Design and Analysis

of Experiments

Chapter 1

INTRODUCTION

Dr. Tran Thanh Hung

Department of Automation Technology,
College of Engineering, Can Tho University
Email: [email protected]
 Learning methods

Learning methods:
• Lecture + Textbook
• Exercise
• Project
Grading:
Exercise (50%) + Project (50%)
(Chapter 2-8: x 1, Chapter 9: x 3)

*Work in group 4-5 students, for both exercise and project2

 Quy ước nộp bài

- Báo cáo tất cả bài tập mỗi chương trên 1

file word, đặt tên file Bai tap chuong
x_Nhom y_Lop z
(Bài báo cáo phải ghi tên các thành viên tham
gia làm bài)
- Mỗi bài tập phải nộp file Minitab
- Nộp vào ĐÚNG thư mục của lớp trong trên
hệ thống E-learning, không cần nén file.
- Đánh x vào bài đã nộp trong danh sách lớp.
- Bài báo cáo giống nhau sẽ bị điểm 0. 3
 Textbooks

[1] Design and Analysis of Experiments by

Douglas C. Montgomery, ebook
[2] A First Course in Design and Analysis of
Experiments by Gary W. Oehlert, ebook
[3] Statistical design and analysis of
experiments with applications to engineering
and science by Robert L. Mason, Richard F.
Gunst, and James L. Hess, ebook
4
 Why Experiment?
In engineering: to study the
In general:
performance of processes
-To answer questions and systems:
-To validate scientific hypotheses - Determine which variables
-To discover something are most influential on the
response y,
- Determine where to set x so
that y is almost always near
the desired value,
- Determining where to set x
so that variability in y is small
- Determine where to set x so
that the effects of z are 5
General model of a process or system minimized.
 Why Design of Experiments
(DOE)?
In engineering perspective: We use DOE to:
• Improved process yields/quality
• Reduce time to design/develop new products
& processes
• Reduced variability and closer conformance to
nominal or target requirements
• Reduced overall costs

6
 Course objectives
At the end of this course, students can:
1. understand the concepts, principles, steps in designing and
analyzing experiments,
2. use methods and tools for designing and analyzing
experiments,
3. identify and resolve problems within their discipline,
4. plan and perform experiments to solve the problems in their
discipline,
5. collect and analyze data to solve problems in their
discipline,
6. evaluate experimental designs.
7
 Chapter objectives

At the end of this chapter, students can:

1. understand the concepts and principles
of DOE,
2. understand experimentation is a
process,
3. list the steps for designing experiments,
4. recognize the key historical figures in
DOE.
8
 Basic Concepts of DOE

1. Factors:
Experimental factors: independent variables
that an experimenter manipulates to capture any
changes in the output of the process.
Other factors of concern are those that are
uncontrollable and those which are controllable
but held constant during the experimental runs.
Nuisance factors: factors that may influence
the experimental response but in which we are
not directly interested.
9
 Basic Concepts of DOE

2. Responses: dependent variables measured to

describe the output of the process.

3. Treatment Combinations (run): experimental

trial where all factors are set at a specified level.

10
 The Basic Principles of DOE

There are 4 basic principles of DOE.

The first three are perhaps the most important.

• Randomization

• Replication

• Blocking

• Multi-factor Designs

11
 The Basic Principles of DOE

1. Randomization:
To eliminate potential biases from the conclusions, the
experimenter have to randomize in:
- selection of the experimental material, people
- order in which the individual runs of the experiment are
performed
- any other factor not under the control of the
experimenter.
 “averaging out” the effects of the extraneous factors
(yếu tố không liên quan) that may be present in the
experiment. 12
 The Basic Principles of DOE

2. Replication
Repetition of a basic experiment without changing any
factor settings, allows the experimenter to estimate the
experimental error (noise) in the system.
It is used to determine whether observed differences in
the data are “real” or “just noise”, allows the experimenter
to obtain more statistical power (ability to identify small
effects).

Replication reflects sources of variability both between

runs and (potentially) within runs.
13
 The Basic Principles of DOE

3. Blocking (đóng khối):

Used to increase the precision of an experiment by

breaking the experiment into homogeneous segments
(blocks) in order to control any potential block to block
variability (multiple lots of raw material, several shifts,
several machines, several inspectors).

Any effects on the experimental results as a result of the

blocking factor will be identified and minimized.
14
 The Basic Principles of DOE

4. Multi-factor Designs:

- In each run, only one factor is changed while the others is

held constants  One-factor design.

- In each run, all factor are changed  Multi-factor design.

Why multi-factor design?

Each factor can interact on the effect of other factor in the

response.

15
 Steps (Guidelines) of DOE

1. Recognition and statement of the

problem:
- What is to be studied?
- The objectives of the experiment?
- The questions one hopes to answer and the
results anticipated?
All these questions need to be clearly answered
before conducting the experiment.
It is also important to keep the overall objectives
of the experiment in mind. 16
 Steps (Guidelines) of DOE

2. Choice of factors, levels, and ranges

The experimenter selects:
- the independent variables/factors to be investigated in the
experiment,
- the number of levels for each factor,
- the levels of each factor chosen
Factors: can be potential design factors (want to vary in the
experiment) or nuisance factors.
- Design factors: selected for study in the experiment.
- Held-constant factors:
- Allowed-to-vary factors: 17

Levels = values of factors.

 Steps (Guidelines) of DOE

3. Selection of the response variable(s)

- Select a response/dependent variable (variables) that
will provide information about the problem under study.
- Propose measurement method for this response
variable, including an understanding of the
measurement system variability.
Remember:
- Chose the variable really provides useful information
about the process under study.
- Recognize important factors that result in desirable
values of the response.
18
- Potential error of measurement.
 Steps (Guidelines) of DOE

4. Choice of experimental design

Choose an appropriate experimental design (relatively
simple design and analysis methods are almost always
best) that will allow your experimental questions to be
answered once the data is collected and analyzed.
- Consideration of sample size (number of replicates),
- Selection of a suitable run order for the experimental
trials,
- Determination of whether or not blocking or other
randomization restrictions are involved.
 Plan of experiment.
19
Remember: keep the experimental objectives in mind.
 Steps (Guidelines) of DOE
5. Conducting the experiment
Perform the experiment as the plan. Remember:
- Monitor the process carefully to ensure that everything
is being done according to the plan:
+ set the values of factors correctly  check before run.
- Collect the data correctly.
- Sometime the experimenter should conduct a few trial
runs or pilot runs before really conduct the whole plan:
+ to see the consistency of experimental material,
+ to check the measurement system,
+ to get a rough idea of experimental error,
20
+ to practice the overall experimental technique.
 Steps (Guidelines) of DOE

6. Statistical analysis of the data

- Statistical methods should be used to analyze the data.
- There are many software packages and tools designed
to assist in data analysis.
- Simple graphical methods play an important role in
data analysis and interpretation.
Remember:
- Statistical methods provide guidelines as to the
reliability and validity of results, but can not prove that
a factor has a particular effect.
- Statistical techniques coupled with good engineering or
process knowledge and common sense will usually 21
lead to good conclusions.
 Steps (Guidelines) of DOE

7. Conclusions, recommendations
Based on the results of the analysis:
- draw conclusions/inferences about the results,
- interpret the physical meaning of these results,
determine the practical significance of the
findings,
- make recommendations for a course of action
including further experiments

22
 History of DOE

1. The agricultural origins, 1918 – 1940s

– R. A. Fisher & his co-workers  3 basic principles
of DOE, statistical thinking and principles, factorial
design concept and the analysis of variance
(ANOVA)
2 books  Profound impact on the use of statistics on
agricultural science.
2. The first industrial era, 1951 – late 1970s
– Box & Wilson (1951)  response surface method
– Applications in the chemical & process industries
– Applications in the plant or manufacturing process
level was still not extremely widespread 23
 History of DOE

3. The second industrial era, late 1970s – 1990

Works on optimal design of experiments began.
- Kiefer and Wolfowitz (1959)  a formal approach to
selecting a design based on specific objective optimality
criteria  model parameters could be estimated with the
best possible precision.
- Taguchi (1987, 1991) robust parameter design,
fractionated factorial designs
DOE: more widely used in the discrete parts industries:
automotive, aerospace manufacturing, electronics,
semiconductors, and many other industries.
 fourth era of statistical design began 24
 History of DOE

4. The modern era, beginning 1990

- Economic competitiveness and globalization is driving
all sectors of the economy to be more competitive.
- Computer software for construction and evaluation of
designs has improved greatly with many new features
and capability.
- Formal education in statistical experimental design is
becoming part of many engineering programs in
universities.

The successful integration of good experimental design

practice into engineering and science is a key factor in
25
future industrial competitiveness.
 What is Statistics?
7.5 9 8.1 5.1 5 8 8.5 9.1 7 8.3 7.7
7.5 7.6 8.1 5.1 5 8 8.5 6 7 8.3 7.7
7.5 7.6 7 5.1 8.5 8 8.5 6 7 8.5 7.7
7.5 7.6 7 5.1 8.4 8 7.7 6 7 8.5 7.7
7.5 7.6 7 8.4 8.4 8.1 7.7 4.4 7 8.5 7.7
7.5 7.6 7 8.4 8.4 8.1 7.7 4.4 6 8.5 7.2
8.6 8.8 7 8.4 8.4 8.1 7.7 4.4 6 8.5 7.2
8.6 8.8 7 8.4 6.7 8.1 7.7 4.4 6 9 7.2
8.6 8.8 7.3 9.3 6.7 8.1 9.5 4.4 6 9 7.2
8.6 8.8 7.3 9.3 6.7 5.8 9.5 6.7 7.8 9 7.2
8.6 7.5 7.3 9.3 6.7 5.8 9.5 6.7 7.8 9
8.6 7.5 7.3 9.3 6.7 5.8 9.5 6.7 7.8 9
6.1 7.5 7.3 9.3 9.2 5.8 9.5 6.7 7.8 7.5
6.1 7.5 7.8 5.4 9.2 5.8 7.6 6.7 9.7 7.5
6.1 7.5 7.8 5.4 5.2 7.8 7.6 8.6 9.7 7.5
6.1 7.5 7.8 5.4 8.7 7.8 7.6 8.6 9.7 7.5
6.1 8.1 7.8 5 8.7 7.8 7.6 8.6 9.7 8.1
9 8.1 7.8 5 8.7 7.8 7.6 8.6 9.7 8.1
9 8.1 7.8 5 8.7 8.5 9.1 8.6 8.3 8.1
9 8.1 5.1 5 8 8.5 9.1 7 8.3 8.1
26
 What is Statistics?

Grade Frequency
1 0
2 0
3 0
4 0
5 11
6 21
7 27
8 65
9 66
10 20

27
 What is Statistics?

• “Statistics is a way to get information from data”

Statistics

Data Information

Data: Facts, especially Information:

numerical facts, Knowledge
collected together for communicated
reference or concerning some
information. particular fact.

Statistics is a tool for creating new understanding from a set of numbers.

1.28Dictionary
Definitions: Oxford English
 Exercise

Download and install software:

• MiniTab: http://www.minitab.com/en-
us/products/minitab/free-trial/
• Read: Introduction to Using Minitab
https://onlinecourses.science.psu.edu/statprogram/minita
b

29