Design of
Experiment
Stat590
Lecture 1
Design of Experiment
Stat590
Lecture 1
Textbook: A First Course in Design and Analysis of
Experiments by Gary Oehlert, Freeman New York, 2000
Source: Chapter 1 a
Design of
Experiment
Stat590
Lecture 8
Design of Experiment
Stat590
Lecture 8
Textbook: A First Course in Design and Analysis of
Experiments by Gary Oehlert, Freeman New York, 2000
Source: Chapter 10
Design of
Experiment
Stat590
Lecture 5
Design of Experiment
Stat590
Lecture 5
Textbook: A First Course in Design and Analysis of
Experiments by Gary Oehlert, Freeman New York, 2000
Source: Chapter 5
M
Design of
Experiment
Stat590
Lecture 3
Design of Experiment
Stat590
Lecture 3
Textbook: A First Course in Design and Analysis of
Experiments by Gary Oehlert, Freeman New York, 2000
Source: Chapter 6
A
Design of
Experiment
Stat590
Lecture 6
Design of Experiment
Stat590
Lecture 6
Textbook: A First Course in Design and Analysis of
Experiments by Gary Oehlert, Freeman New York, 2000
Source: Chapter 7 a
Design of
Experiment
Stat590
Lecture 4
Design of Experiment
Stat590
Lecture 4
Textbook: A First Course in Design and Analysis of
Experiments by Gary Oehlert, Freeman New York, 2000
Source: Chapter 4 a
Design of
Experiment
Stat590
Lecture 7
Design of Experiment
Stat590
Lecture 7
Textbook: A First Course in Design and Analysis of
Experiments by Gary Oehlert, Freeman New York, 2000
Source: Chapter 8 a
DESIGN OF EXPERIMENTS
16:960:590:01
FALL 2014
ARC 105, BC
M 6:40 9:30 PM
Text (free online):
A First Course in Design and
Analysis of Experiments
By Gary W. Oehlert,
W.H. Freeman & Co., 2000
Instructo
Demand Setup Cost Price per unit Min Order Holding Cost EOQ
Feasible?
Source A
20000
100
2.5
0
0.5 2828.427 Yes
Source B
20000
100
2.4
3000
0.48 2886.751 No
Source C
20000
100
2.3
4000
0.46 2948.839 N
DDL
K
l
h
p
x
m
s
75
1976
7.52
400
normal
114
19.74842
$/order
units/year
$/unit/year
$/unit
(38;130)
units/lead time
units/lead time
a
Q
500
1-F(R)
0.004757085
b
EOQ
198.5318453473
Q
198.531845347
20
Homework 4 Pedro Cesar Lopes Gerum
12
=1250*0.18=225
c=$18.5
h=0.25*18.5=4.625
K=$28
2
a) =
= 52 =
= .2311 years = 12.02 weeks
b) Demand is assumed constant, so 6 weeks before running empty, we will
STAT 5200 Handout #17: Nesting and Crossing Examples (Ch. 12)
Example I is a simple nested factorial design. Example II is a bit more complicated.
Example I: Factory Layout
An industrial engineer cond
Logistic Regression
Instructor: Ping Li
Department of Statistics and Biostatitics
Department of Computer Science
Rutgers University
2015
1
Calculus Review: Derivatives
Simple derivatives:
1
[log x] =
Homework 4 Pedro Cesar Lopes Gerum
12
=1250*0.18=225
c=$18.5
h=0.25*18.5=4.625
K=$28
a)
Q=
2K
Q
=52T = =.2311 years=12. 02 weeks
h
b) Demand is assumed constant, so 6 weeks before running empty, we wi
Section 3.4
21)
Section 3.5
24) a)
b)
c) For i = 1,2,3,4
Obj: min
10000.1I i +100H i +200F i
i
s.t.
I i =I i1 + PiDi
I 0 =0
W 0=100
I i , H i W i , Fi 0
Pi=W i0.307971daysi
W
( i1+ H i F i)
W i=
d)
2
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
1 Step Ahead
Forecast
Demand
89
57
144
221
177
280
223
286
212
275
188
312
205.5
225.25
241.5
250.25
249
240.25
MAD
Alpha
Beta
G0
SO
Year 1
Year 2
HW 2 Prod Analysis
Pedro Cesar Lopes Gerum
2.17-2.19
So, according to MAD, the 1 step ahead approach is better.
2.22
The Forecast using alpha equals to 0.4 gave more weight to the most recent demand t
Homework 5
Pedro Cesar Lopes Gerum
Ex. 4.22)
EOQ=
( 2DemandSetup
)
h'
So it is advantageous to order 4000 products from Source C every .2 year.
b) There will be 5 orders placed per year, resulting on
Homework 6 Pedro Cesar Lopes Gerum
10. a)
This problem is similar to a news vendor model, where the overage cost is the emergency shipping and
the underage cost is the holding cost.
Holding Cost =Cu=
Homework 3 Prod Analysis
Pedro Cesar Lopes Gerum
24) I decided to put the constraint of Hires and Fires being integers because the numbers were low
enough it would make a difference.
25)
33) a) 185 em
Activity
A
B
C
D
E
F
G
H
a
2
1
3
2
4
6
1
4
m
5
6
5
4
6
7
2
6
b
9
8
12
12
8
8
6
16
Mean
5.17
5.50
5.83
5.00
6.00
7.00
2.50
7.33
Variance
1.36
1.36
2.25
2.78
0.44
0.11
0.69
4.00
ES
0.00
5.17
5.17
10.67
The solutions for set #5 (Chapters 9 and 10) are:
-Question 9.3 solution:
_
Yij.is an estimator of the (i,j) cell whose parameters are:
mu + alpha i + beta j + alphabetaij
_
Then a contrast such as Su
CNC
Numerical control (NC) is the automation of machine tools that are operated by precisely
programmed commands encoded on a storage medium, as opposed to controlled manually via
hand wheels or lever
STRUTS: Statistical Rules of Thumb
Gerald van Belle
Departments of Environmental Health and Biostatistics
University of Washington
Seattle, WA
98195-4691
Steven P. Millard
Probability, Statistics and
SET #1
Homework #1
This is a permutation test similar to the example included in the recent email (Sept.16), follows
the structure of Exercise # 2.4 from the book, but with simpler numbers:
Treatment
USING R FOR DESIGN OF EXPERIMENTS 16:960:590:01
Once you call the R program you see on the screen a prompt to receive input commands. The default prompt is
for the Windows environment.
The R commands
Design of Experiments (DOE)
Notes
Experimentation by varying one factor at a time began with Francis Bacon (XVI century),
in his experimentum crucis. Modern statistical experiments started with R. A.
HOMEWORK SET # 3
DESIGN OF EXPERIMENTS 590
Due date: October 30, 2017
1,- Solve EXERCISE 5.2 from the book
2.- Solve EXERCISE 5.5 from the book
3.- Solve EXERCISE 6.1 from the book.
Note: from Montgom
Prof. Rojas
Design of Experiments 16:960:590:01
HOMEWORK SET # 1
You can prepare the calculations and present the solutions in any regular piece of paper. You do not need
to print this page to solve t