Lecture-01-DSS

8/7/2019 Lecture-01-DSS

http://slidepdf.com/reader/full/lecture-01-dss 1/82

McGraw-Hill/Irwin Copyright © 2008, The McGraw-Hill Companies, Inc. All rights reserved.

Decision Support Systems

Decision Support Systems

Lecture

1



10-2

� Identify the changes taking place in the form

and use of decision support in business

� Identify the role and reporting alternatives

of management information systems

� Describe how online analytical processing

can meet key information needs of managers

� Explain the decision support system conceptand how it differs from traditional management

information systems

Learning Objectives



10-3

Learning Objectives

� Explain how the following information systems

can support the information needs of executives,

managers, and business professionals

� Executive information systems� Enterprise information portals

� Knowledge management systems

� Identify how neural networks, fuzzy logic,genetic algorithms, virtual reality, and

intelligent agents can be used in business



10-4

Learning Objectives

� Give examples of several ways expert systems

can be used in business decision-making

situations



10-5

Decision Support in Business

� Companies are investing in data-driven decision

support application frameworks to help them

respond to

� Changing market conditions� Customer needs

� This is accomplished by several types of

� Management information� Decision support

� Other information systems



10-6

Case 1: Dashboards for Executives

� Web-based ³dashboards´

� Displays critical information in graphic form

� Assembled from data pulled in real time from

corporate software and databases� Managers see changes almost instantaneously

� Now available to smaller companies

� Potential problems

� Pressure on employees

� Divisions in the office

� Tendency to hoard information



10-7

Case Study Questions

� What is the attraction of dashboards to CEOs

and other executives?

� What real business value do they provide

to executives?� The case emphasizes that managers of small

businesses and many business professionals

now rely on dashboards.

� What business benefits do dashboards provide

to this business audience?



10-8


� What are several reasons for criticism of

the use of dashboards by executives?

� Do you agree with any of this criticism?



10-9

Levels of Managerial Decision Making



10-10

Information Quality

� Information products made more valuable bytheir attributes, characteristics, or qualities

� Information that is outdated, inaccurate, or

hard to understand has much less value� Information has three dimensions

� Time

� Content

� Form



10-11

Attributes of Information Quality



10-12

Decision Structure

� Structured (operational)� The procedures to follow when decision

is needed can be specified in advance

� Unstructured (strategic)� It is not possible to specify in advance

most of the decision procedures to follow

� Semi-structured (tactical)� Decision procedures can be pre-specified,

but not enough to lead to the correct decision







10-15

Business Intelligence Applications





10-17

DSS Components



10-18

DSS Model Base

� Model Base

� A software component that consists of

models used in computational and analytical

routines that mathematically express relations

among variables

� Spreadsheet Examples

� Linear programming

� Multiple regression forecasting

� Capital budgeting present value



10-19

Simple examples of Math Modeling

� Estimate the yield of wheat in India from thestanding crop (without cutting and weighing thewhole of it)

� Estimating the population of India in the year

2025 (without waiting till then).� Estimate the average life span of a bulbmanufactured in a company (without lighting tillfused)

� Estimate the total amount of insurance claims acompany has to pay next year (without waitingtill the end of the year).

� Effect of Immigration and Emigration onpopulation size.



10-20

Technique of Math Modeling

Real

Problem

Mathematical

Problem

Mathematical

Solution

Interpretation

Mathematical Physics

Mathematical Medicine (blood flow)Mathematical Economics

Mathematical Psychology

Mathematical Sociology

Mathematical Engineering

«



10-21

Applications of Statistics and Modeling

� Supply Chain: simulate and optimize supplychain flows, reduce inventory, reduce stock-outs

� Pricing: identify the price that maximizes

yield or profit� Product and Service Quality: detect quality

problems early in order to minimize them

� Research and Development: improve quality,efficacy, and safety of products and services



10-22

Management Information Systems

� The original type of information systemthat supported managerial decision making

� Produces information products that support

many day-to-day decision-making needs� Produces reports, display, and responses

� Satisfies needs of operational and tactical

decision makers who face structured decisions



10-23

Management Reporting Alternatives

� Periodic Scheduled Reports� Prespecified format on a regular basis

� Exception Reports

� Reports about exceptional conditions� May be produced regularly or when an

exception occurs

� Demand Reports and Responses

� Information is available on demand

� Push Reporting� Information is pushed to a networked computer



10-24

Example of Push Reporting



10-25

Online Analytical Processing

� OLAP

� Enables managers and analysts to examine

and manipulate large amounts of detailed and

consolidated data from many perspectives� Done interactively, in real time, with rapid

response to queries



10-26

Online Analytical Operations

� Consolidation� Aggregation of data

� Example: data about sales offices rolled up

to the district level

� Drill-Down

� Display underlying detail data

� Example: sales figures by individual product

� Slicing and Dicing

� Viewing database from different viewpoints

� Often performed along a time axis



10-27

Geographic Information Systems

� GIS

� DSS uses geographic databases to construct

and display maps and other graphic displays

� Supports decisions affecting the geographicdistribution of people and other resources

� Often used with Global Positioning Systems

(GPS) devices



10-28

Data Visualization Systems

� DVS

� Represents complex data using interactive,

three-dimensional graphical forms

(charts, graphs, maps)

� Helps users interactively sort, subdivide,

combine, and organize data while it is in its

graphical form



10-29

Using Decision Support Systems

� Using a decision support system involvesan interactive analytical modeling process

� Decision makers are not demanding

pre-specified information� They are exploring possible alternatives

� What-If Analysis

� Observing how changes to selected variables

affect other variables



10-30

Using Decision Support Systems

� Sensitivity Analysis� Observing how repeated changes to a single

variable affect other variables

� Goal-seeking Analysis� Making repeated changes to selected variables

until a chosen variable reaches a target value

� Optimization Analysis

� Finding an optimum value for selected variables,

given certain constraints



10-31

Data Mining

� Provides decision support through knowledgediscovery� Analyzes vast stores of historical business data

� Looks for patterns, trends, and correlations

� Goal is to improve business performance

� Types of analysis� Regression

� Decision tree� Neural network

� Cluster detection

� Market basket analysis



10-32

Analysis of Customer Demographics



10-33

Market Basket Analysis

� One of the most common uses for data mining

� Determines what products customers purchase

together with other products

� Results affect how companies� Market products

� Place merchandise in the store

� Lay out catalogs and order forms� Determine what new products to offer

� Customize solicitation phone calls



10-34

Executive Information Systems

� EIS

� Combines many features of MIS and DSS

� Provide top executives with immediate and

easy access to information� Identify factors that are critical to accomplishing

strategic objectives (critical success factors)

� So popular that it has been expanded to managers,

analysis, and other knowledge workers



10-35

Features of an EIS

� Information presented in forms tailored to thepreferences of the executives using the system

� Customizable graphical user interfaces

� Exception reports� Trend analysis

� Drill down capability





10-37

Dashboard Example



10-38

Enterprise Information Portal Components



10-39

Enterprise Knowledge Portal



10-40

Case 2: Automated Decision Making

� Automated decision making has been slowto materialize

� Early applications were just solutions looking

for problems, contributing little to improved

organizational performance

� A new generation of AI applications

� Easier to create and manage

� Decision making triggered without humanintervention

� Can translate decisions into action quickly,

accurately, and efficiently



10-41

Case 2: Automated Decision Making

� AI is best suited for � Decisions that must be made quickly and

frequently, using electronic data

� Highly structured decision criteria

� High-quality data

� Common users of AI

� Transportation industry

� Hotels

� Investment firms and lenders



10-42


� Why did some previous attempts to use artificialintelligence technologies fail?

� What key differences of the new AI-based

applications versus the old cause the authors

to declare that automated decision making is

coming of age?

� What types of decisions are best suited for

automated decision making?



10-43


� What role do humans plan in automateddecision-making applications?

� What are some of the challenges faced by

managers where automated decision-making

systems are being used?

� What solutions are needed to meet such

challenges?



10-44

Artificial Intelligence (AI)

� AI is a field of science and technology based on� Computer science

� Biology

� Psychology

� Linguistics� Mathematics

� Engineering

� The goal is to develop computers than cansimulate the ability to think � And see, hear, walk, talk, and feel as well



10-45

Attributes of Intelligent Behavior

� Some of the attributes of intelligent behavior � Think and reason

� Use reason to solve problems

� Learn or understand from experience� Acquire and apply knowledge

� Exhibit creativity and imagination

� Deal with complex or perplexing situations



10-46

Attributes of Intelligent Behavior

� Attributes of intelligent behavior (continued)� Respond quickly and successfully to new

situations

� Recognize the relative importance of elements in a situation

� Handle ambiguous, incomplete, or

erroneous information



10-47

Domains of Artificial Intelligence



10-48

Cognitive Science

� Applications in the cognitive science of AI� Expert systems

� Knowledge-based systems

� Adaptive learning systems

� Fuzzy logic systems

� Neural networks

� Genetic algorithm software

� Intelligent agents

� Focuses on how the human brain works

and how humans think and learn



10-49

Robotics

� AI, engineering, and physiology are the basicdisciplines of robotics� Produces robot machines with computer

intelligence and humanlike physical capabilities

� This area include applications designed to

give robots the powers of � Sight or visual perception

� Touch� Dexterity

� Locomotion

� Navigation



10-50

Natural Interfaces

� Major thrusts in the area of AI and thedevelopment of natural interfaces� Natural languages

� Speech recognition

� Virtual reality

� Involves research and development in� Linguistics

� Psychology� Computer science

� Other disciplines



10-51

Latest Commercial Applications of AI

� Decision Support� Helps capture the why as well as the what of

engineered design and decision making

� Information Retrieval� Distills tidal waves of information into simple

presentations

� Natural language technology

� Database mining



10-52

Latest Commercial Applications of AI

� Virtual Reality� X-ray-like vision enabled by enhanced-reality

visualization helps surgeons

� Automated animation and haptic interfaces

allow users to interact with virtual objects

� Robotics

� Machine-vision inspections systems

� Cutting-edge robotics systems� From micro robots and hands and legs, to cognitive

and trainable modular vision systems



10-53

Expert Systems

� An Expert System (ES)� A knowledge-based information system

� Contain knowledge about a specific, complex

application area� Acts as an expert consultant to end users



10-54

Components of an Expert System

� Knowledge Base� Facts about a specific subject area

� Heuristics that express the reasoning procedures

of an expert (rules of thumb)

� Software Resources� An inference engine processes the knowledge

and recommends a course of action

� User interface programs communicate withthe end user

� Explanation programs explain the reasoning

process to the end user



10-55

Components of an Expert System



10-56

Methods of Knowledge Representation

� Case-Based� Knowledge organized in the form of cases

� Cases are examples of past performance,

occurrences, and experiences

� Frame-Based� Knowledge organized in a hierarchy or

network of frames

� A frame is a collection of knowledge aboutan entity, consisting of a complex package

of data values describing its attributes



10-57

Methods of Knowledge Representation

� Object-Based� Knowledge represented as a network of objects

� An object is a data element that includes both

data and the methods or processes that act on

those data

� Rule-Based

� Knowledge represented in the form of rules

and statements of fact

� Rules are statements that typically take the

form of a premise and a conclusion (If, Then)



10-58

Expert System Application Categories

� Decision Management� Loan portfolio analysis

� Employee performance evaluation

� Insurance underwriting

� Diagnostic/Troubleshooting

� Equipment calibration

� Help desk operations

� Medical diagnosis

� Software debugging



10-59


� Design/Configuration� Computer option installation

� Manufacturability studies

� Communications networks

� Selection/Classification

� Material selection

� Delinquent account identification

� Information classification� Suspect identification

� Process Monitoring/Control



10-60


� Process Monitoring/Control� Machine control (including robotics)

� Inventory control

� Production monitoring

� Chemical testing



10-61

Benefits of Expert Systems

� Captures the expertise of an expert or group of experts in a computer-based information system

� Faster and more consistent than an expert

� Can contain knowledge of multiple experts� Does not get tired or distracted

� Cannot be overworked or stressed

� Helps preserve and reproduce the knowledge

of human experts



10-62

Limitations of Expert Systems

� The major limitations of expert systems� Limited focus

� Inability to learn

� Maintenance problems� Development cost

� Can only solve specific types of problems

in a limited domain of knowledge





10-64

Developing Expert Systems

� Suitability Criteria for Expert Systems� Structure: the solution process must be able

to cope with ill-structured, uncertain, missing,

and conflicting data and a changing problem

situation

� Availability: an expert exists who is articulate,

cooperative, and supported by the management

and end users involved in the developmentprocess



10-65

Development Tool

� Expert System Shell� The easiest way to develop an expert system

� A software package consisting of an expert

system without its knowledge base� Has an inference engine and user interface

programs



10-66

Knowledge Engineering

� A knowledge engineer � Works with experts to capture the knowledge

(facts and rules of thumb) they possess

� Builds the knowledge base, and if necessary,the rest of the expert system

� Performs a role similar to that of systems

analysts in conventional information systems

development



10-67

Neural Networks

� Computing systems modeled after the brain¶smesh-like network of interconnected processing

elements (neurons)

� Interconnected processors operate in parallel

and interact with each other

� Allows the network to learn from the data it

processes



10-68

Fuzzy Logic

� Fuzzy logic� Resembles human reasoning

� Allows for approximate values and

inferences and incomplete or ambiguous data

� Uses terms such as ³very high´ instead of

precise measures

� Used more often in Japan than in the U.S.

� Used in fuzzy process controllers used insubway trains, elevators, and cars



10-69

Example of Fuzzy Logic Rules and Query



10-70

Genetic Algorithms

� Genetic algorithm software� Uses Darwinian, randomizing, and other

mathematical functions

� Simulates an evolutionary process, yielding

increasingly better solutions to a problem

� Being uses to model a variety of scientific,

technical, and business processes

� Especially useful for situations in whichthousands of solutions are possible



10-71

Virtual Reality (VR)

� Virtual reality is a computer-simulated reality� Fast-growing area of artificial intelligence

� Originated from efforts to build natural, realistic,

multi-sensory human-computer interfaces

� Relies on multi-sensory input/output devices

� Creates a three-dimensional world through

sight, sound, and touch

� Also called telepresence



10-72

Typical VR Applications

� Current applications of virtual reality� Computer-aided design

� Medical diagnostics and treatment

� Scientific experimentation� Flight simulation

� Product demonstrations

� Employee training

� Entertainment



10-73

Intelligent Agents

� A software surrogate for an end user or aprocess that fulfills a stated need or activity

� Uses built-in and learned knowledge base

to make decisions and accomplish tasks in

a way that fulfills the intentions of a user

� Also call software robots or bots



10-74

User Interface Agents

� Interface Tutors ± observe user computer operations, correct user mistakes, provide

hints/advice on efficient software use

� Presentation Agents ± show information in a

variety of forms/media based on user preferences

� Network Navigation Agents ± discover paths

to information, provide ways to view it based

on user preferences

� Role-Playing ± play what-if games and other

roles to help users understand information and

make better decisions



10-75

Information Management Agents

� Search Agents ± help users find files anddatabases, search for information, and suggest

and find new types of information products,

media, resources

� Information Brokers ± provide commercialservices to discover and develop information

resources that fit business or personal needs

� Information Filters ± Receive, find, filter,

discard, save, forward, and notify users about

products received or desired, including e-mail,

voice mail, and other information media

C 3 C li d B i I lli



10-76

Case 3: Centralized Business Intelligence

� A reinventing-the-wheel approach to businessintelligence implementations can result in

� High development costs

� High support costs

� Incompatible business intelligence systems

� A more strategic approach

� Standardize on fewer business intelligence tools

� Make them available throughout the organization,

even before projects are planned

C 3 C t li d B i I t lli



10-77

Case 3: Centralized Business Intelligence

� About 10 percent of the 2,000 largest companieshave a business intelligence competency center

� Centralized or virtual

� Part of the IT department or independent

� Cost reduction is often the driving force behind

creating competency centers and consolidating

business intelligence systems

� Despite the potential savings, funding for

creating and running a BI center can be an issue

C St d Q ti



10-78


� What is business intelligence?� Why are business intelligence systems such

a popular business application of IT?

� What is the business value of the variousBI applications discussed in the case?

� Is the business intelligence system an MIS

or a DSS?

C 4 R b t th C D i t



10-79

Case 4: Robots, the Common Denominator

� In early 2004, 22 patients underwent complexlaparoscopic operations

� The operations included colon cancer

procedures and hernia repairs

� The primary surgeon was 250 miles away

� A three-armed robot was used to perform the

procedures

� Left arm, right arm, camera arm

C 4 R b t th C D i t



10-80

Case 4: Robots, the Common Denominator

� Automakers heavily use robotics� Ford has a completely wireless assembly factory

� It also have a completely automated body shop

� BMW has two wireless plants in Europe andis setting one up in the U.S.

� Vehicle tracking and material replenishment

are automated as well

C St d Q ti



10-81


� What is the current and future business valueof robotics?

� Would you be comfortable with a robot

performing surgery on you?� The robotics being used by Ford Motor Co. are

contributing to a streamlining of its supply chain

� What other applications of robots can youenvision to improve supply chain management

beyond those described in the case?



Date post:	08-Apr-2018
Category:	Documents
Upload:	jitender-singh
View:	217 times
Download:	0 times

Lecture-01-DSS

Documents