Microsoft Office Tutorials and References
In Depth Information
some desired
objective
under
constrained
conditions. We will learn more about this
prescriptive analysis
procedure later in this chapter.
Also of great interest will be the Scenarios and Goal Seek tools. The
Scenario
tool is used to implement what we have referred to as
whatif
analysis. Simply put,
with Scenarios we have an efﬁcient tool for automating the variation of inputs for a
problem formulation, and subsequently recording the resulting output. Its function
is to organize and structure. Without the formal structure of Scenarios it is very easy
to lose important information in your analysis.
Goal Seek is also a member of the Whatif subgroup in the Data Tools group.
In situations where we know the outcome we desire from a problem,
Goal Seek
,as
the name implies, is an approach that
seeks
to ﬁnd the input that leads to our
desired
outcome. For example, if we are calculating the constant and periodic payments
for a mortgage, we may ask the question—what interest rate will lead to a monthly
payment of $1,000 for a loan with a term of 240 monthly periods and principal value
of $100,000?
Before we become acquainted with these new tools let us take stock of where we
have been thus far on our analytical journey. We began with a careful classiﬁcation
of data, from categorical to ratio, and we discussed the implications of the data type
on the forms of analysis that could be applied. Our classiﬁcation focused on
quan
titative
and
qualitative
(
non

quantitative
) techniques for presenting, summarizing,
and analyzing data. Through Chap. 6 we assumed that the data under consideration
were available from the collection of primary data
1
or available from a secondary
source. Additionally, our analysis was
descriptive
in nature, generally attempting to
describe some characteristic of a population, by analyzing a sample from that popu
lation; for example, comparing the mean of a sample to the mean of the population,
and then determining if the sample mean could have come from the population, at
some level of statistical signiﬁcance.
In Chaps. 7 and 8, Modeling and Simulation, we created models in which we gen
erated data that we could then analyze. In these chapters, we began our modeling
effort by using the descriptive procedures discussed above to deﬁne our model, and
then we generated data from the model to use for prescriptive purposes. For exam
ple, in Chap. 8 we used our understanding of the operating behavior of Autohaus
to create a Monte Carlo simulation. The data generated by the simulation allowed
us to prescribe to Inez the possible design selections for the Autohaus business
concept.
The models that we create with Solver have an even stronger prescriptive char
acter than those encountered thus far. In using Solver, our goal is to determine the
values of decision variables that will
minimize
or
maximize
some objective, while
adhering to the technological constraints of the problem. Thus, the solution will
prescribe
very speciﬁc action about decision variables. As you can see from this
1
Primary data is collected by someone in the role of collecting data for a speciﬁc purpose and
comes from sources that are generally not available as a result of other studies. For example, a
survey study performed by a company to determine their customers’ satisfaction with their service
is primary data, while data on similar industrywide service that can be purchased from a consulting
ﬁrm is secondary data.
Search JabSto ::
Custom Search