑘 =
0,5 × 𝑆ℎ𝑜𝑟𝑡 𝑡𝑖𝑚𝑒 𝑑𝑖𝑠𝑝𝑒𝑟𝑠𝑖𝑜𝑛 = 3𝜎𝑆𝑇
Example : Cp = IT/C Cp > 1

A=3σ
B Cpk = B/A Cpk < 1
C=6σ Title of Lesson 34

Examples : find the Cp and Cpk for the below cases
(tolerance interval) (tolerance interval)
10+/-,5 0,30 Cp =
Cp = 3 0,15
1/0,3=3,3
Cpk =
Cpk = 3 10,2+3X0,05
0,3/0,15=2
= 10,35
9,5 10 10,5
10,2
Cp = 3 Cp = 3
Cpk = 2
Cpk = 1
Title of Lesson 35
Cm & Cmk parameters

● To determine Cm and Cmk one must:
- Conduct a short period dispersion study [Mother nature/environment]
- In a uniform material [material].
- With the same skilled operator [manpower].
- With the same [methods]
- With the same [means]
Title of Lesson 36
Cp & Cpk parameters

● To determine Cp and Cpk one must:
- Conduct different periods dispersion study [Mother nature/environment]
- In different batch of material [material].
- With different skilled operators [manpower].
- With several [methods]
- With different [means]
Title of Lesson 37
Conclusion about Cm and Cp

In conclusion Cm is about the machine itself as we don’t change the 5M
○ If the capability is bad, there is a problem on the machine
Cp is about the entire process as we modify the 5M during our study.

○ It is about all the different elements of the process and not only the machine.
Title of Lesson 38
Long term capabilities
Title of Lesson 39
short term / long term capability
Time
Short term capability
(can be during just one day)
Long term capability : from few days to few months
Title of Lesson 40
Long-term performance Pp
● The dispersion of the measurements over the long term is evaluated by :
● With the long term dispersion :
For the long-term standard deviation we need a minimum of 100

measurements. We have here: n > 100
Title of Lesson 41
Long-term performance Pp
● The Pp is calculated :
- From a representative sample of the whole production
- Over a period of more than 1 week (representative of the production

delivered to the customer)
- The sample must take into account the variations of instructions as for
example the adjustments
Title of Lesson 42
Long-term adjustment index Ppk

The Ppk allows to evaluate the centering of the process within the limits of the
IT.
● The Ppk is expressed by the calculation :
A process is capable in the long term if Ppk > 1.33
Title of Lesson 43

● Remark : The Pp can be good but the process can still be not correctly
adjusted. This case is observed when the production is decentralized and in
this case we can find parts out of tolerances.
Tolerance interval Tolerance interval
Here the Pp is ok but the

setting is not good.
Dispersion Dispersion
Title of Lesson 44
We speak of a perfect setting if Pp = Ppk.
● The greater the deviation, the greater the difference between Pp and Ppk.
● The dispersion of the pocess is well centered when Pp=Ppk
Title of Lesson 45

● We can conclude that the long-term and short-term capability :
- Give an indication of the dispersion (Cm, Cp, Pp)
- Give an indication of the centering of the process with respect to a target

(Cmk, Cpk, Ppk)
● We can have acceptable Cm, Cmk, Cp, Cpk, Pp and Ppk (>1,33) but the
process can still be improved.
Title of Lesson 46
Conclusion about capabilities

Capability value Comments conclusion
Mass production should not

C < 0,67 Very bad capability
start
Mass production should not
0,67 < C < 1 Bad capability
start
Mass production is risky :
1 < C < 1,33 Medium - capability
requires a perfect setting
Mass production not risky
1,33 < C < 1,67 Medium + capability
but improvable
Mass production quite
1,67 < C < 2 Good capability
comfortable
Mass production easy to
C>2 Very good capability
manage
Title of Lesson 47
4 – Statistical Process Control

(SPC)
Title of Lesson 48
4 – Statistical Process Control (SPC)
A – Introduction
Title of Lesson 49
Hystory of Quality : increasing efficiency to achieve quality
From final Quality Quality Total Quality

inspection control assurance « proactive »
« reaction » « smooth « prevention » (anticipation)
running »
Beg of 20th
Today
century 1930 -1950 1950 -1960 1970 -1980
Sampling , Organization & Training ,

Final inspection
statistic tools procedures motivation
Shewhart Crosby
Taylor Juran
Deming Ishikawa
SPC – Introduction
In a traditional manufacturing process, a final control of the product is
performed. This type of control has many disadvantages:
Informations on
performance Action on the
production
Sorting, rejects, rework, delays...
Title of Lesson 52
Controlling the manufacturing process with statistical techniques has the
following advantages: Deep knowledge of the whole
population characteristics with the
measuring of few samples only
Informations on act on the process before the

Action on the realization of non-conformities
performance
process
Reduction of rejects
Reduction of product costs
Batch available immediately (no

need to wait for final inspection)
No additional costs
Title of Lesson 55
● Even when a machine is perfectly adjusted to the target coast, there is a
dispersion of values around it. This dispersion is due to common causes.
● The set of common causes represents the natural dispersion of the process.
Most of the time, we have no choice but to live with them and it is not
necessary to intervene on the process.
● Examples:
- Gap in a slide(=coulisse)
- Variation of the hardness in a material
- Variation in temperature
- Dimensional variation of the pieces
Title of Lesson 56
● On the other hand, a special cause can have a great influence on the process
and requires an intervention on it. A special cause is an assignable cause, so
one can intervene on it.
● Special causes can have an important effect on the position or dispersion of

the process.
● Examples:
- Tool fatigue
- Chips(=copeaux) under a stop
- Material change
- Poor lubrication
Title of Lesson 57
After eradication of special causes, process result is predictable
During mass production, No part out of the tolerance

interval should occur
the questions are : Mean value of the production
Is the machine setting still good? is not exactly matching with

the distribution mean value
Is there any special cause occuring? Variation due to common causes

of the process
Title of Lesson 58
● Depending on the variations that the process will present, we have to ask
ourselves if we should intervene or not :
What criteria will give the statistic
- "Am I in the presence of a signal?" (special cause)
proof
In this that
case, we must the process has drift
intervene.
- "Am I inand
the presence of noise?it’s
that (common necessary
cause) to
In this case, the intervention is not necessary.
take actions ?
Title of Lesson 59
● The control chart is the tool that will allow us to make the right decisions.
● It is a graphical tool that will allow us to follow :
○ in the case of measurement control charts : the variations of position

and dispersion of the process in time
○ in the case of attribute control charts : the number or proportion of non-

conformities, or non-conforming products.
Title of Lesson 61
- Introduction
- Structure
B – Control chart - Sampling methods
- Driving method
Title of Lesson 62
Measurement control chart example :

3 pieces were measured several times / hour during the manufacturing of the batch:
The range : Max –Min of the 3 pieces is

The mean of the 3 pieces is recorded below
recorded below :
Upper limit
Over the limit Upper limit
X Chart R Chart in the limit

Lower limit
Title of Lesson 63
Principle :
With the help of statistics, looking at the mean and the range of a sampling of few pieces
will confirm or deny that the production is still on the target.
Title of Lesson 64
Control chart – Introduction

● The piloting of a process with the help of a chart allows :
- A systematic control of the manufacturing process with early detection of

any anomaly likely to cause non-conformity
- To have the products immediately available for further operations,
eliminating intermediate controls
- To have a supporting document containing a lot of information
● The keeping of a control chart should, as far as possible, be entrusted to the

operator in order to reduce the inertia of the correction and to increase the
sense of responsibility concerning the quality of the work done.
Title of Lesson 65
Mathematical property
Mean of N samples
Population
𝝈
σ’=
√𝑵
σ
If we take many times N pieces among the whole population, the mean value of the N pieces follows a normal law with :
-same average as the whole population

-standard deviation σ’ 𝝈
σ’=
√𝑵 Title of Lesson 66
- X chart
- R chart
C – Measurement control
charts
- S chart
Title of Lesson 67
Exemple of a control chart

● https://www.youtube.com/watch?v=0CtJJGprG6A
Title of Lesson 68
How to build a X chart (Average chart) and a R chart (Range chart) ?
● The X chart is intended to control the position of the mean of a small sample
(n usually between 3 and 10) taken in a predefined interval.
● The R chart is intended to control the position of the range of a small sample
(n usually between 3 and 10) taken in a predefined interval.
- Step 1: Data collection : A sample is regularly taken and its value or its
difference from the target value is noted.
The mean and range of the measurements are calculated.
- Step 2: Calculation of the limits of the control chart
This step allows to set the limits within which it is natural to find common
causes.
Title of Lesson 69
Step 1 :Example of data collection
…..
x̅ R̅
Title of Lesson 70
Table of coefficients
Title of Lesson 71
Step 2 : Example of limits calculation for X chart

● When the dispersion is estimated based on Rbarre, the formula is :
ഥ±𝑨 ×𝑹
𝑼𝑳𝑪, 𝑳𝑳𝑪 = 𝑿 ഥ
𝟐
As a reminder: Rbarre is the average of the ranges.
● The range and the mean of the ranges are calculated by :
Title of Lesson 72
Step 2 : Example of limits calculation

For our example :
ഥ±𝑨 ×𝑹
𝑼𝑳𝑪, 𝑳𝑳𝑪 = 𝑿 ഥ
𝟐
ULC = 24,97 + 0,577 X 0,73 = 25,39
LLC = 24,97 - 0,577 X 0,73 = 24,55
Title of Lesson 73
X chart (Average chart)
● When the standard deviation of the population is known, the control limits are
defined as follows:
ഥ 𝟑 ഥ±𝑨×𝝈
𝑼𝑳𝑪, 𝑳𝑳𝑪 = 𝑿 ± ×𝝈=𝑿
𝒏
The ratio can be taken over by the coefficient A (see coefficient table).
The target for X charts is the mean of all values of the sample.
Title of Lesson 74
R chart (Range chart)

● Based on the standard deviation we get these limits:
● Based on the mean of the range:
Title of Lesson 75
Step 2 : Example of limits calculation

For our example :
ULC = 2,114 X 0,73 = 1,54
LLC = 0 X 0,73 = 0
Title of Lesson 76
Step 3 : Interpreting the charts and decide what to

do.
Resume manufacturing
stop and change the setting
Title of Lesson 77
Graph Description Decision
Process under control :
-The R curves oscillate on
either side of the mean
PRODUCTION
-- 2/3 of the points are in
the central third of the
chart
Upper limit case Lower limit case
Point out of bounds - Short-term capability - Short-term capability
Table of
(Out of control) deteriorates. The origin must ameliorates
The last point crossed be found and actions taken - Measuring system is blocked
one of the control limits - Measurement error
finite Upper or lower trend

(Run)
7 consecutive points
Graph
Upper limit case
- Short-term capability
Lower limit case
deteriorates. The origin must ameliorates. The origin must
be found and actions taken be found to maintain this
above or below the mean amelioration.
decision Increasing or decreasing

Inscreasing serie case
Decreasing serie case
trend (Trend) deteriorates. The origin must ameliorates. The origin must
rules 7 consecutive points are

steadily
increasing/decreasing
be found and actions taken be found to maintain this
amelioration.
1 point close to the Upper limit case :

limits Monitor the capability. If several points on the map are also
The last point drawn is in close to the limits, the capability deteriorates. The origin must
the 1/6 at the edge of the be found and action taken.
control chart
Title of Lesson 78
Probable causes : Process out of control

● For a process out of control (normality problem):
- Samples include values from two or more production lines

- Mixing of different batches of material
- Over-setting of the machine
- Unstable measurement equipment
- Incorrect calculation of control limits
- Incorrect reporting of control limits
- Manipulations
Title of Lesson 79
Probable causes : Trend

● For a Trend (increasing or decreasing trend) :
- Tool wear/fatigue
- Variation of the temperature
- Dirtiness of the machine, the tool, the measuring device
- Fatigue of the personnel
Title of Lesson 80
Probable causes : Run

● For a Run (upper or lower trend):
- Change in material
- Change of load(=charge)
- Change of personnel, controller
- New tool, new measuring device
- Modification of the setting
Title of Lesson 81
Frequency of sampling
● Determining the frequency of sampling is mostly a matter of common sense.
Other factors, such as sample size, control time or control costs can also be
decisive. The goal is to obtain a correct picture of the process.
● For efficient sampling, a large number of small samples should be taken

rather than a few large samples.
● When starting the process analysis with a control chart, it is wise to have a
relatively high sampling frequency and to decrease it when the process
stabilizes.
Title of Lesson 82
Frequency of sampling
● The following points should be noted:
- The size of a sample must be at least 3 for the industry
- At least 3 samples must be taken between two settings
- 5 to 10% of the parts must be checked for a slow or even medium

production rate
- 2 to 5% of the parts must be checked for a fast production rate
Title of Lesson 83
Rules of verification
Title of Lesson 84
Rules of verification of a setting

Point out of control
Setting to be redone
ULC
Setting to confirmed
Target
Correct setting et the end
Correct setting
LLC
1st sampling after the 2nd sampling after the

setting setting
Title of Lesson 85
Rules of verification of a setting

● Step 0 : A sample is taken and an out-of-control point is observed. It is
necessary to intervene.
● Step 1 : The appropriate adjustment is made
● Step 2 : Immediately after the adjustment, a sample is taken.

○ If the point is in the central 1/3 of the target: the adjustment is correct
and we can continue the production
○ If the point is outside the limits: the adjustment must be resumed
immediately
○ If the point is between the central 1/3 of the target and the limits of the
control chart, it must be confirmed that it is good. An additional sample is
then taken to check that the adjustment is correct: step 2 is performed
again.
Title of Lesson 86
Synthesis :
Driving/Monitoring methods
Title of Lesson 87
Driving methods
● 5 steps allow to summarize the control method with the control charts:
- 1. Take a sample of consecutive parts without intervention / define the

limits for X chart and R chart
- 2. Measure the parts and transfer the points on the control chart
- 3. Interpret according to the control rules
- 4. If necessary, intervene
- 5. Check the effect of the intervention
Title of Lesson 88
Other measurement control

charts
Title of Lesson 89
S chart (Standard deviation

chart)
Title of Lesson 90
S chart (Standard deviation chart)
● This chart is intended to control the position of the dispersion of a small

sample taken in a predefined interval.
● When n > 5, S chart is more relevant than R chart because it uses information
of each measured value while R chart uses only information of Min and Max
value in the sample.
● As standard deviation calculation isn’t easy, the use of this chart is

recommended with a computerized system
Title of Lesson 91
Limit calculation for S chart

● Based on the standard deviation we get these limits:
● Based on the mean of the standard deviation:
Title of Lesson 92
S chart : example of limits calculation

For our example :
ULC = 2,089 X 0,29 = 0,606
LLC = 0 X 0,29 = 0
Title of Lesson 93
CUSUM chart
Title of Lesson 94
CUSUM chart
● The CUSUM chart is more efficient for detecting small or medium amplitude
drifts (1 to 2σ drift) on the average. To detect slow and progressive drifts in a
continuous manufacturing process, the CUSUM chart is very efficient because
it detects a drift on an average number of successive samples.
● The sum of cumulative deviations between the average of a sample and a

target value is indicated.
● The Cusum chart is not relevant for range drifts if the average remains the
same.
Title of Lesson 95
CUSUM chart
● As with the EWMA map, we have a link between past and present averages:
● Based on this value, we will construct two statistics:
With as parameters :
k : filter coefficient
: : the statistic detecting the

positive shifts of the mean
: : the statistic detecting its

decreases
Title of Lesson 96
CuSum chart : example

For our example : see page in the excel file « SPC example »
Same data for both charts : deviation

is more visible with CuSum chart
For this example : k=1 ; h=2,52
Title of Lesson 97
CUSUM chart
● Analyzis :
As long as Si+ and Si- are between h and -h: the process is under control. If they
are not between, the process is out of control.
Three parameters are to be defined: h, k and n. These parameters are defined in

terms of δ: the deregulation.
● We will therefore use the following table to define these different values:
Remenber :
k is link to the deviation (expressed in standard deviation unit) you want to be warned about : the lower it is the more sensitive it will be
Increase sensibility will also increase the number of false alarms
After a new setting on the machine, reset the Si calculation : S0+= S0- =0
Title of Lesson 98
Example of CUSUM chart in

Tutorial Class
Title of Lesson 99
EWMA chart (same idea than

CUSUM chart)
Title of Lesson 100

EWMA chart
The EWMA (Exponentially Weighted Moving Average) control chart allows to
detect small deviations from the target. Indeed, each point established
previously is related to the following point; thus, for each value, the average
weighted by a coefficient λ is obtained :
With as parameters:
λ : constant less than or equal to 1. Generally : λ = 0.33
: initial value = target
: the average of two successive observations
Title of Lesson 101

EWMA chart
We finally have:
The lower the λ, the greater the detection of small drifts but the less sensitive
we will be to sudden and large drifts. λ is therefore chosen according to the
knowledge of the process: most often λ = 0.33.
Title of Lesson 102

EWMA chart
The bounds of the control chart are set to [target ± 3 σ], so if the means are
randomly distributed with a standard deviation equal to then the standard
deviation of the Xi distribution is :
With as parameters :
- i : the number of the sample

- λ : a constant
- n : the number of measurements of the sample
Title of Lesson 103

EWMA chart
One thus takes for lower and upper limits of the limit of when one makes
tend i towards +/- infinity (this limit being reached very quickly when i
increases).
● The control limits are therefore expressed as :
i.e after simplification when i tends towards infinity :
Title of Lesson 104

EWMA chart : example

For our example : see page in the excel file « SPC example »
Same data for both charts : deviation

is more visible with EWMA chart
For this example : L=3 ; λ=0,2
Title of Lesson 105

EWMA chart
● The limits therefore depend on the number of the sample, but they
converge very quickly to a straight line when i increases, which is the reason
for the mathematical simplification. A complementary reasoning in deviations
from the target can be done in the case of small series when the process
performs a repetitive step.
● The parameters L, n and must be defined during the study. These

parameters are defined in terms of δ: the deregulation. The following
Remenber :
summary table is used to select these parameters:
λ is a ponderation between present and past : the lower it is the higher the weight of the past is => the higher the sensitivity to small
drifts is.
However : if λ is small sensitivity to rapid drift is not good.
If you choose λ=1 you are back to Shewart chart (X chart).
Title of Lesson 106

THANKS FOR YOUR

ATTENTION
Title of Lesson 107

Documents Management
Qualité 1 – Chapitre 3 – Coûts d’obtention de la qualité
Title: ID No.:
(COQ)
Autor: Sébastien Lasserre Date: 30/08/2021
Version: 1.0 Format: ppt
Root:
Description:
VERSION DATE AUTOR DESCRIPTION OF MODIFICATIONS
Sébastien
1.0 30/08/2021 Initial document
Lasserre
2,0 19/02/2023 Laurent SESTIER
Title of Lesson 108

Module III - Statistical Process Control

Uploaded by

Copyright:

Available Formats

Module III - Statistical Process Control

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Module III - Statistical Process Control

Uploaded by

Copyright:

Available Formats

ID document: e.g.

2 – Common causes and Special causes

Targets of today lecture :

● Understand what « capability » means for a mass production industry

● Understand how is used a control chart

● Know how to build Shewhart control charts

● Understand the advantage of Cusum and EWME charts compared to Shewhart

Reminder from the last lecture :

Mass production : too costly to check each piece

Measures from the sample of 30 nuts

despite nuts are

2 – Common causes, special causes

What creates dispersion of dimensions ?

What creates dispersion of dimensions ?

-initial thickness of the wood piece -temperature of the machine

Causes we can eradicate Causes we will have to

Special causes Common causes

In industry for any process, causes of

Common causes - Examples

- Gaps in the kinematic chain of the machine

- Defects in the spindle (=broche) of the machine

- Variation of raw material properties in the same batch

● Example of special causes:

Common causes specificity

Any system, subjected to many factors, independent of each other,

Reminder about Gauss curve (normal law)

- 95.44 % of the values are distributed

- 99.73 % of the values are distributed

The last result is important because it

After eradication of special causes, process result is predictable

No part out of the tolerance

Mean value of the production

Variation due to common causes

Requirement for finished products with low rate of

Red component : defective

Light brown : component

Requirement for finished products with low rate of

Having a low rate of defective for finished products requires to be

Even if few % of the production is out of the specification it is not

Capability is the measure establishing the relationship between the actual

A process is capable if C > 1.33

Non-capable process : Capable process :

● On the left, a non-capable process. The dispersion is larger than the

Capability – Different cases

- Analysis: The curve modeling the total

- Action: No adjustment possible.

Capability – Different cases

- Analysis: The curve does not fit within the

- Action: No adjustment is possible, and there are

Capability – Different cases

- Analysis: The curve modeling the total

- Action: Adjustments are possible, as well as a

Capability – Different cases

- Analysis: The fault is due to a major

- Action: Do not exceed the drift (=dérive) and

B – Cp, Cpk, Cm, Cmk, Pp, Ppk - Long-term indicators

- Cp and Cm indicators: Short-term indicators that reflect the dispersion over

- Pp indicators: Long-term indicators that reflect the reality of delivered

Short term capabilities

Short term capability Cm & Cp