Microsoft Office Tutorials and References
In Depth Information
entry fee will be required, etc. A good place to begin the modeling phase is to create
of an Influence Diagram (IFD). IFDs are diagrams that are connected by directed
arrows, much like those in the preliminary process flow planning diagram of Exhibit
7.4. An IFD is a powerful tool that is used by decision analysts to specify influences
in decision models. Though the concept of an IFD is relatively simple, the theoreti-
cal underpinnings can be complicated. For our example, we will develop two types
of IFDs: one very simple and one a bit more complex.
We begin by identifying factors—processes, decisions, outcomes of decisions,
etc. —that constitute the problem. In our first IFD, we will consider the links
between these factors and determine the type of influence between them, either pos-
itive (+) or negative (–). A positive influence (+) suggests that if there is an increase
in a factor, the factor that it influences also has an increase; it is also true that as a
factor decreases so does the factor it influences. Thus, they move in the same direc-
tion. For example, if I increase marketing efforts for a product, we can expect that
sales will also increase. This suggests that we have a positive influence between
marketing efforts and sales. The opposite is true for negative influence (–): factors
move in opposite directions. A negative influence can easily exist between the qual-
ity of employee training and employee errors—the higher the quality of training for
employees the lower the number of errors committed by employees. Note that the
IFD does not suggest the intensity of the influence, only the direction.
Not all models lend themselves to this simple form of IFD, but there will be
many cases where this approach is quite useful. Now, let’s apply the IFD to Fr.
Efia’s problem. Voitech has helped Fr. Efia to create a simple IFD of revenue gen-
eration for Vegas Night at OLPS . It is shown in Exhibit 7.5. Voitech does so by
conducting another interview and having Fr. Efia consider more carefully the struc-
ture of the event, and how elements of the event are related. In order to understand
the movement of one factor due to another, we first must establish a scale for each
factor, from negative to positive. The negative to positive scale used for the levels of
weather quality and attendee good fortune is bad to good .For attendance and rev-
enue , the scale is quite direct: higher levels of attendance or revenue are positive and
lower levels are negative. The IFD in Exhibit 7.5 provides an important view of how
revenues are generated, which of course is the goal of the event. Fr. Efia has speci-
fied six important factors: Weather Quality, Attendance, Attendee Luck or Fortune
in Gambling, Entry Admission Revenue, Gambling Proceeds Revenue, and Total
Revenue. Some factors are uncertain and others are not. For example, weather and
attendee fortune are uncertain, and obviously he hopes that the weather quality will
be good (+) and that attendee good fortune will be bad (
). The effect of these two
conditions will eventually lead to greater revenues (+). Entry Admission Revenues
are known with certainty once we know the attendance, as is the Total Revenue once
we determine Entry Admission Revenue and Gambling Proceeds Revenue.
Note that the model is still quite general, but it does provide a clear understanding
of the factors that will lead to either success or failure for the OLPS event. There
is no final commitment, yet, to a number of the important questions in Table 7.1,
for example, questions 2, the odds of the games , and 3, will all attendees play all
games . But, it has been decided that the three games mentioned in question 2 will
Search JabSto ::




Custom Search