You are on page 1of 60

# Recap Lecture 3: Measured data and Statistics

## Able to do measurement: significant figures, rounding, rules on

multiplication, division, additional and subtraction, rounding rules
& errors
Understand the variations and attributes
Know what statistic is and its applications, distributions and how
they are used in SPC
Able to calculate the mean, median, mode, range and standard
deviation and to draw a histogram

Histogram

Frequency

Interval, i

Mid-point
Upper boundary

Category

Lower boundary

Descriptive Statistics

## Measure of central tendency- describes the center position of

the data (mean, median, mode)

## Measure of dispersion describe the spread of data (range,

variance, standard deviation)

Mean, X

1 N
i 1 X i where Xi is one observation, N is number of sample
N

Median is the middle point of a data series (observation in the middle of sorted
data
Mode the most frequently occuring value

100 91 85 84 75 72 72 69 65
Mean = 79.22

Mode
Median

Descriptive Statistics
Measure of Dispersion - Kurtosis

## Kurtosis Measure the peakness of the data. It is a dimensionless

value. The value must be compared to a normal distribution to determined
if it more peaked or flatter peaked distribution.

a4

f i ( xi X) / n
4

i 1

## Leptokurtic (more peaked)

s4

Mesokurtic (normal)

Platykurtic (flatter)

Descriptive Statistics
Measure of dispersion (range, variance and standard deviation)

## The range is calculated by taking the maximum value and

subtracting the minimum value.
Variance is the squared of the summation of the difference between
each value and the mean divided by number of samples
n

( xi

i 1

= population mean

1, 3, 5, 7, 9, 11
Range = 11-1 = 10

n

(
x

i 1

## If is small, high probability of

getting the values close to mean
value
If is large, high probability of getting
the values away from mean value

Descriptive Statistics
Other measure of dispersion (skewness, kurtosis,
coefficient of variation)

## Skewness - lack of symmetry of data distribution. A negative values

indicate skewed to the left, positive indicates skewed to the right.
<0 left
0 = symmetrical >0 right

fi ( xi x) / n

a3 i 1

## Introduces the QC tools

When to use, how to use the tools
Example problems

Outline
Statistical
Process
Control
Quality
Function
Deployment
Failure Mode
Effect
Analysis
(FMEA)
Taguchis
Quality
Engineering

ISO 9000

Acceptance
Sampling

Reliability

Design of
Experiments
(DOE)

Products
Liability

Quality by
Design

Management
and Planning
Tools

Information Technology
Computer Program

ISO 14000

Benchmarking

Total Productive
Maintenance

Learning Objectives
Construct a Pareto diagram.
Explain how to construct a cause and effect diagram.
Explain how to construct a check sheet.
Know the major sections of ISO 9000 and briefly describe their
emphasis.
Know the objectives, techniques, procedures, and benefits of an internal
audit.
List the quantitative and qualitative TQM tools and techniques

## TQM Tools and Techniques

1. Statistical Process Control (SPC)
Pareto Diagram
Cause-Effect Diagram
Check Sheets
Process Flow Diagram
Scatter Diagram

Histogram
Control Charts

## TQM Tools and Techniques

2. Acceptance Sampling

3. Reliability
4. Design of Experiments (DOE)
5. Taguchis Quality Engineering
6. Failure Mode and Effect Analysis
(FMEA)
7. Quality Function Deployment (QFD)
8. ISO 9000

## TQM Tools and Techniques

9. ISO 14000

10. Benchmarking
11. Total Productive Maintenance
12. Management and Planning Tools
13. Quality by Design
14. Information Technology

## Other TQM Tools

Acceptance Sampling
See

Chapters 8 and 9

Reliability
See

Chapter 11

TQM Tools
Design of Experiments (DOE)
Used

## to determine those variables in a

process that are critical and their target
values

Three

approaches

Classical
Taguchi
Shainin

TQM Tools
Taguchis Quality Engineering
Loss

function concept

Concept

of robustness

Combines

## technology and experience of

people to identify failure modes and
planning for its elimination

It

is a before-the-event action

Two

## types of FMEA-Design and Process

TQM Tools
Quality Function Deployment (QFD)
Ensures

What
Are

## the following questions:

do customers want?

Will

How

## can we change the product, service or

process?

TQM Tools
Quality Function Deployment (QFD) Contd.

How

## does an engineering decision affect

customer perception?

Does

## an engineering change affect other

technical descriptors?

What

## is the relationship to parts deployment,

process planning, and production planning?

## Statistical Process Control

(SPC)
A methodology for monitoring a
process to identify special causes of
variation and signal the need to take
corrective action when appropriate

## Statistical Process Control

Seven Tools:
1.

Pareto Diagram

2.

Cause-Effect Diagram

3.

Check Sheets

4.

5.

Scatter Diagram

6.

Histogram

7.

Control Charts

## The Pareto Principle

Vilfredo Pareto was an economist who is
credited with establishing what is now widely
known as the Pareto Principle or 80/20 rule.
When he discovered the principle, it
established that 80% of the land in Italy was
owned by 20% of the population. Later, he
discovered that the pareto principle was valid
in other parts of his life, such as gardening:
80% of his garden peas were produced by
20% of the peapods.

## The Pareto Principle

Some Sample 80/20 Rule Applications
80% of process defects arise from 20% of the process

issues.

revenues.
80% of delays in schedule arise from 20% of the possible

## causes of the delays.

80% of customer complaints arise from 20% of your

products or services.
(The above examples are rough estimates.)

## The Pareto Diagram

Graph that ranks data classifications in

## descending order from left to right

Pareto diagrams are used to identify the most

important problems
Advantage: Provide a visual impact of those

## vital few characteristics that need attention

Resources are then directed to take the

## The Pareto Diagram

Helps a team focus on causes that have the

greatest impact
Displays the relative importance of problems

## in a simple visual format

Helps prevent shifting the problem where

## the solution removes some causes but

worsens others

Pareto Chart
A plot look like histogram, but prioritize the data according to their significance
in relation to a problem/effect. For example, the types of defects on a painted
surface can be blisters, spot, bubbles, blur mark. These defects can be
organized according to their significance/contribution to the poor paint quality.
Pareto Chart separates the vital few problems from the trivial many.
Example of nonconforming shirt in a week.

## When to use Pareto Chart?

When analysing data about the frequency of problems or causes in a process.
When there are many problems or causes and you want to focus on the most
significant.
When analysing broad causes by looking at their specific components.

Pareto Chart
Exercise: given below the types of complaints and the number of complaints in
customer satisfaction survey at a restaurant, construct a Pareto Chart
Complaints
Cold food
Flimsy Utensils

Number of complaints
105
2
10
94

Complaints
Number of Complaints
Poor Service
13
Food too greasy
9
Lack of courtesy
2
Lack of cleanliness
25

Scatter Diagram

People drowning
Sale of ice-cream

Positive correlation
Sale of ice-cream

Sale of ice-cream

Poor correlation

Negative correlation

People drowning

People drowning

Even there is
correlation, it does
not tell causal
relationship (e.g.
what cause what).
Other tool need to
be used
At best correlation
can tell a potential
relationship which
may be confirmed
by other method

## The degree to which two sets of data are related and

may be mathematically calculated with the correlation
coefficient
The simplest way of visualizing the correlation is with a
Scatter Diagram, where perfect correlation is a straight
line and no correlation is a random cloud of points.
No correlation is a positive result: it tells you clearly
that the two sets of data are not related. The most
frustrating picture is when there is a weak correlation
which only says that there might be a weak or distant
relationship.

Scatter diagram

## For each first item measurement,

there maybe a range of possible
second item measurements, and
vice versa

## Each time the first is measured,

the second item is also measured,
and this pair is plotted on the
Scatter Diagram

## First item measurement

Cause-and-Effect Diagram
It was developed by Dr. Kaoru Ishikawa in 1943
Picture composed of lines and symbols

## designed to represent a meaningful relationship

between an effect and its causes
Effect (characteristics that need improvement)

## on the right and causes on the left

Cause-and-Effect Diagram
People

Materials

Work Methods

Primary
Cause

C
a
u
s
e
s

Quality
Characteristic

Secondary Cause

Environment

Equipment

Measurement

## Figure 3.3 Cause-and-Effect Diagram

E
f
f
e
c
t

Cause-and-Effect Diagram
Enables a team to focus on the content of a

## problem, not on the history of the problem or

differing personal interests of team members
Creates a snapshot of collective knowledge and

## consensus of a team; builds support for solutions

Focuses the team on causes, not symptoms
Used to investigate either a bad effect and to

## take action to correct the causes or a good effect

and to learn those causes responsible

Cause-and-Effect Diagram
Steps in the construction of a
Cause-and-Effect Diagram:
1.

2.

3.

## Determine all the minor causes. Request a

brainstorming session

4.

## Once the diagram is complete, evaluate it to

determine the most likely causes

5.

Develop solutions

Cause-and-Effect Diagram
1.

## Analyzing actual conditions for the purpose of

product or service quality improvement

2.

## Elimination of conditions causing

nonconforming product or service and
customer complaints

3.

operations

4.

## Education and training in decision-making

Cause-and-Effect Diagram
Types of Diagrams:
1.

## Each major branch is filled in completely

before starting work on any of the other
branches. The objective is to analyze the
causes of dispersion or variability

Cause-and-Effect Diagram
Types of Diagrams:
2.

## In order to construct it, it is necessary to

write each step of the production process.
The advantage of this type of diagram is
the ease of construction and its simplicity,
since it follows the production sequence

Cause-and-Effect Diagram
Also Called:, Ishikawa Diagram, Fishbone Diagram
Variations: cause enumeration diagram, process fishbone, time-delay fishbone,
CEDAC (cause-and-effect diagram with the addition of cards), desired-result
fishbone, reverse fishbone diagram.
Identifies many possible causes for an effect or problem. It can be used to
structure a brainstorming session. It immediately sorts ideas into useful categories

Materials

Procedures

Quality
Problem

People

Equipment

## Cause-and-Effect Diagram Procedures

Agree problem statement (effect).
Brainstorm the major categories :
Methods,
Machines (equipment),
People (manpower),
Materials, Measurement,
Environment
Causes as branches from the main arrow.
idea
Causes
categories.
Again ask why does this happen?
sub-causes branching off the causes. Why? generate deeper levels of causes.
Layers of branches indicate causal relationships.
No more ideas
focus attention to places on the chart where ideas are few.
Applications : Problem statements, Interview, Recruitment and etc.

Check Sheet

## Also known as defect concentration diagram, data collection sheet or

tally chart
Generic, flexible for any purpose of data collection and analysis

Check Sheets

27

Check Sheets

28

Check Sheet

## When to Use Check Sheet

data observed and collected repeatedly -same person or location.
collecting data -frequency or patterns of events, problems, defects,
defect location, defect causes, etc.
collecting data -a production process.
Data to be recorded manually-accurate direct interpretation or for
transcription, example into a computer.
recording involves counting, classifying, checking or locating.
checking each measurement as it is recorded, for example within
normal bounds.
the distribution of measures as built up

Bar Charts
Variety of fruits in
6

## Data pattern can

be visualised

4
2

Apple
2

Orange Banana
5
6

Plum
1

Bar Charts
difference between a small set of related data.
flexible, show multiples sets of data in stacked, grouped display.
simple, portray data in a visual and easily-understood form.
Organize numeric measurements -tables -to understand, but lacks visual impact,
with trends and relative sizes difficult to discern.
measurements in discrete physical bars, The area of the bar is proportional to the
size of the measurement, visual impression.
the relative sizes physically by separated bars. Even complex trends across
multiple bars may be apparent (Line Graphs and Control Charts better -measuring
trends).
flexible format -used in a wide number of situations (this may be contrasted with
the specific purpose of the Histogram or Pareto Chart). The independent nature of
each bar enables even quite disconnected items to be compared.

Histogram
Shows data distribution and how frequent a range of value occur
Can tell the capability of a process
Only numerical values can be used

9.
9
10
.9
11
.9
12
.9
13
.9
14
.9
15
.9
M
or
e

8.
9

7.
9

6.
9

5.
9

4.
9

3.
9

2.
9

1.
9

Frequency

Histogram

25

20

15

10

Category

## When to use Histogram

Numerical data.
shape of the datas distribution, determining the output of a process is
distributed approximately normally.
analysing whether a process - meet the customers requirements.
Analysing the output from a suppliers process looks like.
seeing whether a process change has occurred
Determining the outputs of two or more processes are different.
to communicate the distribution of data quickly and easily to others.

## Stem and leave plot

Sort-out data into an array that can be read easily
Easy and efficient method of organizing data from smallest to largest
when data are of the same accuracy and very close to each other.
Example: Create a stem plot from the measurements data for diameter
of a rod:
2.36, 2.39, 2.51, 2.43, 2.56, 2.72, 2.45, 2.83, 2.75, 2.27, 2.63, 2.92, 2.88, 2.44, 2.57, 2.61,
2.58, 2.81, 2.81, 2.43, 2.62, 2.96, 2.48, 2.31, 2.67

2.2
7
2.3
6 9 1
2.4
3 5 4 3 8
2.5
1 6 7 8
2.6
3 1 2 7
2.7
2 5
2.8
3 8 1 1
2.9
2 6

2.2
7
2.3
1 6 9
2.4
3 3 4 5 8
2.5
1 6 7 8
2.6
1 2 3 7
2.7
2 5
2.8
1 1 3 8
2.9
2 6

## Location: Middle value is between 2.5 and 2.6

Spread: Smallest value is 2.27, largest value is 2.96
Shape: Bimodal pattern with high points at the stem 2.4 and 2.8

Flow Diagram/Chart
Flow Chart Graphical description how a work is done. Uses symbols.
Used to describe processes to be improved
Symbols used by
General Motors Corp
I

Inspection

Operation

Storage

Product movement
Delay

Analyst: RMN

Description
of
process

Move to storage

Move to peeler

15

## Soak in water until needed

20

Place in conveyor

20
100 ft
30
50 ft
360
20 ft

10

11

Page 1 0f 3

Distance
(feet)

Time
(min)

Date: 9-30-00

Operation
Transport
Inspect
Delay
Storage

Step

Process Chart

Total

20 ft
30
480

190 ft

## Other Process Descriptive Tool - Story Board

Story board is a recording a related events/effects from the source to the
end (results of those events)
Originally developed by Walt Disney in creating their storyline. Story
boards are put over walls where everyone involved look at it and try
different things.
Allows for tracing back the source of a problem.

## Control Chart Run Chart

Run Chart is a time series plot that monitor the process
changes associated with a particular characteristic over time
Can use either variable or attribute data
x-axis shows time, y-axis shows attribute or variable value
Run Chart is the basis for Control Chart

Control Chart
27
24
UCL = 23.35

Number of defects

21

c = 12.67

18
15
12
9
6

LCL = 1.99

3
2

10

12

Sample number

14

16

Benchmarking
Benchmarking is a new way of doing
business that was developed by Xerox in
1979. The idea is to find another company
that is doing a particular process better than
your company, and then, using that
information to improve the process

## Total Productive Maintenance

Total Productive Maintenance (TPM) is a
technique that utilizes the entire work force to
obtain the optimum use of equipment.
The technical skills in TPM are: daily equipment
checking, machine inspection, fine-tuning
machinery, lubrication, trouble-shooting, and
repair.

## Management and Planning

Tools
Management and Planning Tools (Chapter 12):
1.

Affinity diagram

2.

Interrelationship diagram

3.

Tree diagram

4.

Prioritization matrices

5.

Matrix diagram

6.

7.

## Activity network diagram

Quality by Design
Quality by Design is the practice of using a
multidisciplinary team to conduct product or
service conception, design, and production
planning at one time.
The major benefits are faster product
development, shorter time to market, better
quality, less work-in-process, fewer
engineering change orders, and increased
productivity

Products Liability
Consumers are initiating lawsuits in record numbers
as a result of injury, death, and property damage
from faulty product or service design of faulty
workmanship.
Reasons for injuries:
Behavior or knowledge of a user
Environment where the product is used

## product using safety analysis and quality control

Information Technology
Information Technology is defined as
computer technology (either hardware or
software) for processing and storing
information, as well as communications
technology for transmitting information.

## Thank you for your attention.

Happy Thaipusam 18-1-11
Next week class QC on Thur 27-1-11 at 1112 am DK5.
Then Quiz 1 covers 1 hour 12-1 pm(open
book) at DK5 after 1 hour lecture