High Performance Data Warehousing
Decision Support Systems And Expert Systems
Expert Systems And Data Warehouses
Expert system is a term used very loosely in the computer community
regarding anything from a spreadsheet program to any program
containing an if statement. In general terms, an expert system
models the well-structured decision process of the human mind and
applies that reasoning process to a real-world situation. Any
decision-making process with quantifiable rules can have the rules
stored in an inference engine. An inference engine is used to drive
the information-gathering component of a system, which eventually
arrives at the solution to the problem.
It has been said that an expert system makes a decision for the
user, while a decision support system makes a decision with the
user. This distinction is essentially true, because an expert system
makes no provision for human intuition in the decision-making
process. Many real-world management decisions do not require human
intuition, and the data warehouse can provide answers without end
user input. For example, one of the crucial jobs of a retail manager
is the choice of what goods to order, the quantity, and the ordering
time frame. These decisions can be represented by a model called
economic order quantity (EOQ). If the EOQ equation knows the speed
at which the goods are leaving the retail store, the delivery time
on reorders, the average time goods remain on the shelf, and the
cost of the goods, the computer can confidently produce automatic
daily reports specifying which goods to order and the appropriate
quantity. Also, a DSS pre-summarizes data so the manager can quickly
take a high-level look at the relevant figures (see Figure 1.14).
Figure 1.14 Defining different levels of aggregation.he analysis phase. For more information on this
topic, see Chapter 3, Data Warehouse Analysis.
This is an excerpt from "High Performance
Data Warehousing".
