Managing Information Technology

224 Part II • Information Technology

External

and Internal

Data:

Data

Management

Model

Management

Dialog Management

(User Interface)

Manager

(User)

Other

Computer-Based

Systems

Decision Support System

FIGURE 6.1 Decision Support Systems Components

interface to the DSS. The user interacts with the DSS
through the dialog management component, identifying
the particular model and data set to be used, and then the
DSS presents the results to the user through this same
dialog management component. The model management
and data management components largely act behind the
scenes, and they vary from relatively simple for a typical
spreadsheet model to quite complex for a mathematical
programming-based scheduling model.
An extremely popular type of DSS is a pro forma
financial statement generator. Using a spreadsheet
package such as Microsoft Excel, a manager builds a
model to project the various elements of the organization
or division financial statement into the future. The data
employed are historical financial figures for the organiza-
tion. The initial (base) model incorporates various
assumptions about future trends in income and expense
categories. After viewing the results of the base model, the
manager performs a series of “what-if” analyses by
modifying one or more assumptions to determine their
impact on the bottom line. For example, the manager
might explore the impact on profitability if the sales of a
new product grew by 10 percent per year, rather than the
5 percent incorporated in the base model. Or the manager
might investigate the impact of a higher-than-expected
increase in the price of raw materials, such as 7 percent
per year instead of 4 percent per year. This type of
financial statement generator is a simple but powerful
DSS for guiding financial decision making.

An example of a DSS driven by transactions data is a

police-beat allocation system used by a California city.

This system enables a police officer to display a map

outline and call up data by geographic zone, which shows

police calls for service, types of service, and service times.

The system’s interactive graphics capability lets the officer

manipulate the maps, zones, and data to consider a variety

of police-beat alternatives quickly and easily and takes

maximum advantage of the officer’s judgment.

Other DSS examples include an interactive system

for capacity planning and production scheduling in a large

paper company. This system employs detailed historical

data and forecasting and scheduling models to simulate

overall performance of the company under differing

planning assumptions. A major oil company developed a

DSS to support capital investment decision making. This

system incorporates various financial routines and models

for generating future plans; these plans can be displayed in

either tabular or graphic form to aid in decision making.

A major airline uses a DSS to help aircraft controllers deal

with aircraft shortage problems that might arise at an

airport because of delayed or canceled incoming flights or

mechanical problems for aircraft on the ground. The DSS,

which uses a network optimization modeling technique,

helps controllers use spare aircraft more effectively as well

as evaluate possible delay-and-swap options. Over an

18-month period, this DSS saved the airline more than

$500,000 in delay costs.

All the DSS examples cited are more properly called

specific DSSs. These are the actual applications that assist

in the decision-making process. In contrast, a DSS

generatoris a software package that provides a set of

capabilities to build a specific DSS quickly and easily

(Sprague and Carlson, 1982). In the previous pro forma

financial statement example, Microsoft Excel can be

viewed as a DSS generator, whereas a specific Excel

model to project financial statements for a particular

division of a company is a specific DSS.

Data Mining

In Chapter 5 we introduced data warehousing—the idea of a

company pulling data from its operational systems and

putting the data in a separate data warehouse so that users

may access and analyze the data without interfering with the

operational systems. In that discussion we touched on the

variety of software tools available for analysis of data in

the warehouse but deferred a more complete discussion until

this chapter. Our argument was that the creation and mainte-

nance of the data warehouse is an enterprise system, in that

the data warehouse supports the entire organization by