Transformational Trends in the Energy Storage Market
Thermal and Battery Energy Storage Systems
Poised for Remarkable Growth
9AAE-27
August 2015
Contents
Section
Slide Number
Executive Summary
3
Transformational Trends in Energy Storage
9
Vision for 2025

236
Exercises
(a) State a reasonable model for this experiment, including any assumptions on the
error term.
(b) How would you check the assumptions on your model?
(c) Calculate an analysis of variance table and test any relevant hypotheses, stating
your

235
Exercises
Table 7.20
Data for paper towel strength experiment: A amount of
liquid, B brand of towel, and C liquid type
ABC
111
112
121
122
211
212
221
222
Strength
3279.0
3260.8
2889.6
2323.0
2964.5
3114.2
2883.4
2142.3
(Order)
(3)
(11)
(5)
(1)
(4)
(1

232
Exercises
2. In planning a ve-factor experiment, it is determined that the factors A, B, and C
might interact and the factors D and E might interact but that no other interaction
effects should be present. Draw a line graph for this experiment and giv

7.7
Table 7.14
Using SAS Software
229
SAS program for the rail weld experiment with two empty cells
DATA;
INPUT T V S Y1 Y2;
REP=1; Y=Y1; OUTPUT; * create SAS observation for y=y1;
REP=2; Y=Y2; OUTPUT; * create SAS observation for y=y2;
LINES;
1 1 1 84.0

230
Chapter 7
Table 7.15
Several Crossed Treatment Factors
Output from the rst call of PROC GLM for the rail weld experiment
The SAS System
General Linear Models Procedure
Dependent Variable: Y
Sum of
Source
DF
Squares
Model
5
349.510000
Error
6
417.79000

7.7
Using SAS Software
227
set TORQUE2 containing the mean and variances of the responses at each combination of the
A, B, and M levels is created via the PROC MEANS statement and printed as in Table 7.8,
page 219. An interaction plot for ABM, such as tha

226
Chapter 7
Table 7.12
Several Crossed Treatment Factors
SAS program for the VossWang method for the drill advance experiment
* Data set DRILL contains the original data;
DATA DRILL4; SET DRILL;
* Fit complete model including all main effects and intera

228
Chapter 7
Table 7.13
Several Crossed Treatment Factors
SAS program for analysis of a mixed array and a product array
DATA TORQUE;
INPUT A B M Y;
TC = (3*(A-1) + B;
LINES;
: : : (input data lines here)
;
* calculate the average data values for each ABM

224
Chapter 7
Several Crossed Treatment Factors
we show how to calculate and plot the sample means and variances for robust design, and in
Section 7.7.4, we show the complications that can arise when one or more cells are empty.
7.7.1
Normal Probability P

222
Chapter 7
Several Crossed Treatment Factors
w s
are mutually independent ,
1, . . . , v ;
w
and we can analyze the effects of the design factors averaged over the noise factors via the
usual analysis of variance, but with y w. , w 1, . . . , v, as the

Exercises
239
(a) Make a table similar to that of Table 7.1, page 201, with the rst column containing
the 27 treatment combinations for the washing power experiment in ascending
order. List the contrast coefcients for the main effect trend contrasts: Line

238
Exercises
Table 7.23
Scores for the galling experiment
A
B
C
yijk
1
1
1
2
1
1
2
5
1
2
1
0
1
2
2
2
2
1
1
6
2
1
2
10
2
2
1
4
2
2
2
8
Source: Ertas, A., Carper, H. J., and Blackstone, W. R.
(1992). Published by the Society for Experimental
Mechanics. Rep

Service Business Models and Changing Competitive
Landscape for Energy Management
Industry Convergence and New Technology Drive Business Model
Innovation
9AAF-19
February 2015
Contents
Section
Slide Numbers
Executive Summary
3
Service Business Models for E

Global Demand Response Trends
Expansion of Distributed Generation Intensified by the Need for
Demand Management
MBBC-14
December 2015
Contents
Section
Slide Numbers
Executive Summary
3
Demand Response Analysis
10
Drivers and Restraints
17
Business Models

If a Z, prove that a 2 is not congruent to 2 modulo 4 or to 3 modulo 4.
By the Division Algorithm any a Z must have one of the following forms
a =4k
4k + 1
4k + 2
4k + 3
This implies a2 = 16k2 = 4(4k2 ) = 4q
16k 2 + 8k + 1 = 4(4k 2 + 2k) + 1 = 4r + 1
4(4k

252
Chapter 8
Polynomial Regression
2
where x .
x rx x/n and ssxx
x rx (x x . ) . The corresponding estimators (random
0 and 1 , are normally distributed, since they are linear
variables), which we also denote by
combinations of the normally distributed

8.5
Table 8.2
Test for lack of t of quadratic regression model for hypothetical
data
Source of
Variation
Lack of Fit
Pure Error
Error
Example 8.4.1
251
Analysis of the Simple Linear Regression Model
Degrees of
Freedom
2
10
12
Sum of
Squares
30.0542
132.74

250
Chapter 8
Polynomial Regression
squares from the one-way analysis of variance model. Then the sum of squares for lack of
t is
ssE ssPE .
ssLOF
The sum of squares for pure error has n v degrees of freedom associated with it, whereas
the sum of squares

8.4
8.4
249
Test for Lack of Fit
Test for Lack of Fit
We illustrate the lack-of-t test via the quadratic regression model
0 + 1 x + 2 x 2 .
E[Yxt ]
If data have been collected for only three levels x x1 , x2 , x3 of the treatment factor, then
the tted mod

248
8.3
Chapter 8
Polynomial Regression
Least Squares Estimation (Optional)
In this section, we derive the normal equations for a general polynomial regression model.
These equations can be solved to obtain the set of least squares estimates j of the para

246
Chapter 8
Polynomial Regression
Although regression models can be used to estimate the mean response at values of x
that have not been observed, estimation outside the range of observed x values must be done
with caution. There is no guarantee that th

244
Chapter 8
Polynomial Regression
as a function of x, and it can be used to estimate the mean response or to predict the values
of new observations for any factor level x, including values for which no data have been
collected. We call yx the tted model

8.2
247
Models
In the following optional section, we obtain the least squares estimates of the parameters
0 and 1 in a simple linear regression model. However, in general we leave the determination
of least squares estimates to a computer, since the formu

Chapter 2: Descriptive Statistics
Descriptive statistics are meant to provide quick and efficient summaries of data.
Descriptive statistics are usually differentiated from inferential statistics, which will
be the subject of subsequent chapters. As we wil

ACMS 10145Statistics for Business I, Spring 2015
Course Syllabus
Course:
ACMS 10145, Spring Semester, 2015
When & Where: (Section 03) MWF 9:25AM-10:15AM, DeBartolo Hall 129
Professor:
Dr. Alan Huebner, PhD
E-mail address: Alan.Huebner.10@nd.edu
Office Hou

ACMS 10145 Spring 2015 Help Session Schedule
In addition to my office hours, you can also get help on homework and concepts
from the following individuals at the given locations and times:
Name
Emily
Emily
Day and Time
Sundays 6:00-8:00 pm
Mondays 7:00-9:

How to Access Your CengageNOW Course
ACMS 10145 Spring 2015
Instructor: Alan Huebner
Start Date: 1/12/15
Course Key: E-TWQN4B4HEQ9FW
PURCHASE OPTIONS
*Grace Period or Trial Option should not be used for this course.
Bookstore: Purchase access to CengageNO

Chapter 1: Data and Statistics
1.2 Data
Data are the facts and figures collected, analyzed, and summarized for presentation
and interpretation. Elements are the entities on which data are collected. A variable
is a characteristic of interest for the eleme