|
DECISION SUPPORT SYSTEMS AND EXPERT SYSTEMS |
|
ch 1 Introduction to DSS |
ch 8 Implementing |
ch 15 Expert systems outside |
|
ch 2 Decision Processes |
ch 9 Representational Models |
ch 16 Expert systems inside |
|
ch 3 Systems and Models |
ch 10 Optimizations |
ch 17 Building an Expert Syste |
|
ch 4 Types of DSS's |
ch 11 Group Decision System |
ch 18 Expert System Cases |
|
ch 5 Building an DSS |
ch 12 Executive Info Systems |
Pulling it all together |
|
ch 6 DSS Software tools. |
ch 13 DSS Cases |
|
|
ch 7 DSS Hardware |
ch 14 Artificial Intelligence |
|
CHAPTER 1 INTRODUCTION.
EVOLUTION OF INFORMATION SYSTEMS.
WHAT IS A DSS.
DSS IN THE INFORMATION SYSTEMS PICTURE.
TYPES OF INFORMATION SYSTEMS.
INFORMATION SYSTEMS AND DECISION SUPPORT.
USING COMPUTERS FOR DECISION SUPPORT.
THE VALUE OF COMPUTER BASED DECISION SUPPORT.
SPECIFIC DSS BENEFITS.
IMPROVING PERSONAL EFFICIENCY.
EXPEDITING PROBLEM SOLVING.
FACILITATING INTERPERSONAL COMMUNICATIONS.
PROMOTING LEARNING OR TRAINING.
INCREASING ORGANIZATIONAL CONTROL.
WHY STUDY DSS.
THE PLAN OF THIS BOOK.
SUMMARY.
CHAPTER 2 HUMAN DECISION MAKING PROCESSES.
WHAT IS A DECISION.
THE DECISION PROCESS.
THE INTELLIGENCE PHASE.
THE DESIGN PHASE.
THE IMPORTANCE OF CREATIVITY.
THE CHOICE PHASE.
TYPES OF DECISIONS.
HOW MANAGERS MAKE DECISIONS.
THE RATIONAL MANAGER.
SUBJECTIVE UTILITY.
SYSTEMATIC DECISION MAKING.
SATISFICING.
ORGANIZATIONAL AND POLITICAL DECISION MAKING.
THE IMPACT OF PHYCHOLOGICAL TYPE ON DECISION MAKING.
THE KEPNER-TREGOE DECISION MAKING METHOD.
STATE THE PURPOSE OF THE DECISION.
ESTABLISH OBJECTIVES.
CLASSIFY ACCORDING TO IMPORTANCE.
IDENTIFY MUST OBJECTIVES.
IDENTIFY WANT OBJECTIVES.
QUANTIFY WANT OBJECTIVES.
GENERATE ALTERNATIVES.
EVALUATE ALTERNATIVES AGAINST OBJECTIVES.
COMPARE WITH MUST OBJECTIVES.
COMPARE WITH WANT OBJECTIVES.
UNITE SEPARATE JUDGMENTS.
TENTATIVELY CHOOSE THE BEST ALTERNATIVE.
ASSESS ADVERSE CONSEQUENCES.
LIST POSSIBLE ADVERSE CONSEQUENCES.
WEIGH THE ADVERSE CONSEQUENCES.
MAKE A FINAL CHOICE.
SUMMARY.
CHAPTER 3 SYSTEMS AND MODELS.
INTRODUCTION.
ABOUT SYSTEMS.
INFORMATION SYSTEMS.
PRECOMPUTER INFORMATION SYSTEM TECHNOLOGY.
DATA FLOW DIAGRAMS.
DSS AS INFORMATION SYSTEMS.
MODELS.
TYPES OF MODELS.
SYSTEM VERSUS PROCESS MODELS,WHAT ARE WE MODELING.
STATIC VS. DYNAMIC MODELS: CAUSE AND EFFECT OVER TIME.
CONTINUOUS VS. DISCRETE EVENT MODELS, HOW DO QUANTITIES VARY IN THE SYSTEM.
DETERMINISTIC VS. STOCHASTIC MODELS. STATISTICAL UNCERTAINTY.
SIMPLIFICATION IN MODELS.
SUMMARY
CHAPTER 4 TYPES OF DECISION SUPPORT SYSTEMS.
INTRODUCTION.
THE DSS HIERARCHY.
THE 7 DSS TYPES.
FILE DRAWER SYSTEMS.
DATA ANALYSIS SYSTEMS.
ANALYSIS INFORMATION SYSTEMS.
ACCOUNTING MODELS.
REPRESENTATIONAL MODELS.
OPTIMIZATION MODELS.
SUGGESTION MODELS.
APPLYING THE DSS TYPES TO AIRPLANE YIELD MANAGEMENT.
THE YIELD MANAGEMENT PROBLEM.
APPLYING DSS TO YIELD MANAGEMENT.
GENERALIZING THE DSS CATEGORIES.
MATCHING DSS TO THE DECISION TYPE.
INDIVIDUAL AND GROUP DSS.
MATCHING BENEFITS TO DSS USER COMMUNITY.
MATCHING DSS TO THE DECISION MAKER’S PSYCHOLOGICAL TYPE.
INTROVERSION / EXTRAVERSION.
SENSING / INTUITION.
THINKING / FEELING.
JUDGMENT / PERCEPTION.
COMBINATIONS OF PREFERENCES.
USAGE
MODES.
INSTITUTIONAL VS. AD-HOC DSS.
SUMMARY.
CHAPTER 5 BUILDING A DECISION SUPPORT SYSTEM.
INTRODUCTION.
DEFINING THE DSS ARCHITECTURE.
DSS DEVELOPMENT PROJECT PARTICIPANTS.
THE DSS DEVELOPMENT PROCESS.
THE SDLC APPROACH.
ADVANTAGES OF SDLC.
DRAWBACKS OF SDLC.
PROTOTYPING.
ADVANTAGES OF PROTOTYPING.
DISADVANTAGES OF PROTOTYPING.
END USER DEVELOPMENT.
ADVANTAGES OF END USER COMPUTING FOR DSS DEVELOPMENT.
DISADVANTAGES OF END USER COMPUTING FOR DSS DEVELOPMENT.
DSS USER INTERFACES.
FACTORS TO CONSIDER IN USER INTERFACE DESIGN.
A CAUTIONARY NOTE.
USER INTERFACE STYLES.
HYPERTEXT / HYPERMEDIA.
SUMMARY.
CHAPTER 6 DSS SOFTWARE TOOLS.
INTRODUCTION.
DSS SOFTWARE CATEGORIES.
STANDARD PACKAGES.
SPECIALIZED TOOLS AND GENERATORS.
DATABASE MANAGEMENT SYSTEMS.
INFORMATION
RETRIEVAL PACKAGES.
DATA RETRIEVAL FOR DSS USING SQL.
SPECIALIZED MODELING LANGUAGES.
STATISTICAL DATA ANALYSIS PACKAGES.
FORECASTING PACKAGES.
GRAPHING PACKAGES.
PROGRAMMING LANGUAGES FOR DSS.
THIRD GENERATION PROGRAMMING LANGUAGES.
FOURTH GENERATION PROGRAMMING LANGUAGES.
SUMMARY.
CHAPTER 7 DSS HARDWARE AND OPERATING SYSTEM PLATFORMS.
INTRODUCTION
THE MAJOR OPTIONS.
DSS ON THE CENTRAL CORPORATE SYSTEM.
ADVANTAGES OF USING THE CENTRAL SYSTEM.
DISADVANTAGES OF USING THE CENTRAL SYSTEM.
POPULAR MAINFRAME SYSTEMS.
DSS WITH AN INFORMATION BASE ON A SEPARATE SYSTEM.
ADVANTAGES OF USING AN INFORMATION BASE.
DISADVANTAGES OF USING AN INFORMATION BASE.
POPULAR MINICOMPUTER SYSTEMS FOR INFORMATION BASE USAGE.
DSS AND CLIENT / SERVER COMPUTING.
ADVANTAGES OF USING A CLIENT BASED COMPUTER SYSTEM.
DISADVANTAGES OF USING A CLIENT SERVER COMPUTER SYSTEM.
POPULAR LOCAL AREA NETWORK SYSTEMS.
DSS ON A STAND ALONE SYSTEM.
ADVANTAGES OF USING A STAND ALONE SYSTEM.
DISADVANTAGES OF USING A STAND ALONE SYSTEM.
POPULAR STAND ALONE SYSTEMS.
OPEN SYSTEMS AND DSS.
DOWNSIZING FOR DECISION SUPPORT.
CHOOSING A DSS HARDWARE ENVIRONMENT.
SUMMARY.
CHAPTER 8 IMPLEMENTING DECISION SUPPORT SYSTEMS.
INTRODUCTION.
THE IMPLEMENTATION STAGE.
SYSTEM CONVERSION.
OVERCOMING RESISTANCE TO CHANGE.
UNFREEZING.
MOVING.
REFREEZING.
DSS IMPLEMENTATION ISSUES.
TECHNICAL DSS IMPLEMENTATION ISSUES.
UNFAMILIARITY WITH THIS TYPE OF SYSTEM.
RESPONSE TIME.
RELIABILITY AND AVAILABILITY.
POOR DATA QUALITY.
USER RELATED DSS IMPLEMENTATION.
USER AND MANAGEMENT SUPPORT.
UNSTABLE USER COMMUNITY.
RESPONSE TIME.
TRAINING.
AVAILABILITY OF SUPPORT.
VOLUNTARY OR MANDATORY USE.
USING THE LIST OF ISSUES.
ETHICAL ISSUES IN DSS IMPLEMENTATION.
STORAGE OF INFORMATION.
USE OF INFORMATION.
SHARING OF INFORMATION.
HUMAN JUDGMENT.
COMBINING INFORMATION.
ERROR DETECTION AND CORRECTION.
SUMMARY.
CHAPTER 9 REPRESENTATIONAL MODELS.
INTRODUCTION.
DISCRETE EVENT SIMULATION MODELS.
THE CONCEPT OF DISCRETE EVENT SIMULATION.
DESIGNING A DISCRETE EVENT SIMULATION MODEL.
THE CONCEPT OF "SYSTEM STATE"
ANOTHER SIMULATION EXAMPLE.
COMPLETE SIMULATION STUDIES.
RANDOM AND PSEUDO-RANDOM NUMBERS.
STATIC SIMULATION MODELS
QUEUING MODELS.
QUEUING THEORY CONCEPTS.
A QUEUING THEORY EXAMPLE.
GENERALIZING THE SOLUTION.
ARRIVAL AND DEPARTURE TIME DISTRIBUTIONS.
MARKOV PROCESS MODELS.
COMPUTER CALCULATION FOR MARKOV PROCESSES.
SIMULATION, QUEUING THEORY, AND MARKOV PROCESS COMPARED.
SUMMARY.
CHAPTER 10 OPTIMIZATION.
INTRODUCTION.
TESTING ALTERNATIVES.
COMPLETE ENUMERATION.
RANDOM SEARCH.
THE CALCULUS APPROACH.
LINEAR PROGRAMMING.
NUMERICAL METHODS.
HILL CLIMBING.
BOX’S METHOD.
SUMMARY.
CHAPTER 11 GROUP DECISION SUPPORT SYSTEMS.
INTRODUCTION.
WHAT IS GROUP DSS.
WHY GROUP DSS NOW.
ORGANIZATIONAL REASONS FOR GROUP DSS GROWTH.
GROUP THINK.
TECHNICAL REASONS FOR GROUP DSS GROWTH.
PUTTING THE FACTORS TOGETHER.
GROUP VS. INDIVIDUAL ACTIVITIES.
TYPES OF GROUP DSS.
GROUPWARE.
GROUP DSS IN USE TODAY.
ELECTRONIC MEETING SYSTEMS.
USING OPTION FINDER.
WORK FLOW SYSTEMS.
FOUR GROUP DSS PRODUCTS.
ACCESS TECHNOLOGY’S FORCOMMENT.
ON TECHNOLOGY INSTANT UPDATE.
LOTUS NOTES.
GROUP BULL FLOWPATH
SUMMARY.
CHAP 12 EXECUTIVE INFORMATION SYSTEMS
INTRODUCTION
WHO ARE THE EXECUTIVES.
WHAT IS AN EXECUTIVE INFORMATION SYSTEM.
WHY USE AN EXECUTIVE INFORMATION SYSTEM OR AN EXECUTIVE SUPPORT SYSTEM.
EIS CHARACTERISTICS. GENERAL FEATURES.
EIS DESIGN APPROACHES.
EIS ISSUES.
WHO IS THE USER.
THE EIS SPONSOR.
COST OF THE EIS
MANAGEMENT RESISTANCE TO THE EIS.
EMPLOYEE RESISTANCE TO THE EIS.
FROM EIS TO ESS.
IMPLEMENTING THE EIS / ESS.
SUMMARY.
CHAP 13 DECISION SUPPORT SYSTEMS CASES.
MBTA PASSENGER WAITING TIME SYSTEM
MEDIQUAL
JOCK^2
OPTION PRICING WITH BLACK-SCHOLES
GEOGRAPHIC INFORMATION SYSTEMS GIS
GIS EXAMPLE 1 YELLOW FREIGHT
SERVICE MAP CREATION PROCESS AUTOMATED WITH GIS.
GIS USED TO TRACK TERMINAL SERVICE AREAS.
GIS APPLICATIONS EXPECTED TO GROW WITH FUTURE SYSTEM DEVELOPMENT.
GIS EXAMPLE 2 METROPOLITAN LIFE INSURANCE.
DISCUSSION
CHAP 14 ARTIFICIAL INTELLIGENCE, EXPERT SYSTEMS, AND DSS.
INTRODUCTION.
ABOUT A.I.
HISTORY OF A.I.
ONE HUNDRED DEFINATIONS OF A.I.
THE TURING TEST
ARTIFICIAL INTELLIGENCE TODAY.
THE ELECTRONIC PSYCHIATRIST.
ROBOTICS
MACHINE VISION.
NATURAL LANGUAGE INTREPRETATION
VOICE RECOGNITION.
NEURAL NETS.
EXPERT SYSTEMS.
AMERICAN AIRLINES
THE BASIC IDEA.
A SIMPLE EXPERT SYSTEM.
EXPERT SYSTEMS AND DSS.
EXPERT SYSTEMS AND NEURAL NETWORKS.
SUMMARY.
EXAMPLE RENT A CAR AT ALAMO.
CHAPTER 15 EXPERT SYSTEMS FROM THE OUTSIDE.
INTRODUCTION.
PROS AND CONS OF EXPERT SYSTEMS.
ADVANTAGES OF EXPERT SYSTEMS.
DRAWBACKS OF EXPERT SYSTEMS.
CHOOSING A GOOD EXPERT SYSTEM APPLICATION.
PROBLEM RELATED (TASK RELATED) CRITERIA.
EXPERT RELATED CRITERIA.
KEEPING THE CRITERIA IN PERPECTIVE.
THE EXPERT SYSTEM HUMAN INTERFACE.
INTERFACES OF AN EXPERT SYSTEM.
THE USER CONTROL AND INPUT INTERFACES.
THE USER OUTPUT INTERFACE.
THE EXPLANATION INTERFACE.
HOW AND WHY EXPLANATIONS.
WHAT IF EXPLANATIONS.
THE KNOWLEDGE ENGINEER INTERFACE.
DEFINING THE KNOWLEDGE BASE.
TRACE OUTPUT.
EXPERT SYSTEMS AND LEGAL ISSUES.
SUMMARY.
EXAMPLE ENVOYANCE, WHAT METHODS TO USE GET THIS FOOL !
CHAPTER 16 EXPERT SYSTEMS FROM THE INSIDE.
INTRODUCTION.
VARIABLES.
PRODUCTION RULES.
COMBINING PRODUCTION RULES INTO A KNOWLEDGE BASE.
INFERENCING METHODS.
BACKWARD CHAINING.
FORWARD CHAINING.
CHOOSING BETWEEN BACKWARD AND FORWARD CHAINING.
STRENGTHS AND WEAKNESSES OF PRODUCTION RULES.
FRAMES.
SEMANTIC NETWORKS.
OBJECT ORIENTED PROGRAMMING.
EXAMPLE OF OBJECTS.
BENEFITS OF OBJECT ORIENTED PROGRAMMING.
FRAME CONCEPTS.
REASONING WITHIN THE FRAMES.
BEYOND THE INDIVIDUAL FRAMES.
PREDICATE CALCULUS.
DATABASES.
CONFIDENCE FACTORS.
WHERE CONFIDENCE FACTORS COME FROM.
ALTERNATIVE APPROACHES TO CONFDENCE FACTORS.
FUZZY LOGIC.
USING FUZZY LOGIC FOR COMMERCIAL LOAN ANALYSIS.
SUMMARY.
CHAPTER 17 BUILDING AN EXPERT SYSTEM.
INTRODUCTION
HARDWARE PLATFORMS FOR EXPERT SYSTEMS.
EXPERT SYSTEMS AND MICROCOMPUTERS.
EXPERT SYSTEMS AND MINI COMPUTERS.
SYMBOLIC PROCESSORS.
EXPERT SYSTEMS AND MAINFRAMES.
EXPERT SYSTEMS AND PARALLEL PROCESSORS.
SOFTWARE TOOLS FOR EXPERT SYSTEMS.
PACKAGES.
SHELLS.
LANGUAGES
SPECIALIZED AI LANGUAGES
LISP
PROLOG.
CONCLUDING COMMENTS.
GENERAL PURPOSE LANGUAGES IN EXPERT SYSTEMS.
KNOWLEDGE ACQUISITION.
WHAT KNOWLEDGE ACQUISTION IS.
FINDING THE EXPERTS.
ONE EXPERT OR MORE.
CHARACTERISTICS OF THE KNOWLEDGE ENGINEER.
STARTING THE KNOWLEDGE ACQUISITION PROCESS.
WORKING WITH THE EXPERTS.
FROM THE EXPERT TO THE COMPUTER.
INDUCTION : AN ALTERNATIVE TO THE KNOWLEDGE ELICITATION.
BEYOND THE KNOWLEDGE ENGINEER.
COMBINIG EXPERT SYSTEMS AND DSS.
SUMMARY.
CHAPTER 18 EXPERT SYSTEMS CASES.
INTRODUCTION.
XCON.
HOW XCON WORKS.
XCON PERFORMANCE.
XCON BENEFITS.
KEY ROLES IN XCON.
LESSONS LEARNED FROM XCON.
EXPERT SYSTEM CALCULATES SPACE SHUTTLE PAYLOAD CONFIGURATION.
ON THE INSIDE.
TRUCK BRAKE BALANCING.
CHAPTER 19 PULLING IT ALL TOGETHER.
SYSTEMS INTEGRATION AND THE FUTURE OF DSS.
INTRODUCTION.
COMBINING THE PIECES.
WHAT IS SYSTEMS INTEGRATION.
A SYSTEMS INTEGRATION EXAMPLE.
TYPES OF SYSTEMS INTEGRATION.
SINGLE SYSTEM VISIBILITY VERSUS MULTI SYSTEM VISIBILITIES.
ONE HARDWARE PLATFORM VS. SEVERAL HARDWARE PLATFORM.
ONE LOCATION VERSUS MANY MULTIPLE LOCATIONS.
TRENDS IN SYSTEMS INTEGRATION.
HIRING A SYSTEMS INTEGRATOR.
THE FUTURE OF DSS.
DSS CAPABILITY WILL BECOME A STANDARD AND EXPECTED PART OF INFORMATION SYSTEMS.
DECISION MAKERS WILL BECOME COMPUTER LITERATE.
HARDWARE TECHNOLOGY WILL CONTINUE TO EVOLVE.
NEW SOFTWARE TECHNOLOGIES WILL ENTER THE MAINSTREAM.
USAGE WILL CONTINUE AWAY FROM CHARACTER BASED INTERFACES.
THE JOB MARKET FOR INFORMATION SYSTEMS PROFESSIONALS WITH A BUSINESS PERPECTIVE WILL REMAIN STRONG.
IN CONCLUSION.
SUMMARY
1. Name the components of DSS and briefly describe each.
User interface.
Modeling based systems.
Dbms systems.
Knowledge based systems.
1. Name the different types of user interfaces and briefly describe each.
Command driven.
Menu driven.
Icon based.
Natural language.
Voice based.
2. What is the origin of the term DSS and where did it come from.
3. How do we measure the integration of the DSS. 2 types.
Unitergrated= how ez can operator move around applications.
Quasi integrated. = how ez can data be accessed across platforms.
4. Define a relation.
A named 2 dimension table of data.
5. Define a NORMALIZED relation.
Domain is String### or number.
Process of turning complex structures into simple stable ones.
6. Define a relational datamodel.
Is a collection of normalized relations of associated degrees.
Data model that states data in the form of tables or relations.
7. Define 1NF, 2NF, 3NF.
1NF = a table with no repeating values.
3NF= a normal relation is said to be in 3NF if the non-key domains
1.Mutually independent. 2. Fully functionally dependent on the primary key.
Look for abnormalities such as deleting a customer and losing the salesman.
Insertion / deletion / modification.
Insertion ,deletion will have to be made in several entries.
A 2NF will lose data like above.
A 2NF will have transitive dependency. 2 or more.
Will have to delete data in several different tables. Or delete/ modify.
Can delete important data while not knowing, delete customer lose salesman.
A 3NF will only lose the customer. Break down into tables.
Each table has it’s own primary key. Each his own. Or her.
8. Given a Relational data model, look at relations,show functional dependency.
Like a 2 NF form. Different attributes answer to different primary keys.
9. Given this model ,decide if 1NF,2NF,3NF.
1 NF will be a table with no repeating groups. P215
2NF will be perfect except abnormalities (delete,add,modify.)
Two many specifics in each table. Break down into tables for each.
For 2NF live one or two many.
10. Change this model to 3NF.
Read above put in access or paradox form. Like labs.
11. Fill in general descriptions of Paradox with a main menu. ~ to retrieve data,choose what. View ask report create modify images form tools scripts help exit
12. Describe the function of the Paradox Personal Programmer.
To help develop ,help is usually at bottom, this is a linear process.similar to paradox itself. Maintains ease of use.
13. What is the purpose of the Wait Table Command. Todays notes on edit.
14. Personal Programmer top menu fill in. no menu.
Create modify summarize review play tools
15. Given a Paradox Relation model,fill in query form to retrieve specific information. Use check marks.
Look at graphics.
16. Steps to develop an application in Personal Programmer define Main Menu and attach the Building Blocks.
17. Define a Mathematical Model.
A abstract form of engineering using mathematical tools. Reality.
18. 2 benefits of a Mathematical Model.
Cheap, easier, more models,gains more information about object.
Eliminate clutter details.
19. Given Model Statements, list what type.
Static model which time is not involved or (explicitly) used.
Dynamic time is explicitly used.
Deterministic variable are not specifically represented.
Stochastic variables are specifically represented.
20. Given Objective Function and Constraints , Graph it.
See graphs.
Multiply one constraint by a number to bring one variable to zero.
Then solve for the one and substitute into either original constraint.
These are your intersection point.
21. Formula for finding Intersection Point .
look on graphics.
22. Simulation Model. Come up with best solution. See handout..