03Sdss13.ppt 303KB May 20 2010 09:29:42 PM

(1)

Chapter 13

Decision Support Systems

(2)

Simon’s Types of Decisions

 Programmed decisionsProgrammed decisions

– repetitive and routine_{repetitive and routine}

– have a definite procedure_{have a definite procedure}

 Nonprogrammed decisionsNonprogrammed decisions

– _{Novel and unstructured}_{Novel and unstructured}

– _{No cut-and-dried method for handling problem}_{No cut-and-dried method for handling problem}

 Types exist on a continuumTypes exist on a continuum

(3)

Simon’s Problem Solving

Phases

 IntelligenceIntelligence

– Searching environment for conditions calling for a solution_{Searching environment for conditions calling for a solution}

 DesignDesign

– Inventing, developing, and analyzing possible courses of _{Inventing, developing, and analyzing possible courses of} action

action  ChoiceChoice

– _{Selecting a course of action from those available}_{Selecting a course of action from those available}

 ReviewReview

(4)

Definitions of a Decision

Support System (DSS)

General definition -

General definition - a system providing both a system providing both

problem-solving and communications capabilities problem-solving and communications capabilities

for semistructured problems for semistructured problems

Specific definition -

Specific definition - a system that supports a a system that supports a single manager or a relatively small group of single manager or a relatively small group of

managers working as a problem-solving team in managers working as a problem-solving team in

the solution of a semistructured problem by the solution of a semistructured problem by

providing information or making suggestions providing information or making suggestions

concerning

concerning specific specific decisions.decisions.

(5)

The DSS Concept

 Gorry and Scott Morton coined the phrase ‘DSS’ in Gorry and Scott Morton coined the phrase ‘DSS’ in 1971, about ten years after MIS became popular

1971, about ten years after MIS became popular

 Decision types in terms of problem structureDecision types in terms of problem structure

– Structured problems can be solved with algorithms and _{Structured problems can be solved with algorithms and} decision rules

decision rules

– _{Unstructured problems have no structure in Simon’s}_{Unstructured problems have no structure in Simon’s} phases

phases

– _{Semistructured problems have structured and}_{Semistructured problems have structured and} unstructured phases

(6)

Degree of Degree of problem problem structure structure

The Gorry and Scott Morton Grid

(7)

Alter’s DSS Types

 In 1976 Steven Alter, a doctoral student In 1976 Steven Alter, a doctoral student

built on Gorry and Scott-Morton framework built on Gorry and Scott-Morton framework

– _{Created a taxonomy of six DSS types}_{Created a taxonomy of six DSS types} – _{Based on a study of 56 DSSs}_{Based on a study of 56 DSSs}

 Classifies DSSs based on “degree of Classifies DSSs based on “degree of

(8)

Levels of Alter’s DSSs

 Level of problem-solving support from Level of problem-solving support from lowest to highest

lowest to highest

– _{Retrieval of information elements}_{Retrieval of information elements} – _{Retrieval of information files}_{Retrieval of information files}

– _{Creation of reports from multiple files}_{Creation of reports from multiple files} – _{Estimation of decision consequences}_{Estimation of decision consequences} – _{Propose decisions}_{Propose decisions}

– _{Make decisions}_{Make decisions}

(9)

Importance of Alter’s

Study

 Supports concept of developing systems Supports concept of developing systems

that address particular decisions that address particular decisions

 Makes clear that DSSs need not be Makes clear that DSSs need not be

(10)

Retrieve Retrieve information information elements elements Analyze Analyze entire entire files files Prepare Prepare reports reports from from multiple multiple files files Estimate Estimate decision decision consequen-ces ces Propose Propose decisions decisions Degree Degree of of problem problem solving solving support support Degree of Degree of complexity of the complexity of the problem-solving problem-solving

system system

Little

Little MuchMuch

Alter’s DSS Types

Make

decisions

decisions

(11)

Three DSS Objectives

1.

1. Assist in solving semistructured problemsAssist in solving semistructured problems 2.

2. Support, not replace, the manager Support, not replace, the manager 3.

3. Contribute to decision effectiveness, rather Contribute to decision effectiveness, rather than efficiency

(12)

GDSS software Mathematical Mathematical Models Models OtherOther

group group members members Database Database GDSS GDSS software software Environment Environment IndividualIndividual problemproblem solverssolvers Decision support system EnvironmentEnvironment

Legend: Data Communication Information

A DSS Model

A DSS Model

(13)

Database Contents

 Used by Three Software Subsystems_{Used by Three Software Subsystems}

– Report writers Report writers

»Special reportsSpecial reports »Periodic reportsPeriodic reports »COBOL or PL/ICOBOL or PL/I »DBMSDBMS

– Mathematical modelsMathematical models

»Simulations Simulations

(14)

Group Decision Support

Systems

 Computer-based system that supports groups of Computer-based system that supports groups of

people engaged in a common task (or goal) and

that provides an interface to a shared environment.

that provides an interface to a shared environment.  Used in problem solvingUsed in problem solving

 Related areasRelated areas

– _{Electronic meeting system (EMS)}_{Electronic meeting system (EMS)}

– _{Computer-supported cooperative work (CSCW)}_{Computer-supported cooperative work (CSCW)} – _{Group support system (GSS)}_{Group support system (GSS)}

– _Groupware_Groupware

(15)

How GDSS Contributes

to Problem Solving

 Improved communicationsImproved communications

 Improved discussion focusImproved discussion focus

(16)

GDSS Environmental

Settings

 Synchronous exchange Synchronous exchange

– Members meet at same time_{Members meet at same time}

– _{Committee meeting is an example}_{Committee meeting is an example}

 Asynchronous exchangeAsynchronous exchange

– Members meet at different times_{Members meet at different times} – _{E-mail is an example}_{E-mail is an example}

 More balanced participation.More balanced participation.

(17)

GDSS Types

 Decision roomsDecision rooms

– Small groups face-to-face_{Small groups face-to-face} – Parallel communication_{Parallel communication} – Anonymity_Anonymity

 Local area decision networkLocal area decision network

– _{Members interact using a LAN}_{Members interact using a LAN}

 Legislative sessionLegislative session

– Large group interaction_{Large group interaction}

 Computer-mediated conferenceComputer-mediated conference

(18)

Smaller Larger

GROUP

GROUP SIZE_SIZE

Face-to-face

Dispersed

Decision Room

Local Area Decision Network

Legislative Session

Computer-Mediated Conference

MEMBER

MEMBER

PROXIMITY

Group Size and Location Determine

GDSS Environmental Settings

(19)

Groupware

 FunctionsFunctions

– E-mail_E-mail – _FAX_FAX

– Voice messaging_{Voice messaging} – _{Internet access}_{Internet access}

 Lotus Notes Lotus Notes

(20)

Electronic mail X X X X

FAX X X O X

Voice messaging O X

Internet access X X O X

Bulletin board system X 3 O

Personal calendaring X X 3 X

Group calendaring X X O X

Electronic conferencing O X 3 3

Task management X X 3 X

Desktop video conferencingO

Database access O X 3

Workflow routing O X 3 X

Reengineering O X 3

Electronic forms O 3 3 O

Group documents O X X O

Main Groupware Functions

IBM TeamWARE Lotus Novell IBM TeamWARE Lotus Novell Function Workgroup Office Notes GroupWise

Function Workgroup Office Notes GroupWise

X = standard feature

(21)

Artificial Intelligence (AI)

The activity of providing such machines as computers with the ability to display behavior that

(22)

History of AI

 Early historyEarly history

– _{John McCarthy coined term, AI, in 1956, at}_{John McCarthy coined term, AI, in 1956, at} Dartmouth College conference.

Dartmouth College conference.

– Logic Theorist (first AI program. Herbert Simon _{Logic Theorist (first AI program. Herbert Simon}

played a part)

– General problem solver (GPS)_{General problem solver (GPS)}

 Past 2 decadesPast 2 decades

– _{Research has taken a back seat to MIS and DSS}_{Research has taken a back seat to MIS and DSS} development

development

(23)

Areas of Artificial Intelligence

Expert Expert

systems

systems AIAI

(24)

Appeal of Expert Systems

 Computer program that codes the Computer program that codes the

knowledge of human experts in the form of knowledge of human experts in the form of

heuristics heuristics

 Two distinctions from DSSTwo distinctions from DSS

– _{1. Has potential to extend manager’s problem-}_{1. Has potential to extend manager’s} problem-solving ability

solving ability

– _{2. Ability to explain how solution was reached}_{2. Ability to explain how solution was reached}

(25)

Know-ledge base

User

User interface

Instructions & information

Solutions &

explanations Knowledge

Inference engine

Problem

Domain

Development engine

Expert

system

(26)

Expert System Model

 User interfaceUser interface

– Allows user to interact with system_{Allows user to interact with system}

 Knowledge baseKnowledge base

– Houses accumulated knowledge_{Houses accumulated knowledge}

 Inference engineInference engine

– Provides reasoning_{Provides reasoning}

– Interprets knowledge base_{Interprets knowledge base}

 Development engineDevelopment engine

– Creates expert system_{Creates expert system}

(27)

User Interface

 User enters:User enters:

– Instructions_Instructions

– _Information_Information

 Expert system provides:Expert system provides:

– Solutions_Solutions

– _{Explanations of}_{Explanations of}

(28)

Knowledge Base

 Description of problem domainDescription of problem domain

 RulesRules

– _{Knowledge representation technique}_{Knowledge representation technique} – _‘_‘_{IF:THEN’ logic}_{IF:THEN’ logic}

– _{Networks of rules}_{Networks of rules}

» Lowest levels provide evidence_{Lowest levels provide evidence}

» _{Top levels produce 1 or more conclusions}_{Top levels produce 1 or more conclusions} » Conclusion is called a Goal variable._{Conclusion is called a Goal variable.}

(29)

Conclusion

Conclusion

Evidence Evidence Evidence Evidence

Conclusion

A Rule Set That Produces One Final

(30)

Rule Selection

 Selecting rules to efficiently solve a Selecting rules to efficiently solve a

problem is difficult problem is difficult

 Some goals can be reached with only a few Some goals can be reached with only a few

rules; rules 3 and 4 identify bird rules; rules 3 and 4 identify bird

(31)

Inference Engine

 Performs reasoning by using the contents of Performs reasoning by using the contents of

knowledge base in a particular sequence knowledge base in a particular sequence

 Two basic approaches to using rulesTwo basic approaches to using rules – _{1. Forward reasoning (data driven)}_{1. Forward reasoning (data driven)}

(32)

Forward Reasoning

(Forward Chaining)

 Rule is evaluated as: Rule is evaluated as:

– (1) true, (2) false, (3) unknown_{(1) true, (2) false, (3) unknown}

 Rule evaluation is an iterative processRule evaluation is an iterative process

 When no more rules can fire, the reasoning When no more rules can fire, the reasoning

process stops even if a goal has not been process stops even if a goal has not been

reached reached

Start with inputs and work to solution

(33)

Rule 1 Rule 1 Rule 3 Rule 3 Rule 2 Rule 2 Rule 4 Rule 4 Rule 5 Rule 5 Rule 7 Rule 7 Rule 8 Rule 8 Rule 9 Rule 9 Rule 10 Rule 10 Rule 11 Rule 11 Rule 12 Rule 12 IF A THEN B IF C THEN D IF M THEN E IF K THEN F IF G

IF B OR D THEN K

IF E THEN L

IF K AND L THEN N

IF M

IF N OR O THEN P

F

IF (F AND H) OR J

IF (F AND H) OR J The The Forward Forward Reasoning Reasoning Process Process T T T T T T T T T Legend: Legend:

(34)

Reverse Reasoning Steps

(Backward Chaining)

 Divide problem into subproblemsDivide problem into subproblems

 Try to solve one subproblemTry to solve one subproblem

 Then try anotherThen try another

Start with solution and work back to inputs

(35)

T Rule 1 Rule 2 Rule 3 Step 4 Step 3 Step 2 Step 1 Step 5

IF A THEN B

IF B OR D THEN K

IF K AND L THEN N

IF N OR O THEN P IF C THEN D IF M THEN E IF E THEN L T

The First Five

(36)

If K Then F Legend: Problems to be solved If G Then H If I Then J If M Then O Step 8

Step 9 Step 7 Step 6

Rule 4

Rule 5

Rule 11 Rule 6

T

IF (F And H) Or J Then M T Rule 9 T T Rule 12 T If N Or O

Then P

The Next Four Problems Are

Identified

(37)

Forward Versus Reverse

Reasoning

 Reverse reasoning is faster than forward Reverse reasoning is faster than forward reasoning

reasoning

 Reverse reasoning works best under certain Reverse reasoning works best under certain conditions

conditions

– _{Multiple goal variables}_{Multiple goal variables} – Many rules_{Many rules}

(38)

Development Engine

 Programming languages Programming languages

– _Lisp_Lisp – Prolog_Prolog

 Expert system shellsExpert system shells

– Ready made processor that can be tailored to a _{Ready made processor that can be tailored to a}

particular problem domain

 Case-based reasoning (CBR)Case-based reasoning (CBR)  Decision treeDecision tree

(39)

Expert System Advantages

 For managersFor managers

– Consider more alternatives_{Consider more alternatives} – _{Apply high level of logic}_{Apply high level of logic}

– _{Have more time to evaluate decision rules}_{Have more time to evaluate decision rules} – Consistent logic_{Consistent logic}

 For the firmFor the firm

(40)

Expert System

Disadvantages

 Can’t handle inconsistent knowledgeCan’t handle inconsistent knowledge

 Can’t apply judgment or intuitionCan’t apply judgment or intuition

(41)

Keys to Successful ES

Development

 Coordinate ES development with strategic planningCoordinate ES development with strategic planning

 Clearly define problem to be solved and understand Clearly define problem to be solved and understand

problem domain

 Pay particular attention to ethical and legal feasibility Pay particular attention to ethical and legal feasibility

of proposed system

 Understand users’ concerns and expectations Understand users’ concerns and expectations

concerning system

Employ management techniques designed to retain

(42)

Neural Networks

 Mathematical model of the human brainMathematical model of the human brain

– Simulates the way neurons interact to process _{Simulates the way neurons interact to process} data and learn from experience

data and learn from experience

 Bottom-up approach to modeling human Bottom-up approach to modeling human

intuition intuition

(43)

The Human Brain

 Neuron -- the information processorNeuron -- the information processor

– Input -- dendrites_{Input -- dendrites}

– Processing -- soma_{Processing -- soma}

– _{Output -- axon}_{Output -- axon}

(44)

Soma (processor)

Axon

Synapse

Dendrites (input)

Axonal Paths (output)

Simple Biological Neurons

(45)

Evolution of Artificial

Neural Systems (ANS)

 McCulloch-Pitts mathematical neuron McCulloch-Pitts mathematical neuron

function (late 1930s) was the starting point function (late 1930s) was the starting point

 Hebb’s learning law (early 1940s)Hebb’s learning law (early 1940s)

 NeurocomputersNeurocomputers

(46)

Current Methodology

 Mathematical models don’t duplicate Mathematical models don’t duplicate

human brains, but exhibit similar abilities human brains, but exhibit similar abilities

 Complex networksComplex networks

 Repetitious trainingRepetitious training

– _{ANS “learns” by example}_{ANS “learns” by example}

(47)

y1

y2

y3

y w1

w2

w3

w

(48)

The The Multi-Layer Layer Perceptron Perceptron Y_n2 IN

IN_n_n

OUT

OUTnn

OUT

OUT₁₁

IN

IN₁₁

Y

Y₁₁

(49)

Knowledge-based Systems

in Perspective

 Much has been accomplished in neural nets Much has been accomplished in neural nets

and expert systems and expert systems

 Much work remainsMuch work remains

 Systems abilities to mimic human Systems abilities to mimic human

intelligence are too limited and regarded as intelligence are too limited and regarded as

(50)

Summary [cont.]

 AIAI

– Neural networks_{Neural networks} – Expert systems_{Expert systems}

 Limitations and promiseLimitations and promise