DESAIN EKSPERIMEN
INTRODUCTION
Semester Genap 2017/2018 Jurusan Teknik Industri
OUTLINE
•
Introduction to DOX
•
Four Eras in the History of DOX
•
Engineering Experiments
•
Typical Applications of Experimental Design
•
Strategy of Experimentation
•
Factorial Design
•
Factorial Designs with Several Factors and Fractional
Factorial
•
The Basic Principles of DOX
•
Planning, Conducting & Analyzing an Experiment
Introduction to DOX (1/2)
•
An experiment is a test or a series of tests
•
Experiments are used widely in the engineering world
– Process characterization & optimization
– Evaluation of material properties
– Product design & development
– Component & system tolerance determination
•
“
All experiments are designed experiments, some are
poorly designed, some are well-designed
”
Introduction to DOX (2/2)
•
Experimentation is a vital part of scientific (or
engineering) method
•
For any experiment, questions to be asked:
• Are only these methods available?
• Are there any other factors that might affect the results?
• How many samples are needed for the experiment?
• How should the samples be assigned to each experiment?
• What is the order that the data should be collected?
• What method of data analysis should be used?
• What difference in average observed results between method,
material, machines,…?
03/04/17 Desain Eksperimen - Semester Genap
Four Eras in the History of DOX
• The agricultural origins, 1918 – 1940s
– W.S. Gossett and the t-test (1908)
– R. A. Fisher & his co-workers
– Profound impact on agricultural science
– Factorial designs, ANOVA
• The first industrial era, 1951 – late 1970s
– Box & Wilson, response surfaces
– Applications in the chemical & process industries
• The second industrial era, late 1970s – 1990
– Quality improvement initiatives in many companies
– Taguchi and robust parameter design, process
robustness
• The modern era, beginning circa 1990
5
03/04/17 Desain Eksperimen - Semester Genap
2017/2018
R. A. Fisher (1890 – 1962)
Engineering Experiments (1/2)
• In general, experiments are used to
study the process and systems.
• The process can be the combination of
operations, machines, methods, people, and other resources (often materials) that transfer some input into output that has one or more observable response
variables y.
• Some of the process variables and
material properties, x1, …, xp are controllable.
• Some of them are uncontrollable
(although they may be controllable for purposes of a test).
6/33
Engineering Experiments (2/2)
• Determining which variables
are most influential on the response y
• Determining where to set the
influential x’s so that y is almost always near the desired nominal value.
• Determining where to set the
influential x’s so that variability in y is small.
• Determining where to set the
influential x’s so that the effects of the uncontrollable variables z1, …., zq are
minimized.
7
Typical Applications
of Experimental Design
• Improve process yield
• Reduce variability and closer conformance to nominal or target
requirements
• Reduce development time
• Reduce overall costs
• Evaluate and compare basic design configurations
• Evaluate material alternatives
• Select design parameters
• Determine key product design parameters
• Formulate new product
03/04/17 Desain Eksperimen - Semester Genap
Strategy of Experimentation (1/4)
•
Golf example-factor to influence
the score
• Driver – oversized or regular
• Ball – balata or three piece
• Mode of travel — walking or riding a
golf cart
• Beverage – water or beer
• …….
Strategy of Experimentation (2/4)
•
“
Best-guess
”
experiments
– Used a lot
– More successful than you might suspect, but there are
disadvantages…
•
One-factor-at-a-time (OFAT) experiments
– Sometimes associated with the “scientific” or “engineering”
method
– Devastated by interaction, also very inefficient
•
Statistically designed
experiments
– Based on Fisher’s factorial concept
Desain Eksperimen - Semester Genap
2017/2018 10
Strategy of Experimentation (3/4)
11/33
Strategy of Experimentation (4/4)
12/33
Factorial Design (1/3)
• In a factorial experiment, all
possible combinations of factor levels are tested
• The golf experiment:
– Type of driver – Type of ball – Walking vs. riding – Type of beverage – Time of round – Weather
– Type of golf spike
– Etc, etc, etc
13/33
Factorial Design (2/3)
14/33
Factorial Design (3/3)
15/33
Factorial Designs with Several Factors
16/33
Factorial Designs with Several Factors A
Fractional Factorial
The Basic Principles of DOX
03/04/17 Desain Eksperimen - Semester Genap 2017/2018 18
Randomization
Running the trials in an experiment in random order
Notion of balancing out effects of “lurking”
variables
Replication
Sample size (improving precision of effect estimation, estimation of error or background noise)
Replication versus repeat measurements? Replication reflects sources of variability both between runs and within runs
Blocking
Planning, Conducting & Analyzing an
Experiment
1. Recognition of & statement of problem
2. Choice of factors, levels, and ranges
3. Selection of the response variable(s)
4. Choice of design
5. Conducting the experiment
6. Statistical analysis
7. Drawing conclusions, recommendations
8. Get statistical thinking involved early
9. Your non-statistical knowledge is crucial to success
10. Pre-experimental planning (steps 1-3) vital
11. Think and experiment sequentially (use the KISS principle)
03/04/17 Desain Eksperimen - Semester Genap
2017/2018 19
Kekeliruan Eksperimen
Menyatakan kegagalan dari dua unit eksperimen identik yang dikenai perlakuan untuk memberikan hasil yang sama.
Contoh: kekeliruan waktu menjalankan eksperimen, kekeliruan pengamatan, variasi bahan eksperimen, variasi antara unit
eksperimen dan pengaruh gabungan semua faktor tambahan
Untuk mengurangi kekeliruan dapat menggunakan bahan
eksperimen yang homogen, menggunakan informasi yang sebaik-baiknya tentang variabel yang telah ditentukan dengan tepat,
melakukan eksperimen dengan teliti, menggunakan DE yang lebih efisien
03/04/17 Desain Eksperimen - Semester Genap
2017/2018 20
Efek dan interaksi
•
Dalam keperluan desain eksperimen, variabel bebas
dinamakan faktor, dan nilai-nilai atau
klasifikasi-klasifikasi dari sebuah faktor dinamakan taraf faktor.
•
Faktor-faktor dinyatakan dengan huruf kecil a, b, c, d,
dst.., sedangkan taraf faktor dinyatakan dengan angka 1,
2, 3, dst yang dituliskan sebagai indeks untuk faktor
yang bersangkutan.
03/04/17 Desain Eksperimen - Semester Genap
2017/2018 21
Kontrol Lokal
Pengelompokan
• Penempatan sekumpulan unit eksperimen ke dalam kelompok-kelompok agar kelompok-kelompok yang berbeda memungkinkan untuk mendapatkan perlakuan yang berbeda pula.
• Contohnya ketika meneliti pengaruh tiga macam cara penyampaian materi terhadap mahasiswa. Terdapat 11 mahasiswa, yang dibagi menjadi 3 kelompok, (4-4-3). Penempatan atau pemilihan
mahasiswa ke dalam kelompok-kelompok dilakukan secara acak. Masing-masing kelompok diberikan materi dengan cara berbeda (A-B-C)
Pemblokan
• Pengalokasian unit-unit eksperimen ke dalam blok sehingga unit-unit dalam blok secara relatif bersifat homogen sedangkan sebagian besar dari variasi yang dapat
diperkirakan di antara unit-unit telah terbaur dengan blok.
• Variasi yang dapat diduga tidak menjadi bagian daripada kekeliruan eksperimen, sehingga desain lebih efisien.
• Contohnya:: Ada 12 komponen mesin. Enam jenis X, tiga jenis Y, dan tiga jenis Z. Jenis diambil sebagai blok. Blok pertama terdiri dari 6 buah komponen jenis X. Blok kedua terdiri dari 3 buah komponen jenis Y. Blok ketiga terdiri dari 3 buah komponen jenis Z.
Penyeimbangan
• Usaha memperoleh unit
eksperimen, usaha pengelompokan, pemblokan dan penggunaan
perlakuan sedemikian sehingga dihasilkan konfigurasi formasi yang seimbang
03/04/17 Desain Eksperimen - Semester Genap 2017/2018 22