Decision support and expert system, Slides of Management Information Systems

Download Decision support and expert system and more Slides Management Information Systems in PDF only on Docsity!

UNIT 7: DECISION

SUPPORT AND EXPERT

SYSTEMS

James 1:5-

If any of you lacks wisdom, let him ask

God, who gives generously to all

without reproach, and it will be given

him. But let him ask in faith, with no

doubting, for the one who doubts is like

a wave of the sea that is driven and

tossed by the wind.

Decision Support

• (^) Organization’s success depends on the quality of decisions made by its employees.
• (^) Computer-based systems are beneficial to make decision from large amounts of information or Intensive processing.
• (^) Decision support aids is based on
  • (^) Decision support systems (DSSs) and
  • (^) Expert systems (ESs)
• (^) Decision support modules are often part of larger enterprise applications. For example, ERP (enterprise resource planning) systems support decision making in such areas as production capacity planning, logistics, and inventory replenishment.

• (^) Decision support modules today are part of larger enterprise applications and it serve not for decision making only but as business analysis tools or business intelligence applications.
• (^) Decision support systems and expert systems are especially designed to streamline the decision- making process by providing either a single optimal solution to a question or problem, or a narrow set of solutions from which decision makers can select.
• (^) Data warehouses and online processing (OLAP) technologies have enhanced the ability to use data for decision making.

Figure 10.1 The three phases of decision making

• (^) Businesses collect data internally and externally. They use models to analyze data. A model is an abstraction of reality.
• (^) Decision makers either use universal models, such as certain statistical models, or design their own models to analyze data.
• (^) Then they select what they perceive as the best course of action.

Structured and Unstructured Problems

• (^) Structured problem is one in which an optimal solution can be reached through a single set of steps.
• (^) The steps must be followed in a known sequence, solving a structured problem with the same data always yields the same solution.
• (^) Mathematicians call a sequence of steps an algorithm and the categories of data that are considered when following those steps parameters.
• (^) Unstructured problem : one for which there is no algorithm that leads to an optimal solution - (^) May not be enough information - (^) May be a large number of potential factors

• Figure 10.2 Examples of structured and semi-structured problems in business © Cengage Learning

• (^) The terms “decision support systems” and “expert systems” are mentioned less frequently these days.
• (^) The concepts of modeling decision-making processes and automating them and the transformation of human expertise into software are alive and thriving.
• (^) While many situations exist in which only an experienced professional can make good decisions, much of the decision-making process can be automated through use of computer-based decision aids.
• (^) The raw materials for many decisions are already in corporate databases and data warehouses, and they can be accessed through ISs such as supply chain management systems.

The Data Management Module

• (^) A DSS’s Data management module is a database or data warehouse that provides data for the intelligence phase - (^) Accesses the data - (^) Provides a means to select data by specified criteria
• (^) A data management module accesses the data and provides a means for the DSS to select data according to certain criteria: type of stock, range of years, and so on.

The Model Management Module

• (^) Model management module : turns data into useful information offered from either a single fixed model , a dynamically - (^) A fixed variable model does not change. - (^) A dynamically modified model is one that is automatically adjusted based on changing relationships among variables.
• (^) A model might be a widely used method to predict performance, such as best-fit linear analysis, or it might be built by the organization, using the experience that employees in the firm have accumulated over time.

The Dialog Module

• (^) Dialog module : part of a DSS that allows user interaction with the program. - (^) Prompts the user to select a model and data to process - (^) Allows the user to change parameters and view the results of the changes - (^) Displays the results of the analysis in textual, tabular, or graphical format
• (^) Many DSSs are available through the Internet
• (^) The dialog module is also responsible for displaying the results of the analysis.
• (^) DSSs use various textual, tabular, and graphical displays from which the decision maker can choose. Take the previous advertising effort scenario, for example, where the company’s marketing manager is trying to decide how to spend promotional dollars.

Sensitivity Analysis

• (^) An outcome is almost always affected by more than one parameter; for instance, the sales volume of a product is affected by the number of salespeople, the number of regional sales representatives, the amount spent on national and local television advertising, price, competition, and so on.
• (^) However, outcomes rarely respond in equal measure to changes in parameters.
• (^) Sensitivity analysis is conducted to test the degree to which the total profit grows or shrinks: - (^) If one or more of the factors is increased or decreased
• (^) Also called what-if analysis; DSS can perform sensitivity analysis on multiple parameters simultaneously

Agriculture

• (^) When a farmer enters basic information about a particular crop and the pest problem— weeds, insects, or diseases—the Planner outlines options such as chemicals and ways to apply them.
• (^) Farmers can describe their spraying equipment, size of the field, and current chemical prices from local suppliers, Tax Planning
• (^) Based on the taxable income and the combination of deductions taken, the applications warn users about their chances of being audited by the Internal Revenue Service and give them a chance to modify deductions. Web Site Planning and Adjustment
• (^) Because so many companies use the Web for marketing, selling, and customer support, decisions on how to design Web sites are extremely important. Some companies offer DSSs specifically designed to analyze shoppers’ behavior at their sites based on captured data such as pages viewed, options clicked, and the sequence of pages shoppers view.

Expert Systems

• (^) It is not always possible to exploit expertise by coupling quantitative data from a database with decision models.
• (^) An expert system (ES) is developed to emulate the knowledge of an expert to solve problems and make decisions in a relatively narrow domain.
• (^) A domain is a specific area of knowledge.
  • (^) For example, in medicine a domain is often a diagnosis of a specific disease or a family of related diseases, such as bacterial diseases.
• (^) The purpose of ESs is to replicate the unstructured and undocumented knowledge of the few (the experts), and put it at the disposal of the many other people who need the knowledge, often novices or professionals in the same domain but with far less expertise.