ID document: e.g.

PPT/L-TCIO-1

Chapter IV –
Statistical Process
Control & Cost of
Operational quality and Lean
management - S7 EENG 4
quality
Title of Lesson
Title of Lesson 1
ID document: e.g. PPT/L-TCIO-1

Chapter organization
1 – Introduction / reminders about control charts

2 – Attribute control charts

3 – Cost of Quality

Title of Lesson 2
ID document: e.g. PPT/L-TCIO-1

Targets of today’s lecture :

● Understand the difference between measurement control chart and attribute
control chart

● Understand how is used an attribute control chart SPC

● Know how to build attribute control charts

● Understand what are internal and external anomalies

● Make the difference between prevention costs and detection costs COQ

● Understand what’s the meaning of « Cost of Quality »

Title of Lesson 3
ID document: e.g. PPT/L-TCIO-1

Reminder from the last lectures :

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 4
 Common causes specificity
ID document: e.g. PPT/L-TCIO-1

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 5
ID document: e.g. PPT/L-TCIO-1

Reminder about Gauss curve (normal law)

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

- 95.44 % of the values are distributed

around the mean between ± 2 standard
deviation

- 99.73 % of the values are distributed

around the mean between ± 3 standard
deviation

The last result is important because it

represents the dispersion
Title of Lesson 6
ID document: e.g. PPT/L-TCIO-1

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 7
ID document: e.g. PPT/L-TCIO-1

SPC – Introduction
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 8
ID document: e.g. PPT/L-TCIO-1

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 9
ID document: e.g. PPT/L-TCIO-1

4 – Statistical Process Control

-n chart

-p chart
D – Attribute control charts
-c chart

-u chart

Title of Lesson 10
ID document: e.g. PPT/L-TCIO-1

Quality is sometimes not measured through

dimensions
Exemple with electronic soldering

Result of the checking :

compliant or
non compliant

Components soldering is key for electronic circuit

Title of Lesson 11
ID document: e.g. PPT/L-TCIO-1

dimensions measurement can be too difficult

Exemple with diameter measurement with a gauge instead of a
caliper
Side 1 : should enter

Result of the checking :

compliant or
non compliant

Side 2 : should not enter

Title of Lesson 12
ID document: e.g. PPT/L-TCIO-1

Target : same as for Measurement

control chart
During mass production,
What criteria will give the statistic
the questions are :
Is the machine setting still good?
proof
Attributethat the controlprocess charthas is the
drift
Is there any special cause occuring?

statistic
and that
tool it’s
to make
necessarythe decision
to
The context is different as we don’t have measurement results
take
After checking a sample, weactions
have a number?of compliant parts
and a number of non compliant parts
Title of Lesson 13
ID document: e.g. PPT/L-TCIO-1

Attribute control charts set up

● As with the use of a measurement chart, the purpose of attribute charts is to
separate the extraordinary from the ordinary variations, alerting the
operator to a special cause

● The method of setting up an attribute chart is the same as for a measurement

chart :

- Observation (charts without limits, at least 20 samples)

- Calculation of the limits (corresponds to the maximum fluctuations of the
common causes)
- Control (documented intervention when a special cause appears)
- Improvement (reduction of variations and recalculation of limits).

Title of Lesson 14
ID document: e.g. PPT/L-TCIO-1

Attribute control charts driving rules

● The principle of the driving rules is identical to that of an measurement
chart.

● However, one must be aware that in the case of a judgment made by an

individual, a change in the control criterion can be the cause of a trend
(increasing, decreasing, lower, higher).

● To ensure effective attribute-based control, it is necessary to minimize

variations in judgment by clearly defining what is a compliant product, and
what is a non-compliant product.

Title of Lesson 15
ID document: e.g. PPT/L-TCIO-1

Attribute control charts

● When establishing conformity criteria, it is therefore necessary to ensure that
certain factors of variation are controlled:

- Catalog of defects (samples, photos, sketches, ...)

- Personnel capacity (fatigue, formation, ...)

- Training (detection technique, judgment, ...)

- Environment (lighting, noise, ...)

- Capability of the measurement process (homogeneity test)

Title of Lesson 16
ID document: e.g. PPT/L-TCIO-1

Attribute control charts

● If we are interested in the non-conforming product, we will use an n-card or a
p-card.

- n chart: chart with the number of non-conforming products

- p chart: chart with the proportion of non-conforming products

● If we are interested in defects, we will use a c-card or a u-card.

- c chart: chart with the number of defects

- u chart: chart of the number of defects per unit (proportion)

The n and c charts use a constant sample size, while with the p and u charts, the
sample size can vary slightly ( 25%).
Title of Lesson 17
ID document: e.g. PPT/L-TCIO-1

Different attribute control charts

2 defects on the same 2 defects on the same

circuit board = 1 non circuit board =
compliant product 2 non compliancies

Non compliant Non compliant

products units
Number n chart c chart
Proportion p chart u chart

Based on binomial law Based on Poisson law

Title of Lesson 18
ID document: e.g. PPT/L-TCIO-1

Non compliant Non compliant

products units

n chart Number
Proportion
n chart
p chart
c chart
u chart

Title of Lesson 19
ID document: e.g. PPT/L-TCIO-1

n chart
● The calculation of the control limits is done according to the binomial law.

With as parameters :

- n : number of pieces in the sample

- p : average proportion of non-conforming products
- q = 1-p

Title of Lesson 20
ID document: e.g. PPT/L-TCIO-1

Example:
Step 1 : data collection Step 2 : limit calculation
Production and use of the chart

Title of Lesson 21
ID document: e.g. PPT/L-TCIO-1

n chart
● The lower control limit does not have to be shown on the control chart.

● It is indeed difficult to decide to intervene when the proportion of non-

conforming units has decreased. However, this limit is interesting for the
analyst because it indicates an improvement of the process. A tightening of
the limits is necessary if this is really the case.

● The sample size is based on the average number of non-conforming

products, in order to find several non-conforming products per sample.

Title of Lesson 22
ID document: e.g. PPT/L-TCIO-1

Non compliant Non compliant

products units

p chart Number
Proportion
n chart
p chart
c chart
u chart

Title of Lesson 23
ID document: e.g. PPT/L-TCIO-1

p chart
● This chart differs from the map with the number of non-compliant products
only by the scale of the vertical axis which is in % (proportion of non-
compliant products in the sample) instead of representing a number of non-
compliant products.

● The main advantage of this map is that the sample size can vary within 25%
without affecting the control limits.

● We say that the sample is pseudo-constant. This advantage compensates for

the additional calculation generated by this card.

Title of Lesson 24
ID document: e.g. PPT/L-TCIO-1

p chart
If there are large variations in the sample size (> 25%), it is possible to plot several
control limits on the same chart.

● The control limits for the p-chart are obtained using the formula :

Title of Lesson 25
ID document: e.g. PPT/L-TCIO-1

Example:
Step 1 : data collection Step 2 : limit calculation
Production and use of the chart

Title of Lesson 26
ID document: e.g. PPT/L-TCIO-1

Non compliant Non compliant

products units

c chart Number
Proportion
n chart
p chart
c chart
u chart

Title of Lesson 27
ID document: e.g. PPT/L-TCIO-1

c chart

● This chart, established on the basis of Poisson's law.

● The determination of the control limits is made from the average number of
non-conformities per unit (m) on the same basis as for a chart with the
number of non-conforming units (products) since the number of non-
conformities can only be an integer (discontinuous variable).

Title of Lesson 28
ID document: e.g. PPT/L-TCIO-1

c chart
● The control limits are obtained using the formula :

With as parameter:

- m : average number of non-conformities per unit

Title of Lesson 29
ID document: e.g. PPT/L-TCIO-1

Example : painted components

Different kinds of defects can be found

drip spot bubles Orange skin

50 pieces are checked and the number of defect (whatever

the kind of defect) is recorded

The ULC is then calculated for further checking of the

production

More than 13 defect on the same component will require to stop

production and look for causes of deviation

Title of Lesson 30
ID document: e.g. PPT/L-TCIO-1

c chart

● The non-conformance chart (also called a multi-attribute chart) is extremely

practical when several non-conformances may appear on the same unit.

● However, the determination of control limits only makes sense if all non
conformities have the same weight.

Title of Lesson 31
ID document: e.g. PPT/L-TCIO-1

Non compliant Non compliant

products units

u chart Number
Proportion
n chart
p chart
c chart
u chart

Title of Lesson 32
ID document: e.g. PPT/L-TCIO-1

u chart
● As with the p-chart, this chart differs from the non-conformity chart only in
the scale of the vertical axis, which is in % (proportion of non-conformities
in the sample).

● The sample size can also vary within a range of 25% without influencing the
position of the control and monitoring limits too much.

Title of Lesson 33
ID document: e.g. PPT/L-TCIO-1

u chart

● These limits are obtained by the following formula:

LSCu , LLCu =

u̅ : rate of non-conformities per unit

n : number of pieces in the sample

Title of Lesson 34
ID document: e.g. PPT/L-TCIO-1

4 – Costs of obtaining quality

(COQ)

Title of Lesson 35
ID document: e.g. PPT/L-TCIO-1

Costs of obtaining quality – Goals

● The cost of obtaining quality is the amount of money that must be spent to achieve
an optimal quality approach. It is also known as COQ.

● The goal is to invest just what is needed to obtain optimum quality. It is a matter
of getting the best return on investment.

Title of Lesson 36
Back to the example of the fast food restaurant …

Remember quality for my hamburger :

-size (as big as on the advertisement) -no risk for health (no bacteria contamination in the food)
-tasty -nutritious benefits (protein, vitamin etc.)
-at the good temperature
-delivered in a short time
Etc.

Examples of costs of obtaining quality in the fast food :

Training of employees
Time spent for hand washing everytime entering the kitchen
Checking of the steaks good cooking
Checking of the oil quality in the fryers
…
ID document: e.g. PPT/L-TCIO-1

5 – Costs of obtaining quality

- Costs of obtaining quality

- Costs of non-quality
A –Costs of obtaining quality
and non-quality

Title of Lesson 38
ID document: e.g. PPT/L-TCIO-1

Costs of obtaining quality and non-quality

In 1986, AFNOR (French Association for normalization) published a documentation
brochure in which the classification of quality costs is as follows

- Internal anomalies (or internal non-conformities)

The costs of non quality:
- External anomalies (or external non-conformities)

- Detection
The costs of obtaining quality:
- Prevention

Title of Lesson 39
Internal issues , external issues
Internal anomalies : problem detected by the company before the delivery to the customer
Comes from : Consequences :
Products design difficult to manufacture
Management of manufacturing Loss of time and money to scrap or rework goods
Machines stability Delay in the deliveries
Suppliers bad quality Loss of competitivness in the market
Etc. Etc.

external anomalies : problem detected by the customer (the company failed to detect
it before delivery)

Consequences :
Comes from :
Loss of time and money to settle issues with customers.
Above causes + Products design not robust
After sales costs
+ weak control process
Bad brand image, decrease credibility for customers
Etc.
 Back to the example of the fast food restaurant …
Internal anomaly or External anomaly ?

5 over cooked steaks had to be put in the bin by the employee Internal anomaly

After a mistake in the boxes an employee spent time to repack some sandwiches Internal anomaly

A milkshake was given free of charge to a customer after he came back to claim that a sandwich was missing when
he got his order
External anomaly

When the restaurant closed, 2 remaining sandwhiches not sold on the shelves after a mistake in the orders

Internal anomaly
ID document: e.g. PPT/L-TCIO-1

Costs of obtaining quality

(COQ)

Title of Lesson 42
ID document: e.g. PPT/L-TCIO-1

Costs of obtaining quality

● To achieve quality, additional expenses are required:

- Value analysis in the design phase

- Product certification costs
- Staff awareness/training
- Prevention/anticipation costs (preventive maintenance)
- Continuous improvement

Title of Lesson 43
ID document: e.g. PPT/L-TCIO-1

Costs of obtaining quality

These expenses include prevention and detection:

● Prevention :

- Staff awareness

- Internal training

- Evaluation of suppliers (quality of raw materials, subcontracting)

- Management costs associated with an ISO standard (certification)

Title of Lesson 44
ID document: e.g. PPT/L-TCIO-1

Costs of obtaining quality

These expenses include prevention and detection:

● Detection :

- Verification of product conformity

- Purchase of control and measurement equipment

- Calibration (= étalonnage) of control and measurement equipment

- Training of personnel to implement the controls

Title of Lesson 45
 Back to the example of the fast food restaurant …
Prevention cost or Detection cost?

1 hour spent for a new employee training about hygiene procedure Prevention cost

Checking of the temperature of the frozen food into the freezer Detection cost

Checking of the oil quality in the fryers Detection cost

Time spent to write a new procedure for the food stock management to avoid scraps

Prevention cost
ID document: e.g. PPT/L-TCIO-1

Costs of obtaining quality

•It is not necessary to have the perfect quality either.

corresponding to the expectations of the customer

 not to spend too much and reach an over-quality and thus lose money.

Title of Lesson 47
ID document: e.g. PPT/L-TCIO-1

5 - Iceberg of the quality

Title of Lesson 48
ID document: e.g. PPT/L-TCIO-1

Iceberg of the quality – Definitions

90% of the mass of the iceberg is below the surface of the sea, invisible to the eye, as it
is often the case in business.

This means that 90% of the company and its environment is partially or completely
unknown to the company's managers and employees, despite the sophistication of
the information systems and management methods.

What we see, what we know, what we understand and what we want to see, know
and understand in a company or organization concerning its context, strategies,
processes, resources and intangible factors is 10%, regardless of the position or
department.
Title of Lesson 49
Examples of non Quality hidden costs
A company decides to rework a batch of products after discovering a quality problem

The time spent by an employee to rework the products will probably be recorded

BUT : the below costs will very likely not be included in the rework cost in a big company

Time spent to unload the products Containers hiring because Warehouse hiring because
no more available for the no more space to store the

…
Scraps of packaging other batches product in the current
facility

The costs are recorded in separate IT system which makes it difficult to merge
ID document: e.g. PPT/L-TCIO-1

Iceberg of the quality – Total costs

Scraps, wastes
We talk about the iceberg of the Visible
quality, because only one part of Repairs, maintaining
costs
the costs is visible : Client returns
-scrap,
-repairs, Cost of time spent to handle
quality issues unvisible
-customer returns,
costs
- etc. Logistic costs
Disorganization :
consequences like delay in Customer loss
But many other costs, which are deliveries
not directly identifiable, take a Bad brand image
Useless scraps
much larger share. Useless energy and raw
materials consumption

Title of Lesson 51
Iceberg of the quality
It’s crucial for a company to track hidden costs to take good decisions for quality

1rst situation : Low tracking of

hidden cost. Only
50k€ of Non Quality
50k€
cost is identified
Changing this
Decision : No Go
old machine will
eradicate quality
issue. The cost Better tracking of
2nd situation :
of the new hidden cost. 120k€ of
Non Quality cost is
machine = 100k€ identified
120k€

Decision : Go
ID document: e.g. PPT/L-TCIO-1

THANKS FOR YOUR ATTENTION

Title of Lesson 53
ID document: e.g. PPT/L-TCIO-1

Documents Management
Title: Qualité 1 – Chapitre 3 – Coûts d’obtention de la qualité (COQ) ID No.:

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 05/03/2023 Laurent Sestier

Title of Lesson 54