EFRAIM TURBAN and JAY E. ARONSON

EFRAIM TURBAN and JAY E. ARONSON
DECISION SUPPORT SYSTEMS AND INTELLIGENT SYSTEMS PRENTICE HALL, UPPER SADDLE RIVER, NJ, 1998 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

DECISION MAKING AND COMPUTERIZED SUPPORT
Management Support Systems (MSS) Collection of computerized technologies Objectives to support managerial work to support decision making Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 1: Management Support Systems-- An Overview
Emerging and Advanced Computer Technologies for Supporting the Solution of Managerial Problems These Technologies are Changing Organizational Structure Enabling Business Process Reengineering Changing Management Methods Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.1 Opening Vignette: Decision Support at Roadway Package System (RPS)
Business-to-business small package delivery Extremely competitive Roadway Package System (RPS) started operations in March 1985 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Decision Support System(s) (DSS)
Location Problem Solution: Quantitative location model via a DSS using SAS Major Growth lead to 50 Decision Support Applications in 3 Critical Application Areas: Market Planning and Research Operation and Strategic Planning Sales Support Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

DSS and EIS Upshot Both Managers and Customers access information
Information processing is vital to survival Help management make timely and effective decisions Providing reliable and relevant information, in the correct format, at the right time Help conduct analysis easily and quickly Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. 2 Managers and Decision Making Why Computerized Support
1.2 Managers and Decision Making Why Computerized Support? For RPS Opening Vignette Competition End-user and customer access Special data warehouse Two major technologies DSS (supports managers and marketing analysts) EIS (supports top managers) Diversity of decisions Much internal and external data DSS applications separate from the TPS, but use TPS data Statistical and other quantitative models The MANAGERS are responsible for decision making Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Nature of Managers’ Work
What Do Managers Do? Mintzberg [1980] (Table 1.1) Roles of Managers Interpersonal Figurehead Leader Liason Informational Monitor Disseminator Spokesperson Decisional Entrepreneur Disturbance Handler Resource Allocator Negotiator Managers need information and use computers to support decision making Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.3 Managerial Decision Making and Information Systems
Management is a process by which organizational goals are achieved through the use of resources Resources: Inputs Goal Attainment: Output Measuring Success: Productivity = Outputs / Inputs Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

All managerial activities revolve around decision making
The manager is a decision maker Decision making was considered a pure art--a talent Individual styles used, not systematic quantitative scientific methods Now fast changing, complex environment See Figure 1.1: Major Factors, Trends and Results Impacting on Managerial Decision Making Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Difficult to make decisions
Many alternatives due to improved technology and communication systems Cost of making errors can be very large Information may be difficult to access Decisions must be made quickly Trial-and-error approach to management - Not so Good! Use new tools and techniques Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.4 Managers and Computerized Support
Computer technologies evolve and expand Impacts on organizations and society increasing Interaction and cooperation between people and machines rapidly growing Computerized systems are now used in complex managerial areas Information Technology is vital to the business (Survey says “#1 issue) and executives are implementing technologies extensively (Caldwell [1995]) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Computer Applications Evolving from TPS and MIS
to Proactive Applications (DSS) New Modern Management Tools in Data access On-line analytical processing (OLAP) Internet / Intranet / Web for decision support Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.5 The Need for Computerized Decision Support and the Supporting Technologies
Speedy computations Overcoming cognitive limits in processing and storage Cognitive limits may restrict an individual’s problem solving capability Cost reduction Technical support Quality support Competitive edge: business processes reengineering and empowerment Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Primary Decision Support Technologies
Management Support Systems (MSS) Decision Support Systems (DSS) Group Support Systems (GSS), including Group DSS (GDSS) Executive Information Systems (EIS) Expert Systems (ES) Artificial Neural Networks (ANN) Hybrid Support Systems Cutting Edge Intelligent Systems (Genetic Algorithms, Fuzzy Logic, Intelligent Agents, ...) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.6 A Classic Framework for Decision Support
Figure 1.2 (Proposed by Gorry and Scott Morton [1971]) Combination of Simon [1977] Taxonomy and Anthony [1965] Taxonomy Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Simon: Decision-making along a continuum
Highly structured (programmed) decisions to Highly unstructured (nonprogrammed) decisions Simon: Three Phase decision-making process Intelligence--searching for conditions that call for decisions Design--inventing, developing, and analyzing possible courses of action Choice--selecting a course of action from those available Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Unstructured problem has no structured phases
Semistructured problem has some (or some parts with) structured phases Structured problem has all structured phases Procedures for obtaining the best solution are known Objectives are clearly defined Management support systems such as DSS and ES can be useful Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Unstructured problem often solved with human intuition
Semistructured problems fall in between. Solve with both standard solution procedures and human judgment A Decision Support System can help managers understand problems in addition to providing solutions Goal of DSS: Increase the Effectiveness of Decision Making Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Anthony’s Taxonomy [1965] Broad Categories encompass ALL managerial activities Strategic planning Management control Operational control Combine Anthony and Simon’s Taxonomies Decision Support System (DSS) for semistructured and unstructured decisions MIS and management science (MS) approaches insufficient Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Computer Support for Structured Decisions
Since the 1960s Repetitive in nature High level of structure Can abstract and analyze them, and classify them into prototypes Solve with quantitative formulas or models Method: called Management Science (MS) or Operations Research (OR) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Management Science Scientific approach to automate managerial decision making 1.Define the problem 2.Classify the problem into a standard category 3.Construct a mathematical model 4.Find and evaluate potential solutions to model 5.Choose and recommend a solution to problem Modeling: Transforming the real-world problem into an appropriate prototype structure Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.7 Concept of Decision Support Systems (DSS)
Scott Morton [1971] DSS are interactive computer-based systems, which help decision makers utilize data and models to solve unstructured problems [1971] Keen and Scott Morton [1978] Decision support systems couple the intellectual resources of individuals with the capabilities of the computer to improve the quality of decisions. It is a computer-based support system for management decision makers who deal with semi-structured problems. DSS: Content-free expression, i.e., means different things to different people There is no universally accepted definition of DSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

DSS as an Umbrella Term DSS sometimes describes any computerized system used to support decision making Narrow definition: specific technology Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major Characteristics of a DSS
(DSS In Action 1.5: Houston Minerals Case) Initial risk analysis based on decision maker’s definition of the situation using a management science approach Model scrutiny using experience, judgment, and intuition Initial model mathematically correct, but incomplete DSS provided very quick analysis DSS: flexible and responsive to allow managerial intuition and judgment Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Why Use a DSS? Perceived benefits (Udo and Guimaraes [1994])
decision quality improved communication cost reduction increased productivity time savings improved customer and employee satisfaction Highly Correlated Factors degree of competition industry size of company user-friendliness Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major Reasons of Firestone Tire & Rubber Co. [1982]
Unstable economy Increasing competition Increased difficulty in tracking numerous business operations. Existing computer system did not support objectives IS Department could not handle needs or ad hoc inquiries Business analysis functions were not inherent within existing systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Why major corporations started large-scale DSS Hogue and Watson [1983]
Factors Cited by percent Accurate information is needed 67 DSS is viewed as an organizational winner 44 New information is needed 33 Management mandated the DSS 22 Timely information is provided 17 Cost reduction is achieved 6 Plus: End-user computing, especially due to PC’s and inexpensive DSS software development platforms Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.8 Group Support Systems (GSS)
Decisions often made by groups Supports (improves) the work of groups, anytime, anyplace GSS also called Groupware Electronic meeting systems Collaborative systems Group Decision Support Systems (GDSS) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.9 Executive Information (Support) Systems (EIS, ESS)
Strictly for Executive Use! Provide an organizational view Serve the information needs of executives and other managers. Provide an extremely user friendly (user seductive) interface Interface customized to individual decision styles Provide timely and effective tracking and control Provide quick access to detailed information behind text, numbers, or graphics (Drill Down) Filter, compress, and track critical data and information Identify problems (opportunities) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

EIS Started in mid-1980s in large corporations Going global
Becoming affordable to smaller companies Serving many managers as enterprise-wide systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.10 Expert Systems Experts solve complex problems
Experts have specific knowledge and experience Experts are aware of alternatives chances of success benefits and costs Expert systems attempt to mimic human experts An Expert System (ES) is a decision-making and/or problem-solving computer package that reaches or exceeds a level of performance comparable to a human expert in some specialized and usually narrow problem area Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.11 Artificial Neural Networks
Can work with partial, incomplete, or inexact information Artificial Neural Networks (ANN) are mathematical models of the human brain Neural networks learn patterns in data Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Cutting Edge Intelligent Systems
Genetic Algorithms Work in an evolutionary fashion Fuzzy Logic Continuous logic (NOT just True / False) Intelligent Agents Generally in search engines, also in electronic commerce Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.12 Hybrid Support Systems
Combination of MSS Information Technologies Integrated Systems can use strengths of each Goal: successful solution of the managerial problem not the use of a specific tool or technique Some Hybrid Approaches Use each tool independently Use several loosely integrated tools Use several tightly integrated Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Tools can support each other
(e.g., ES adds intelligence to a DSS database) More intelligence is being added (by ES, neural networks, fuzzy logic, intelligent agents) to traditional MSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.13 The Evolution and Attributes of Computerized Decision Aids
Summary of the development of computerized procedures used as aids in decision making (Table 1.2) The support to specific questions provided by DSS (Table 1.3) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson
Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Evolutionary view of computer-based information systems (CBIS)
1. Time Sequence mid-1950s Transaction Processing Systems (TPS) 1960s MIS 1970s Office Automation Systems DSS 1980s DSS Expanded Commercial applications of expert systems Executive Information Systems 1990s Group Support Systems Neural Computing Integrated, hybrid computer systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2. Computer evolved over time 3
2. Computer evolved over time 3. Systemic linkages in how each system processes data into information See Figure 1.3 Relationship among these and other technologies Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Relationship among these and other technologies
Each technology unique Technologies interrelated Each supports some aspects of managerial decision making Evolution and creation of the newer tools help expand the role of information technology for the betterment of management in organizations Interrelationship and coordination among these tools still evolving Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary DSS has many definitions
Complexity of managerial decision making is increasing Computer support for managerial decision making is a must There are several MSS technologies, including hybrids Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Questions for the Opening Vignette
1.Identify the specific decisions cited in the case. Why, in your opinion, do such decisions need computerized support (be specific, provide an answer for each decision)? 2.It is said that the DSS/EIS helped the company to compete on prices, quality, timeliness and service. Visit a competing company such as UPS and explain the importance of this type of competition. 3.Find information about RPS (use the Internet). Find the volume of their business in terms of packages delivered and their financial statements. How successful is this company? 4.It is said that decision support tools empower employees and customers. Can you identify such empowerment in this case? 5.It is said that the TPS is separate from the DSS/EIS in this case. What kind of TPS can you envision in this type of a shipping company and why is it different from the decision support systems? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Exercise Find information on the use of computers to support decisions versus TPS. Each member collects an application in a different industry (e.g., banking, insurance, food, etc.). The group summarizes the findings, points out similarities and differences of the applications. Sources: Companies where students are employed, trade magazines, Internet news groups, and vendor advertisements. Prepare: A class presentation on the findings. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Case Application 1.1: Manufacturing and Marketing of Machine Devices
Part A: The 1995 Crisis - Case Questions 1. Which MSS technique is most likely to be selected by Ms. Chen, and why? (If several techniques can be used, rank them by a descending order of likelihood of success and explain your ranking.) 2. Should MSS tool(s) be used in this situation at all? Part B - Case Questions Examine the following three alternatives: 1. Hire Mr. Morgan; forget MSS. 2. Accelerate the evaluation of MSS technologies; forget Mr. Morgan. 3. Combine alternatives (1) and (2). Discuss the plusses and minuses of each alternative. Which one would you select and why? 4. Which MSS technique is most likely to be selected by Ms. Chen, and why? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.3 Systems 2.3 Systems A SYSTEM is a collection of objects such as people, resources, concepts, and procedures intended to perform an identifiable function or to serve a goal System Levels (Hierarchy): All systems are subsystems interconnected through interfaces A SYSTEM is a collection of objects such as people, resources, concepts, and procedures intended to perform an identifiable function or to serve a goal System Levels (Hierarchy): All systems are subsystems interconnected through interfaces 47 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ 47 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 2: Decision Making, Systems, Modeling, and Support
Conceptual Foundations of Decision Making The Systems Approach How Support is Provided 2.1 Opening Vignette: How to Invest $1,000,000 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.2 Introduction and Definitions
Typical Business Decision Aspects Decision may be made by a group Several, possibly contradictory objectives Hundreds or thousands of alternatives Results can occur in the future Attitudes towards risk “What-if” scenarios Trial-and-error experimentation with the real system: may result in a loss Experimentation with the real system can only be done once Changes in the environment can occur continuously Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

What methodologies can be applied?
How are decisions made??? What methodologies can be applied? What is the role of information systems in supporting decision making? DSS Decision Support Systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Decision Making Decision Making: a process of choosing among alternative courses of action for the purpose of attaining a goal or goals Managerial Decision Making is synonymous with the whole process of management (Simon [1977]) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Decision Making versus Problem Solving
Simon’s 4 Phases of Decision Making 1. Intelligence 2. Design 3. Choice 4. Implementation Decision making and problem solving are interchangeable Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.3 Systems A SYSTEM is a collection of objects such as people, resources, concepts, and procedures intended to perform an identifiable function or to serve a goal System Levels (Hierarchy): All systems are subsystems interconnected through interfaces Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Structure of a System
Three Distinct Parts of Systems (Figure 2.1) Inputs Processes Outputs Systems Are surrounded by an environment Frequently include a feedback mechanism A human, the decision maker, is usually considered part of the system Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

System Environment Output(s) Input(s) Processes Feedback Boundary

Inputs are elements that enter the system
Processes convert or transform the inputs into outputs Outputs describe the finished products or the consequences of being in the system Feedback is the flow of information from the output to the decision maker, who may modify the inputs or the processes (closed loop) The Environment contains the elements that lie outside but impact the system's performance Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

How to Identify the Environment?
Answer Two Questions (Churchman [1975]) 1. Does the element matter relative to the system's goals? [YES] 2. Is it possible for the decision maker to significantly manipulate this element? [NO] Environmental Elements Can Be Social Political Legal Physical Economical Often Other Systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Boundary Separates a System From Its Environment
Boundaries may be physical or nonphysical (by definition of scope or time frame) Information System Boundaries are Usually Directly Defined! Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Closed and Open Systems
Defining manageable boundaries is closing the system A Closed System is totally independent of other systems and subsystems An Open System is very dependent on its environment

An Information System Collects, processes, stores, analyzes, and disseminates information for a specific purpose Is often at the heart of many organizations Accepts inputs and processes data to provide information to decision makers and helps decision makers communicate their results Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

System Effectiveness and Efficiency
Two Major Classes of Performance Measurement Effectiveness is the degree to which goals are achieved Doing the right thing! Efficiency is a measure of the use of inputs (or resources) to achieve outputs Doing the thing right! MSS emphasize effectiveness Often: several non-quantifiable, conflicting goals Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.4 Models Major Component of DSS
Use Models instead of experimenting on the real system A model is a simplified representation or abstraction of reality. Reality is generally too complex to copy exactly Much of the complexity is actually irrelevant in problem solving Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Degrees of Model Abstraction
(Least to Most) Iconic (Scale) Model: Physical replica of a system Analog Model behaves like the real system but does not look like it (symbolic representation) Mathematical (Quantitative) Models use mathematical relationships to represent complexity Used in most DSS analyses Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

An MSS employs models because
Benefits of Models An MSS employs models because 1. Time compression 2. Easy model manipulation 3. Low cost of construction 4. Low cost of execution (especially that of errors) 5. Can model risk and uncertainty 6. Can model large and extremely complex systems with possibly infinite solutions 7. Enhance and reinforce learning, and enhance training. Computer graphics advances: more iconic and analog models (visual simulation) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.5 The Modeling Process-- A Preview
Example: How Much to Order for the Ma-Pa Grocery? The Owners: Bob and Jan The Question: How much bread to stock each day? Several Solution Approaches Trial-and-Error Simulation Optimization Heuristics Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Decision-Making Process
Systematic Decision-Making Process (Simon [1977]) Intelligence Design Choice Implementation (See Figure 2.2) Modeling is Essential to the Process Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Intelligence phase Design phase Choice phase Implementation phase
Reality is examined The problem is identified and defined Design phase Representative model is constructed The model is validated and evaluation criteria are set Choice phase Includes a proposed solution to the model If reasonable, move on to the Implementation phase Solution to the original problem Failure: Return to the modeling process Often Backtrack / Cycle Throughout the Process Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.6 The Intelligence Phase
Scan the environment to identify problem situations or opportunities Find the Problem Identify organizational goals and objectives Determine whether they are being met Explicitly define the problem Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Problem Classification
According to the Degree of Structuredness Programmed versus Nonprogrammed Problems Simon [1977]) Nonprogrammed Programmed Problems Problems

Problem Ownership Outcome: Problem Statement
Problem Decomposition: Divide a complex problem into (easier to solve) subproblems Sometimes called Chunking - (Salami Approach) Some seemingly poorly structured problems may have some highly structured subproblems Problem Ownership Outcome: Problem Statement Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.7 The Design Phase Generating, developing, and analyzing possible courses of action Includes Understanding the problem Testing solutions for feasibility A model is constructed, tested, and validated Modeling Conceptualization of the problem Abstraction to quantitative and/or qualitative forms Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Mathematical Model Identify Variables
Establish Equations describing their Relationships Simplifications through Assumptions Balance Model Simplification and the Accurate Representation of Reality Modeling: An Art and Science Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Quantitative Modeling Topics
Model Components Model Structure Selection of a Principle of Choice (Criteria for Evaluation) Developing (Generating) Alternatives Predicting Outcomes Measuring Outcomes Scenarios Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Components of Quantitative Models
(Figure 2.3) Decision Variables Uncontrollable Variables (and/or Parameters) Result (Outcome) Variables Mathematical Relationships or Symbolic or Qualitative Relationships Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Results of Decisions are Determined by the
Uncontrollable Factors Relationships among Variables Result Variables Reflect the level of effectiveness of the system Dependent variables Examples - Table 2.2 Decision Variables Describe alternative courses of action The decision maker controls them Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Uncontrollable Variables or Parameters
Factors that affect the result variables Not under the control of the decision maker Generally part of the environment Some constrain the decision maker and are called constraints Examples - Table 2.2 Intermediate Result Variables Reflect intermediate outcomes

The Structure of Quantitative Models
Mathematical expressions (e.g., equations or inequalities) connect the components Simple financial-type model P = R - C Present-value model P = F / (1+i)n Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Product-Mix Linear Programming Model
Example The Product-Mix Linear Programming Model MBI Corporation Decision: How many computers to build next month? Two types of computers Labor limit Materials limit Marketing lower limits Constraint CC7 CC8 Rel Limit Labor (days) <= 200,000 / mo Materials $ 10,000 15,000 <= 8,000,000/mo Units 1 >= 100 Units 1 >= 200 Profit $ 8,000 12,000 Max Objective: Maximize Total Profit / Month Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Linear Programming Model
(DSS In Focus 2.1) Components Decision variables Result variable Uncontrollable variables (constraints) Solution X1 = X2 = 200 Profit = $5,066,667 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Principle of Choice
What criteria to use? Best solution? Good enough solution? Selection of a Principle of Choice A decision regarding the acceptability of a solution approach Normative Descriptive Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Normative Models The chosen alternative is demonstrably the best of all Optimization process Normative decision theory is based on rational decision makers Humans are economic beings whose objective is to maximize the attainment of goals; that is, the decision maker is rational In a given decision situation, all viable alternative courses of action and their consequences, or at least the probability and the values of the consequences, are known Decision makers have an order or preference that enables them to rank the desirability of all consequences of the analysis Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Suboptimization Narrow the boundaries of a system
Consider a part of a complete system Leads to (possibly very good, but) non-optimal solutions Viable method Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Descriptive Models Describe things as they are, or as they are believed to be Extremely useful in DSS for evaluating the consequences of decisions and scenarios No guarantee a solution is optimal Often a solution will be "good enough” Simulation: Well-known descriptive modeling technique Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Satisficing (Good Enough)
Most human decision makers will settle for a good enough solution There is a tradeoff between the time and cost of searching for an optimum versus the value of obtaining one A good enough or satisficing solution may be found if a certain goal level is attained (Simon [1977]) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Why Satisfice? Bounded Rationality (Simon)
Humans have a limited capacity for rational thinking They generally construct and analyze a simplified model Their behavior with respect to the simplified model may be rational But, the rational solution for the simplified model may NOT BE rational in the real-world situation Rationality is bounded not only by limitations on human processing capacities, but also by individual differences Bounded rationality is why many models are descriptive, not normative Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Developing (Generating) Alternatives
In Optimization Models: Automatically by the Model! Not Always So! Issue: When to Stop? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Predicting the Outcome of Each Alternative
Must predict the future outcome of each proposed alternative Consider what the decision maker knows (or believes) about the forecasted results Classify Each Situation as Under Certainty Risk Uncertainty Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Decision Making Under Certainty
Assumes that complete knowledge is available (deterministic environment) Example: U.S. Treasury bill investment Typically for structured problems with short time horizons Sometimes DSS approach is needed for certainty situations Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Decision Making Under Risk (Risk Analysis)
(Probabilistic or stochastic decision situation) Decision maker must consider several possible outcomes for each alternative, each with a given probability of occurrence Long-run probabilities of the occurrences of the given outcomes are assumed known or can be estimated Decision maker can assess the degree of risk associated with each alternative (calculated risk) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Risk Analysis Calculate the expected value of each alternative
Selecting the alternative with the best expected value. Example: Poker game with some cards face up (7 card game - 2 down, 4 up, 1 down) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Decision Making Under Uncertainty
Situations in which several outcomes are possible for each course of action BUT the decision maker does not know, or cannot estimate, the probability of occurrence of the possible outcomes More difficult - insufficient information Modeling involves assessing the decision maker's (and/or the organizational) attitude toward risk Example: Poker game with no cards face up (5 card stud or draw) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Measuring Outcomes Goal attainment Maximize profit Minimize cost
Customer satisfaction level (Minimize number of complaints) Maximize quality or satisfaction ratings (found by surveys) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Scenarios Useful in Simulation What-if analysis

Importance of Scenarios in MSS
Help identify potential opportunities and/or problem areas Provide flexibility in planning Identify leading edges of changes that management should monitor Help validate major assumptions used in modeling Help check the sensitivity of proposed solutions to changes in scenarios Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Possible Scenarios Many, but …
Worst possible (Low demand, High costs) Best possible (High demand, High Revenue, Low Costs) Most likely (Typical or average values) The scenario sets the stage for the analysis Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.8 The Choice Phase Search Approaches
Search, evaluation, and recommending an appropriate solution to the model Specific set of values for the decision variables in a selected alternative The problem is considered solved after the recommended solution to the model is successfully implemented Search Approaches Analytical Techniques Algorithms (Optimization) Blind and Heuristic Search Techniques Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.9 Evaluation: Multiple Goals, Sensitivity Analysis, "What-If," and Goal Seeking
Evaluation (coupled with the search process) leads to a recommended solution Multiple Goals Complex systems have multiple goals Some may conflict Typical quantitative models have a single goal Can transform a multiple-goal problem into a single-goal problem Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Common Methods Utility theory Goal programming
Expression of goals as constraints, using linear programming Point system Computerized models can support multiple goal decision making Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Types of Sensitivity Analyses
Sensitivity Analysis Change inputs or parameters, look at model results Sensitivity analysis checks relationships Types of Sensitivity Analyses Automatic Trial and error Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Trial and Error Change input data and re-solve the problem
Better and better solutions can be discovered How to do? Easy in spreadsheets (Excel) what-if goal seeking Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

What-If Analysis Goal Seeking
Figure SSpreadsheet example of a what-if query for a cash flow problem Goal Seeking Backward solution approach Example: Figure 2.9 Example: What interest rate causes an the net present value of an investment to break even? In a DSS the what-if and the goal-seeking options must be easy to perform Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.10 The Implementation Phase
There is nothing more difficult to carry out, nor more doubtful of success, nor more dangerous to handle, than to initiate a new order of things (Machiavelli [1500s]) *** The Introduction of a Change *** Important Issues Resistance to change Degree of top management support Users’ roles and involvement in system development Users’ training Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.11 How Decisions Are Supported
Specific MSS technologies relationship to the decision making process (see Figure 2.10) Intelligence: DSS, ES, ANN, MIS, Data Mining, OLAP, EIS, GDSS Design and Choice: DSS, ES, GDSS, Management Science, ANN Implementation: DSS, ES, GDSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.12 Human Cognition and Decision Styles
Cognition Theory Cognition: Activities by which an individual resolves differences between an internalized view of the environment and what actually exists in that same environment Ability to perceive and understand information Cognitive models are attempts to explain or understand various human cognitive processes Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Cognitive Style The subjective process through which individuals perceive, organize, and change information during the decision-making process Often determines people's preference for human-machine interface Impacts on preferences for qualitative versus quantitative analysis and preferences for decision-making aids Cognitive style research impacts on the design of management information systems Analytic decision maker Heuristic decision maker (See Table 2.4) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The manner in which decision makers
Decision Styles The manner in which decision makers Think and react to problems Perceive Their cognitive response Their values and beliefs Varies from individual to individual and from situation to situation Decision making is a nonlinear process The manner in which managers make decisions (and the way they interact with other people) describes their decision style There are dozens Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Some Decision Styles Heuristic Analytic Autocratic Democratic
Consultative (with individuals or groups) Combinations and variations For successful decision making support, an MSS must fit the Decision situation Decision style Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Different decision styles require different types of support
The system should be flexible and adaptable to different users have what-if and goal-seeking have graphics have process flexibility An MSS should help decision makers use and develop their own styles, skills, and knowledge Different decision styles require different types of support Major factor: individual or group decision maker Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2.13 The Decision Makers Individuals Groups
Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Individuals May still have conflicting objectives
Decisions may be fully automated

Groups Most major decisions in medium and large organizations are made by groups Conflicting objectives are common Variable size People from different departments People from different organizations The group decision making process can be very complicated Consider Group Support Systems (GSS) Organizational DSS can help in enterprise-wide decision making situations

Summary Managerial decision making is synonymous with the whole process of management Problem solving also refers to opportunity's evaluation A system is a collection of objects such as people, resources, concepts, and procedures intended to perform an identifiable function or to serve a goal DSS deals primarily with open systems A model is a simplified representation or abstraction of reality Models enable fast and inexpensive experimentation with systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary (cont.) Modeling can employ optimization, heuristic, or simulation techniques Decision making involves four major phases: intelligence, design, choice, and implementation What-if and goal seeking are the two most common sensitivity analysis approaches Computers can support all phases of decision making by automating many of the required tasks Human cognitive styles may influence human-machine interaction Human decision styles need to be recognized in designing MSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. Identify the conflicting objectives 2. Identify the uncertainties 3. Identify the alternative courses of action (can they be combined?) 4. What are the possible results of the decision? Why may the results be difficult to predict? 5. What kind of risk is associated with the decision? 6. What were the decision-makers different “attitudes” toward risk? How could this influence the decision? 7. What would you do and why? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Project Interview an individual who was recently involved in making a business decision. Try to identify: 1. The scope of the problem being solved 2. The individuals involved in the decision (explicitly identify the problem owner(s)) 3. Simon’s phases (you may have to ask the individual specific questions such as how he or she identified the problem, etc.) 4. The alternatives (choices) and the decision chosen 5. How the decision was implemented 6. How computers were used or why they were not used Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Produce a detailed report describing an analysis of the above and clearly state how closely the real-world decision making process compares to Simon’s suggested process. Also, clearly identify how computers were used or why they were not used. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Part 2: Decision Support Systems
Decision Support Methodology Technology Components Construction Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 3: Decision Support Systems: An Overview
Capabilities Structure Classifications Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.1 Opening Vignette: Gotaas-Larsen Shipping Corp. (GLSC)
Strategic planning Not a structured decision situation Cargo ship voyage planning DSS: Data and Models Large-scale DSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Opening Vignette Illustrates that for the GLSC,
3.2 DSS Configurations Opening Vignette Illustrates that for the GLSC, the DSS Supports an entire organization Supports several interrelated decisions Is used repeatedly and constantly Has two major components: data and models Utilizes a simulation model Uses both internal and external data Has “what-if” capabilities Uses several quantitative models Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

DSS Definitions Little [1970] “model-based set of procedures for processing data and judgments to assist a manager in his decision making” Assumption: that the system is computer-based and extends the user’s capabilities. Alter [1980] Contrasts DSS with traditional EDP systems (Table 3.1) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Moore and Chang [1980] 1.extendible systems 2.capable of supporting ad hoc data analysis and decision modeling 3.oriented toward future planning 4.used at irregular, unplanned intervals Bonczek et al. [1980] A computer-based system consisting of 1. a language system -- communication between the user and DSS components 2. a knowledge system 3. a problem-processing system--the link between the other two components Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Central Issue in DSS support and improvement of decision making
Keen [1980] DSS apply “to situations where a `final’ system can be developed only through an adaptive process of learning and evolution” Central Issue in DSS support and improvement of decision making Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Working Definition of DSS
A DSS is an interactive, flexible, and adaptable CBIS, specially developed for supporting the solution of a non-structured management problem for improved decision making. It utilizes data, it provides easy user interface, and it allows for the decision maker’s own insights DSS may utilize models, is built by an interactive process (frequently by end-users), supports all the phases of the decision making, and may include a knowledge component Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.4 Characteristics and Capabilities of DSS
DSS (Figure 3.1) 1. Provide support in semi-structured and unstructured situations 2. Support for various managerial levels 3. Support to individuals and groups 4. Support to interdependent and/or sequential decisions 5. Support all phases of the decision-making process 6. Support a variety of decision-making processes and styles Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7. Are adaptive 8. Have user friendly interfaces 9
7. Are adaptive 8. Have user friendly interfaces 9. Goal is to improve the effectiveness of decision making 10. The decision maker controls the decision-making process 11. End-users can build simple systems 12. Utilizes models for analysis 13. Provides access to a variety of data sources, formats, and types Decision makers can make better, more consistent decisions in a timely manner Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.5 DSS Components 1. Data Management Subsystem 2. Model Management Subsystem 3. Knowledge Management Subsystem 4. User Interface Subsystem 5. The User (Figure 3.2) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.6 The Data Management Subsystem
DSS database Database management system Data directory Query facility (Figure 3.3) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

DSS Database Issues Data warehouse Special independent DSS databases
Extraction of data from internal, external and private sources Web browser access of data Multimedia databases Object-oriented databases Commercial database management systems (DBMS) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.7 The Model Management Subsystem
Mirrors the database management subsystem (Figure 3.4) Model Management Issues Model level: Strategic, managerial (tactical) and operational Modeling languages Lack of standard MBMS activities. WHY? Use of AI and Fuzzy logic in MBMS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.8 The Knowledge Management Subsystem
Provides expertise in solving complex unstructured and semi-structured problems Expertise provided by an expert system or other intelligent system Advanced DSS have a knowledge management component Leads to intelligent DSS Example: Data mining Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.9 The User Interface (Dialog) Subsystem
Includes all communication between a user and the MSS To most users, the user interface is the system Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.10 The User Different usage patterns for the user, the manager, or the decision maker Managers Staff specialists Intermediary: 1.Staff assistant 2.Expert tool user 3.Business (system) analyst 4.Group DSS Facilitator Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.11 DSS Hardware Major Hardware Options
Evolved with computer hardware and software technologies Major Hardware Options organization’s mainframe computer minicomputer workstation personal computer client/server system Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.12 Distinguishing DSS from Management Science and MIS
DSS is a problem solving tool and is frequently used to address ad hoc and unexpected problems Different than MIS DSS evolve as they develop Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3.13 DSS Classifications Alter’s Output Classification [1980]
Degree of action implication of system outputs (supporting decision) (Table 3.3) Holsapple and Whinston’s Classification 1.Text-oriented DSS 2.Database-oriented DSS 3.Spreadsheet-oriented DSS 4.Solver-oriented DSS 5.Rule-oriented DSS 6.Compound DSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Other Classifications
Institutional DSS vs. Ad Hoc DSS Institutional DSS deals with decisions of a recurring nature Ad Hoc DSS deals with specific problems that are usually neither anticipated nor recurring Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Other Classifications (cont’d.)
Degree of Nonprocedurality (Bonczek, et al. [1980]) Personal, Group, and Organizational Support (Hackathorn and Keen [1981]) Individual versus Group DSS Custom-made versus Ready-made Systems

Summary Fundamentals of DSS GLSC Case Components of DSS
Major Capabilities of the DSS Components Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Exercises 1. Susan Lopez was promoted to be a director of the transportation department in a medium-size university. ... Susan’s major job is to schedule vehicles for employees, and to schedule the maintenance and repair of the vehicles. Possibility of using a DSS to improve this situation. Susan has a Pentium PC, and Microsoft Office, but she is using the computer only as a word processor. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Projects 1. Design and implement a DSS for either the problem described in Exercise 1 above or a similar, real-world one. Clearly identify data sources and model types, and document the problems your group encountered while developing the DSS. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 4 Data Management: Warehousing, Access and Visualization
MSS foundation New concepts Object-oriented databases Intelligent databases Data warehouse Online analytical processing Multidimensionality Data mining Internet / Intranet / Web Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2-3 million data records are processed monthly
4.1 Opening Vignette: Data Warehousing and DSS at Group Health Cooperative 2-3 million data records are processed monthly How to manage? How to use for decision support? How to hold down costs? How to improve customer service? How to utilize resource effectively? How to improve service quality? Answers Develop a comprehensive database (data warehouse) and DSS approach Very effective Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4.2 Data Warehousing, Access, Analysis and Visualization
What to do with all the data that organizations collect, store and use? Information overload! Solution Data warehousing Data access Data mining Online analytical processing (OLAP) Data visualization Data sources Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4.3 The Nature and Sources of Data
Data: Raw Information: Data organized to convey meaning Knowledge: Data items organized and processed to convey understanding, experience, accumulated learning, and expertise DSS Data Items Documents Pictures Maps Sound Animation Video Can be hard or soft Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Data Sources Internal External Personal

4.4 Data Collection and Data Problems
Summarized in Table 4.1 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4.5 The Internet and Commercial Database Services
For External Data The Internet: Major supplier of external data Commercial Data “Banks”: Sell access to specialized databases Can add external data to the MSS in a timely manner and at a reasonable cost Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Internet/Web and Corporate Databases and Systems
Use Web Browsers to Access vital information by employees and customers Implement executive information systems Implement group support systems (GSS) Database management systems provide data in HTML Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4.6 Database Management Systems in DSS
DBMS: Software program for entering (or adding) information into a database; updating, deleting, manipulating, storing, and retrieving information A DBMS combined with a modeling language is a typical system development pair, used in constructing DSS or MSS DBMS are designed to handle large amounts of information Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4.7 Database Organization and Structure
Relational Databases Hierarchical Databases Network Databases Object-oriented Databases Multimedia-based Databases Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4.8 Data Warehousing Physical separation of operational and decision support environments Purpose: to establish a data repository making operational data accessible Transforms operational data to relational form Only data needed for decision support come from the TPS Data are transformed and integrated into a consistent structure Data warehousing (or information warehousing): a solution to the data access problem End users perform ad hoc query, reporting analysis and visualization Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Data Warehousing Benefits
Increase in knowledge worker productivity Supports all decision makers’ data requirements Provide ready access to critical data Insulates operation databases from ad hoc processing Provides high-level summary information Provides drill down capabilities Yields Improved business knowledge Competitive advantage Enhances customer service and satisfaction Facilitates decision making Help streamline business processes Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Data Warehouse Architecture and Process
Two-tier architecture Three-tier architecture (Figure 4.3) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Data Warehouse Components
Large physical database Logical data warehouse Data mart Decision support systems (DSS) and executive information system (EIS)

DW Suitability For organizations where Data are in different systems
Information-based approach to management in use Large, diverse customer base Same data have different representations in different systems Highly technical, messy data formats Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Characteristics of Data Warehousing
1. Data organized by detailed subject with information relevant for decision support 2.Integrated data 3.Time-variant data 4.Non-volatile data Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4.9 OLAP: Data Access and Mining, Querying and Analysis
Online Analytical processing (OLAP) DSS and EIS computing done by end-users in online systems Versus online transaction processing (OLTP) OLAP Activities Generating queries Requesting ad hoc reports Conducting statistical analyses Building multimedia applications Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

OLAP uses the data warehouse and a set of tools, usually with multidimensional capabilities
Query tools Spreadsheets Data mining tools Data visualization tools Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Using SQL for Querying SQL (Structured Query Language) Data language English-like, nonprocedural, very user friendly language Free format Example: SELECT Name, Salary FROM Employees WHERE Salary >2000 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Data Mining For Knowledge discovery in databases Knowledge extraction
Data archeology Data exploration Data pattern processing Data dredging Information harvesting Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major Data Mining Characteristics and Objectives
Data are often buried deep Client/server architecture Sophisticated new tools--including advanced visualization tools--help to remove the information “ore” Massaging and synchronizing data Usefulness of “soft” data End-user minor is empowered by “data drills” and other power query tools with little or no programming skills Often involves finding unexpected results Tools are easily combined with spreadsheets etc. Parallel processing for data mining Example in Figure 4.4 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Data Mining Application Areas
Marketing Banking: Retailing and sales Manufacturing and production Brokerage and securities trading Insurance Computer hardware and software Government and defense Airlines Health care Broadcasting Law Enforcement Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4.10 Data Visualization and Multidimensionality
Data Visualization Technologies Digital images Geographic information systems Graphical user interfaces Multidimensions Tables and graphs Virtual reality Presentations Animation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Multidimensionality 3-D + Spreadsheets
Data can be organized the way managers like to see them, rather than the way that the system analysts do Different presentations of the same data can be arranged easily and quickly Dimensions: products, salespeople, market segments, business units, geographical locations, distribution channels, country, or industry Measures: money, sales volume, head count, inventory profit, actual versus forecasted Time: daily, weekly, monthly, quarterly, or yearly Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Multidimensionality Limitations
Extra storage requirements Higher cost Extra system resource and time consumption More complex interfaces and maintenance Multidimensionality is especially popular in executive information and support systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4.11 Intelligent Databases and Data Mining
Developing MSS applications requires access to databases AI technologies (ES, ANN) to assist database management Integration Example in Figure 4.5 Link ES to large databases Example: query optimizer Natural language interfaces Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Intelligent Data Mining
Use intelligent search to discover information within data warehouses that queries and reports cannot effectively reveal Find patterns in the data and infer rules from them Use patterns and rules to guide decision-making and forecasting Five common types of information that can be yielded by data mining: 1) association, 2) sequences, 3) classifications, 4) clusters, and 5) forecasting Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Main Tools Used in Intelligent Data Mining
Case-based Reasoning Neural Computing Intelligent Agents Other Tools decision trees rule induction data visualization Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary Data for decision making come from internal and external sources The database management system is one of the major components of most management support systems Familiarity with the latest developments is critical Data contain a gold mine of information if they can dig it out Organizations are warehousing and mining data Multidimensional analysis tools and new enterprise-wide system architectures are useful OLAP tools are also useful Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary (cont’d.) Object-oriented approach to systems analysis, design, and implementation may prove useful New data formats for multimedia DBMS Internet and intranets via Web browser interfaces for DBMS access Built-in artificial intelligence methods in DBMS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.Use the Holsapple and Whinston classification system and identify the categories of the DSS applications in the case. 2.Identify the driving forces that led to the creation of the data warehouse. 3.Comment on the sources of data. 4.Identify the decisions supported by the data warehouse. 5.Read the article: Braley, D. (1996, February). “System Purchases Support Vendors’ Visions.” Health Management Technology. Vol. 17. No Compare the evolution and developments described in the article to those in the Opening Vignette. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Exercise 4 The U.S. government spends millions of dollars gathering data on its population every 10 years (plus some mid-decade corrections). Census data are critical in determining the representation of each state in the House of Representatives and the number of Electoral College votes to which each state is entitled for Presidential elections. More importantly, census data provide information about U.S. markets. The demographics indicate family and gender make up, incomes, education level, etc. for the states, metropolitan statistical areas (MSA), and counties. Such data are available from various sources including books, disk, CD-ROM and the World Wide Web (see Internet Exercise 5). In this exercise, we take a real-world view of external, but readily available data. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. Find an electronic source of standard census data files for the states and MSAs. 2. Access the data and examine the file structures. Do the contents and organization of each make sense? Why or why not? If not, suggest improvements. 3. Load the state P1 data population table into a spreadsheet file (Excel if possible) and into a database file (ACCESS if possible). How difficult was this? How could this have been made easier? Don’t forget to delete the comments and U.S. totals (if present) at the top, for later use. Note that Washington, DC is listed as well. Print the table. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4. Using the state P1 population data, sort the data, based on population size. What are the five most populated states, and the five least populated states? Which five states have the largest and smallest population densities? Which state has the most males and which state has the most females? Which three states have the most people living on farms, and which state has the least lonely people? Which file type (spreadsheet or database) did you use and why? What features made it easy to do these analyses? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5. Load the State Basic Table P6 (Household Income) into a spreadsheet or database file. Which five states have the most people earning $100,000 or more per year? Which five states have the highest percentages of people earning $100,000 or more per year? Combining these data with data from Table P1, which five states have the most people per square mile earning $100,000 or more per year? Which file type (spreadsheet or database) did you use and why? What features made it easy to do these analyses? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6. Data warehousing and data mining are used to combine data and identify patterns. Using data (load and save them into spreadsheet or database files) from files: a) P1 Population b) P3 Persons by Age c) P4 Households by Size d) P6 Household Income e) P8 Other Income Measures f) P9 Level of Education. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Synthesize these tables into a usable set and determine if there are any relationships at the state level between: Population per square mile and education Income and age Household size and education Can you think of any other relationships to explore? Do so. What made this task difficult or easy? Explain. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7. Examine the MSA data tables and see if any of the relationships found for the state data above hold. 8. How does your MSA (or one closest to where you live) compare to your state’s census profile and that of the entire United States? How did you determine this? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Exercise One of the most difficult tasks in any large city is traffic law enforcement. According to PCWeek, Nov. 13, 1993, p. 63, a solution to the problem can be found in a client/server-based data warehousing system. Read the article and then visit your local traffic enforcement agency. a) Review the current information system. b) Identify problems in the existing system. c) Explain how a system like the one described in the PCWeek story can help your local agency. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Case Application 4.1: Data Warehousing at the Canadian Imperial Bank of Commerce
2nd largest bank in Canada One of the top 10 banks in North America Decision support applications supported by a data warehouse Data warehouse provides diverse decision-making support Analyses supported include customer traffic patterns at branches Data warehouse evolved over time The secret is to hold the data at the event level and summarize them to the level of granularity appropriate for specific queries Statistical modeling and consulting Supports EIS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Benefits of the Data Warehouse Structure
Data integrity Consistency across time lines High efficiency Low operating costs Can store data at different levels of summarization Can give customers quick turnaround Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Appendix W4-A: Object-oriented Systems Analysis, Design and Programming
W4-A.1 Introduction to the Object-Oriented Approach Objects are created and manipulated, rather than ‘items’ in programs Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Objects have Certain features or attributes Exhibit certain behaviors
Interact

Objects can be grouped and classified, like real-world objects
Specific objects (a specific person) have certain attributes by being in a class (employees, citizens, customers, etc.) An object knows what it is and what it can do Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

W4-A.2 Object Think The system analyst focuses on the user requirements that lead directly to the definition and subsequent development of objects Objects have characteristics that they exhibit, and inherit characteristics directly from their class, and from their “parents” Example: University library Class of objects called “books” Class of objects called “borrowers See Figure 4-A.1 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

W4-A.3 Important Object-Oriented Approach Terminology
Object: a thing - a specific instance An object knows what it is and what it can do Class: a type of thing, and all specific things that fit the general definition of the class belong to the class Like a data entity type when modeling data A class is the general category and an object is a specific instance Attributes of a Class: The attributes that all of the objects in the class share define a class of objects The attribute values are part of what an object knows when using the object think approach Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Association Relationships: Objects may be related to other objects These are similar to relationships in a data model. A relationship is an association based on the context in which we view objects, e.g., a natural association These relationships have names, can be optional or mandatory, and have cardinality Whole-part Relationships: Stronger than association relationships Strong relationships between an object and other objects that are its parts Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Methods or Services of a Class:
A method is something the object knows how to do Service is something that the object knows how to do for a requester Standard services: all objects know how to do Complex services: custom designed for the class of objects Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Encapsulation or Information Hiding:
Encapsulation means packaging several items together into one unit Packaging both the attributes and services of the class together so that the object knows things (attributes) and how to do things (services). We hide the internal structure of an object from the environment Message Sending: End users can send messages to objects to perform a service Objects can send messages to other objects Messages may also be triggered temporally Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Polymorphism (Multiple forms):
Different kinds of related items Inheritance, Classification Hierarchies and Reuse: Classification hierarchies allow classes of objects to inherit attributes from larger classes Allows for object reuse Pre-defined classes of interface objects Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

W4-A.4 The Object-Oriented System Development Cycle
1. Object-Oriented Analysis 2. Object-Oriented Design 3. Object-Oriented Implementation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Object-Oriented Analysis
Define system requirements through scenarios or use cases Then, build an object model with the capability to satisfy the requirements Output: requirements model Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Object-Oriented Design
The requirements model created in the analysis phase is enhanced in the design phase. Sometimes more attributes and services are added Interface objects are added Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Object-Oriented Implementation
Usable system is developed Use object-oriented programming languages If needed, provide links to a separate database management system Object-Oriented CASE Tools New capabilities are being developed Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

W4-A.5 Object-Oriented Programming Languages
Pure Smalltalk Hybrid: C++ Also: Object-oriented Cobol Ada Objective C Object Pascal Actor Eiffel and more Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

W4.A-6 Object Oriented Database Management Systems
The database system must 1. Support complex objects 2. Support object identity 3. Allow objects to be encapsulated 4. Support types or classes 5. Support inheritance 6. Avoid premature binding 7. Be computationally complete Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8. Be extensible 9. Be able to remember data locations 10
8. Be extensible 9. Be able to remember data locations 10. Be able to manage large databases 11. Accept concurrent users 12. Be able to recover from hardware/software failures 13. Support data query in a simple way Norman [1996] Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Strengths and Weaknesses of an Object-Oriented Database
1. Data Modeling 2. Nonhomogeneous data 3. Variable length and long strings 4. Complex objects 5. Version control 6. Schema evolution 7. Equivalent objects 8. Long transactions 9. User Benefits Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Weaknesses 1 .New problem solving approach 2. Lack of a common data model with a strong theoretical foundation 3. Limited success stories Norman [1996] Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Companies Versant Object Technology Corp. (Menlo Park, CA - Versant ODBMS) KE Software Inc. (Vancouver, BC try the demo) O2 Technology (Palo Alto, CA try the demo) Object Design Inc. (Burlington, MA) Hewlett-Packard Co. (OpenODB) Itasca Systems Inc. (Itasca Distributed Management System) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Object Design Inc. (ObjectStore) Objectivity Inc. (Objectivity/DB)
Ontos Inc. (Ontos DB) Servio Corp. (Gemstone) UniSQL Inc. (UniSQL/X, UniSQL/M) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

W4-A.7 Commercial Applications of the Object-Oriented Approach
From enterprise information systems, maintenance management and financial applications to Geographical Information Systems EDS’s Maintenance Management System (MMS) Time Warner Communications: a variety of business applications Sprint Corp. developed an object-oriented order-entry sales system to speed the providing of phone service Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

W4-A.8 Summary and Conclusions
Many demonstrated successes But a paradigm shift is required The entire organization must adopt object-think Revolutionary change Transition may be bumpy Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 5: Modeling and Analysis
Major component the model base and its management Caution Familiarity with major ideas Basic concepts and definitions Tool--the influence diagram Modeling directly in spreadsheets Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Structure of some successful models and methodologies
decision analysis decision trees optimization heuristic programming simulation New developments in modeling tools and techniques Important issues in model base management. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Saved SSI over $75 million each year
5.1 Opening Vignette: Siemens Solar Industries Saves Millions by Simulation Clean room contamination-control technology No experience Use simulation: a virtual laboratory Major benefit: knowledge and insight Improved the manufacturing process Saved SSI over $75 million each year Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.2 Modeling for MSS Modeling Key element in most DSS
A necessity in a model-based DSS Frazee Paint Company (Appendix A Three model types 1. Statistical model (regression analysis) 2. Financial model 3. Optimization model Several models Standard Custom made Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major Modeling Issues Problem identification Environmental analysis
Variable identification Forecasting Multiple model use Model categories (or selection) [Table 5.1] Model management Knowledge-based modeling Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.3 Static and Dynamic Models
Static Analysis Single snapshot Dynamic Analysis Dynamic models Evaluate scenarios that change over time Are time dependent Show trends and patterns over time Extended static models Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.4 Treating Certainty, Uncertainty, and Risk
Certainty Models Uncertainty Risk Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.5 Influence Diagrams Graphical representations of a model to assist in model design, development and understanding Provide visual communication to the model builder or development team Serve as a framework for expressing the MSS model relationships Rectangle = a decision variable Circle = uncontrollable or intermediate variable Oval = result (outcome) variable: intermediate or final Variables connected with arrows Example in Figure 5.1 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.6 MSS Modeling in Spreadsheets
(Electronic) spreadsheet: most popular end-user modeling tool Powerful financial, statistical, mathematical, logical, date/time, string functions External add-in functions and solvers Important for analysis, planning, modeling Programmability (macros) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Figure 5.2: Simple loan calculation model (static) Figure 5.3: Dynamic
What-if analysis Goal seeking Seamless integration Microsoft Excel Lotus 1-2-3 Figure 5.2: Simple loan calculation model (static) Figure 5.3: Dynamic Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.7 Decision Analysis of Few Alternatives (Decision Tables and Trees)
Single Goal Situations Decision tables Decision trees Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Decision Tables Investment Example
One goal: Maximize the yield after one year Yield depends on the status of the economy (the state of nature) Solid growth Stagnation Inflation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. If there is solid growth in the economy, bonds will yield 12 percent; stocks, 15 percent; and time deposits, 6.5 percent 2. If stagnation prevails, bonds will yield 6 percent; stocks, 3 percent; and time deposits, 6.5 percent 3. If inflation prevails, bonds will yield 3 percent; stocks will bring a loss of 2 percent; and time deposits will yield 6.5 percent Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

View problem as a two-person game
Payoff Table 5.2 Decision variables (the alternatives) Uncontrollable variables (the states of the economy) Result variables (the projected yield) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Treating Uncertainty Optimistic approach Pessimistic approach

Treating Risk Use known probabilities (Table 5.3)
Risk analysis: Compute expected values Can be dangerous Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Other Methods of Treating Risk
Decision Trees Other Methods of Treating Risk Simulation Certainty factors Fuzzy logic. Multiple Goals Table 5.4: Yield, safety, and liquidity Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.8 Optimization via Mathematical Programming
Linear programming (LP) used extensively in DSS Mathematical Programming Family of tools to solve managerial problems in allocating scarce resources among various activities to optimize a measurable goal Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

LP Allocation Problem Characteristics
1. Limited quantity of economic resources 2. Resources are used in the production of products or services. 3. Two or more ways (solutions, programs) to use the resources 4. Each activity (product or service) yields a return in terms of the goal 5. Allocation is usually restricted by constraints Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

LP Allocation Model Rational Economic Assumptions
1. Returns from different allocations can be compared in a common unit 2. Independent returns 3. Total return is the sum of different activities’ returns 4. All data are known with certainty 5. The resources are to be used in the most economical manner Optimal solution: the best, found algorithmically Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Linear Programming Decision variables Objective function
Objective function coefficients Constraints Capacities Input-output (technology) coefficients Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.9 Heuristic Programming
Cuts the search Gets satisfactory solutions more quickly and less expensively Finds rules to solve complex problems Heuristic programming finds feasible and "good enough" solutions to some complex problems Heuristics can be Quantitative Qualitative (in ES) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

When to Use Heuristics 1. Inexact or limited input data
2. Complex reality 3. Reliable, exact algorithm not available 4. Simulation computation time too excessive 5. To improve the efficiency of optimization 6. To solve complex problems 7. For symbolic processing 8. For solving when quick decisions are to be made Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Advantages of Heuristics
1. Simple to understand: easier to implement and explain 2. Help train people to be creative 3. Save formulation time 4. Save programming and storage requirements on the computers 5. Save computer running time (speed) 6. Frequently produce multiple acceptable solutions 7. Usually possible to develop a measure of solution quality 8. Can incorporate intelligent search 9. Can solve very complex models Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Limitations of Heuristics
1. Cannot guarantee an optimal solution 2. There may be too many exceptions 3. Sequential decision choices can fail to anticipate future consequences of each choice 4. Interdependencies of subsystems can influence the whole system Heuristics successfully applied to vehicle routing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.10 Simulation A technique for conducting experiments with a computer on a model of a management system Frequently used DSS tool Major Characteristics of Simulation Simulation imitates reality and capture its richness Simulation is a technique for conducting experiments Simulation is a descriptive not normative tool Simulation is often used to solve very complex, risky problems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Advantages of Simulation
1. Theory is straightforward 2. Time compression 3. Descriptive, not normative 4. Intimate knowledge of the problem forces the MSS builder to interface with the manager 5. The model is built from the manager's perspective 6. No generalized understanding is required of the manager. Each model component represents a real problem component Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7. Wide variation in problem types
8. Can experiment with different variables 9. Allows for real-life problem complexities 10. Easy to obtain many performance measures directly 11. Frequently the only DSS modeling tool for handling nonstructured problems 12. Monte Carlo add-in spreadsheet packages Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Limitations of Simulation
1. Cannot guarantee an optimal solution 2. Slow and costly construction process 3. Cannot transfer solutions and inferences to solve other problems 4. So easy to sell to managers, may miss analytical solutions 5. Software is not so user friendly Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Simulation Methodology
Set up a model of a real system and conduct repetitive experiments 1. Problem Definition 2. Construction of the Simulation Model 3. Testing and Validating the Model 4. Design of the Experiments 5. Conducting the Experiments 6. Evaluating the Results 7. Implementation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Simulation Types Probabilistic Simulation Discrete distributions
Continuous distributions Probabilistic simulation via Monte Carlo technique Time Dependent versus Time Independent Simulation Simulation Software Visual Simulation Object-oriented Simulation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.11 Multidimensional Modeling
From a spreadsheet and analysis perspective 2-D to 3-D to multiple-D Multidimensional modeling tools: 16-D + Multidimensional modeling: four views of the same data (Figure 5.5) Tool can compare, rotate, and "slice and dice" corporate data across different management viewpoints Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.12 Visual Spreadsheets User can visualize the models and formulas using influence diagrams Not cells, but symbolic elements (Figure 5.6) English-like modeling Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.13 Financial and Planning Modeling
Special tools to build usable DSS rapidly, effectively, and efficiently The models are algebraically oriented Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Definition and Background of Planning Modeling
Fourth generation programming languages Models written in an English-like syntax Models are self-documenting Model steps are nonprocedural Examples Visual IFPS / Plus ENCORE Plus! SORITEC Some are embedded in EIS and OLAP tools Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

List of typical applications of planning models (DSS In Focus 5.6).
Major differences between financial modeling-based tools and DBMS-based tools (Table 5.6) Visual IFPS/Plus model from the influence diagram model in Figure 5.1 (Figure 5.7) List of typical applications of planning models (DSS In Focus 5.6). Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.14 Visual Modeling and Simulation
Visual interactive modeling (VIM) (DSS In Action 5.8) Also called: Visual interactive problem solving Visual interactive modeling Visual interactive simulation Use computer graphics to present the impact of different management decisions. Users perform sensitivity analysis Static or a dynamic (animation) systems (Example: Figure 5.8) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Visual Interactive Simulation (VIS)
Decision makers interact with the simulated model and watch the results over time Visual Interactive Models and DSS VIM (Case Application W5.1 on the Book’s Web Site) Queuing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.15 Ready-made Quantitative Software Packages
Preprogrammed models can expedite the programming time of the DSS builder Some models are building blocks of other quantitative models Statistical Packages Management Science Packages Financial Modeling Other Ready-Made Specific DSS (Applications) including spreadsheet add-ins Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5.16 Model Base Management MBMS: capabilities similar to that of DBMS
But, there are no comprehensive model base management packages Each organization uses models somewhat differently There are many model classes Some MBMS capabilities require expertise and reasoning Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Desirable Capabilities of MBMS
Control Flexibility Feedback Interface Redundancy Reduction Increased Consistency Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

MBMS Design Must Allow the DSS User to
1. Access and retrieve existing models. 2. Exercise and manipulate existing models 3. Store existing models 4. Maintain existing models 5. Construct new models with reasonable effort Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Object-oriented Model Base and Its Management
Modeling Languages Relational MBMS Object-oriented Model Base and Its Management Models for Database and MIS Design and their Management Enterprise and Business Process Reengineering Modeling and Model Management Systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

SUMMARY Models play a major role in DSS
Models can be static or dynamic. Analysis is under assumed certainty, risk, or uncertainty Influence diagrams Electronic spreadsheets Decision tables and decision trees Optimization tool: mathematical programming Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

SUMMARY (cont’d.) Linear programming: economic-base
Heuristic programming Simulation Simulation can deal with more complex situations Expert Choice Forecasting methods Multidimensional modeling Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

SUMMARY (cont’d.) Built-in quantitative models (financial, statistical) Special financial modeling languages Visual interactive modeling Visual interactive simulation (VIS) Spreadsheet modeling and results in influence diagrams MBMS are like DBMS AI techniques in MBMS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.Explain how simulation was used to evaluate a nonexistent system. 2.What was learned, from using the simulation model, about running the clean room? 3.How could the time compression capability of simulation help in this situation? 4.How did the simulation results help the SSI engineers learn about their decision making problem? Were they able to focus better on the structure of the real system? How did this save development and operating costs of the real clean room? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Debate Some people believe that managers do not need to know the internal structure of the model and the technical aspects of modeling. “It is like the telephone or the elevator, you just use it.” Others claim that this is not the case and the opposite is true. Debate the issue. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Class Exercises 3. Everyone in the class Write your weight, height and gender on a piece of paper (no names please!). Create a regression (causal) model for height versus weight for the whole class, and one for each gender. If possible, use a statistical package and a spreadsheet and compare their ease of use. Produce a scatterplot of the three sets of data. Do the relationships appear linear? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

How accurate were the models (R2)?
Does weight cause height; does height cause weight; or does neither really cause the other? Explain? How can a regression model like this be used in building design; diet / nutrition selection? in a longitudinal study (say over 50 years) in determining whether students are getting heavier and not taller, or vice-versa? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6. DSS generators are English-like and have a variety of analysis capabilities.
a. Identify the purpose and the analysis capabilities of the following IFPS program: MODEL FIRST COLUMNS 1-5 INVESTMENT = LAND + BUILDING RETURN = SALES - COSTS PRESENT VALUE = NPVC(RETURN, DISCOUNT RATE, INVESTMENT) INTERNAL RATE OF RETURN = IRR(RETURN, INVESTMENT) \ INPUT DATA LAND = 200, 0 BUILDING = 100, 150, 0 SALES = 500, PREVIOUS + 100 COSTS = SUM(MATERIALS THRU LABOR) MATERIALS = * SALES OVERHEAD = .10 * SALES LABOR = * SALES DISCOUNT RATE = 0.20, PREVIOUS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

What do these statements do to this new model?
b. Change sales to be under assumed risk, that is, replace the SALES line and insert a line following it: 9 SALES = NORRANDR(EXPECTED SALES, EXPECTED SALES/10) EXPECTED SALES = 500, PREVIOUS + 100 and use MONTE CARLO 200 COLUMNS 5 HIST PRESENT VALUE, INTERNAL RATE OF RETURN FREQ PRESENT VALUE, INTERNAL RATE OF RETURN NONE What do these statements do to this new model? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

12. Use the Expert Choice software to select your next car
12. Use the Expert Choice software to select your next car. Evaluate cars on ride (from poor to great), looks (from attractive to ugly), and acceleration (seconds per first 50 yards). Consider three final cars on your list. Develop: a. Problem hierarchy b. Comparison of the importance of the criteria against the goal c. Comparison of the alternative cars for each criterion d. An overall ranking (synthesis of leaf nodes with respect to goal) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

e. A sensitivity analysis.
Maintain the inconsistency index lower than 0.1. If you initially had an inconsistency index greater than 0.1, what caused it to be that high? Would you really buy the car you selected? Why or why not? Also develop a spreadsheet model using estimated weights and estimates for the intangible items, each on a scale from 1-10 for each car. Compare the conclusions reached with this method to those found in using the Expert Choice Model. Which one more accurately captures your judgments and why? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

14. Job Selection Using Expert Choice
14. Job Selection Using Expert Choice. You are on the job market (use your imagination, if necessary). List the names of four or five different companies that have offered you a job (or from which you expect to get an offer). (As an alternative, your instructor may assign Graduate or Undergraduate Program Selection.) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Write down all the factors that may influence your decision as to which job offer you will accept. Such factors may include but need not be limited to geographic location, salary, benefits, taxes, school system (if you have children), and potential for career advancement. Some of these factors (criteria, attributes) may have sub-criteria. For instance, location may be sub-divided further into climate, urban concentration, cost of living, etc. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

If you, in fact, do not yet have a dollar salary figure associated with a job offer, you should just guess a "reasonable" figure. Perhaps your classmates can help you in determining realistic figures. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

a. Model this problem in a spreadsheet (Excel) using some kind of Weighted Average Methodology [you set the criteria weights first] (see the current Rand-McNally Places Rated Almanac for an example). b. Construct an Expert Choice model for your decision problem, and use the pairwise comparisons to arrive at the "best" job opportunity. c. Compare the two approaches. Did they yield the same results? Why or why not? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

d. Write a short report (one or two typed pages) explaining the results including those of the Weighted Average Methodology, and for Expert Choice: each criterion, sub-criterion (if any) and alternative. Describe (briefly) which options and capabilities of Expert Choice you used in your analysis, and show the numerical results of your analysis. To this purpose, you may want to include printouts of your AHP tree, but make sure you circle and explain the parts of interest on these printouts. Discuss the nature of the tradeoffs you encountered during the evaluation process. You may want to include a (meaningful) sensitivity analysis of the results, but this is optional (for this assignment). Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

To think about: Was the Expert Choice analysis helpful in structuring your preferences? Do you think it will be a helpful aid in your actual decision making process? Comment on all these issues in your report. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Term Paper Select a current DSS technology or methodology. Write up a 5 page report detailing the origins of the technology, what need prompted the development of the technology, and what the future holds for it over the next 2, 5 and 10 years. Use electronic sources, if possible, to identify companies providing the technology. If demo software is available, acquire it and include a sample run in your paper Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 6: Knowledge-based Decision Support and Artificial Intelligence
Managerial Decision Makers are Knowledge Workers They Use Knowledge in Decision Making Issue: Accessibility to Knowledge Knowledge-Based Decision Support Through Applied Artificial Intelligence Tools Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6.1 Opening Vignette: A Knowledge-based DSS in a Chinese Chemical Plant
The Problem Dalian Dyestuff plant Managers determined own production plans The Solution DSS with a knowledge-base component Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Subsystems Production Planning Accounting and Cost Control
Financing and Budgeting Inventory and Material Management Control Information Services LP-based production planning model in model base Two Expert Systems (ES) to Plan monthly production and Analyze working capital Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Expert System Generates a proposed plan Models the working capital

ES Advantages Combines quantitative and qualitative analysis
Provides flexibility and adaptability Involved decision makers Allows better and more efficient decisions Increased profit by more than $1 million / year (about a 10% increase) Allows users to express preferences and expertise Improves service to customers Improved the competitive position of the plant Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6.2 Artificial Intelligence (AI)

AI Concepts and Definitions
Encompasses Many Definitions AI Involves Studying Human Thought Processes (to Understand What Intelligence Is) AI Deals with Representing Thought Processes on Machines Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Artificial Intelligence
Artificial intelligence is behavior by a machine that, if performed by a human being, would be called intelligent (well-publicized) "Artificial Intelligence is the study of how to make computers do things at which, at the moment, people are better" (Rich and Knight [1991]) AI is basically a theory of how the human mind works (Mark Fox) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Objectives of Artificial Intelligence
(Winston and Prendergast [1984]) Make machines smarter (primary goal) Understand what intelligence is (Nobel Laureate purpose) Make machines more useful (entrepreneurial purpose) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

(Continued on next page)
Signs of Intelligence Learn or understand from experience Make sense out of ambiguous or contradictory messages Respond quickly and successfully to new situations Use reasoning to solve problems (Continued on next page) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Signs of Intelligence (cont’d)
Deal with perplexing situations Understand and Infer in ordinary, rational ways Apply knowledge to manipulate the environment Think and reason Recognize the relative importance of different elements in a situation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Turing Test for Intelligence
A computer can be considered to be smart only when a human interviewer, “conversing” with both an unseen human being and an unseen computer, could not determine which is which Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Symbolic Processing Use Symbols to Represent Problem Concepts
Apply Various Strategies and Rules to Manipulate these Concepts Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

AI: Represents Knowledge as Sets of Symbols
A symbol is a string of characters that stands for some real-world concept Examples Product Defendant 0.8 Chocolate Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Symbol Structures (Relationships)
(DEFECTIVE product) (LEASED-BY product defendant) (EQUAL (LIABILITY defendant) 0.8) tastes_good (chocolate). Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

AI Programs Manipulate Symbols to Solve Problems
Symbols and Symbol Structures Form Knowledge Representation Artificial intelligence is the Branch of Computer Science Dealing Primarily with Symbolic, Nonalgorithmic Methods of Problem Solving Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Characteristics of Artificial Intelligence
Numeric versus Symbolic Algorithmic versus Nonalgorithmic Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Heuristic Methods for Processing Information
Search Inferencing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Reasoning - Inferencing from Facts and Rules using heuristics or other search approaches
Pattern Matching Attempt to describe objects, events, or processes in terms of their qualitative features and logical and computational relationships Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6.3 Artificial Intelligence versus Natural Intelligence

Commercial Advantages of AI Over Natural Intelligence
AI is more permanent AI offers ease of duplication and dissemination AI can be less expensive AI is consistent and thorough AI can be documented AI can execute certain tasks much faster than a human can AI can perform certain tasks better than many or even most people Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Natural Intelligence Advantages over AI
Natural intelligence is creative People use sensory experience directly Can use a wide context of experience in different situations AI - Very Narrow Focus Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Information Processing
Computers can collect and process information efficiently People instinctively Recognize relationships between things Sense qualities Spot patterns that explain relationships BUT, AI technologies can provide significant improvement in productivity and quality Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6.4 Knowledge in Artificial Intelligence
Knowledge encompasses the implicit and explicit restrictions placed upon objects (entities), operations, and relationships along with general and specific heuristics and inference procedures involved in the situation being modeled Of data, information, and knowledge, KNOWLEDGE is most abstract and in the smallest quantity Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Uses of Knowledge Knowledge consists of facts, concepts, theories, heuristic methods, procedures, and relationships Knowledge is also information organized and analyzed for understanding and applicable to problem solving or decision making Knowledge base - the collection of knowledge related to a problem (or opportunity) used in an AI system Typically limited in some specific, usually narrow, subject area or domain The narrow domain of knowledge, and that an AI system must involve some qualitative aspects of decision making (critical for AI application success) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Knowledge Bases Search the Knowledge Base for Relevant Facts and Relationships Reach One or More Alternative Solutions to a Problem Augments the User (Typically a Novice) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6.5 How Artificial Intelligence Differs from Conventional Computing

Conventional Computing
Based on an Algorithm (Step-by-Step Procedure) Mathematical Formula or Sequential Procedure Converted into a Computer Program Uses Data (Numbers, Letters, Words) Limited to Very Structured, Quantitative Applications (Table 6.1) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Table 6.1: How Conventional Computers Process Data
Calculate Perform Logic Store Retrieve Translate Sort Edit Make Structured Decisions Monitor Control Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

AI Computing Based on symbolic representation and manipulation
A symbol is a letter, word, or number represents objects, processes, and their relationships Objects can be people, things, ideas, concepts, events, or statements of fact Create a symbolic knowledge base Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

AI Computing (cont’d) Uses various processes to manipulate the symbols to generate advice or a recommendation AI reasons or infers with the knowledge base by search and pattern matching Hunts for answers (Algorithms often used in search) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

AI Computing (cont’d) (Table 6.2) Caution: AI is NOT magic
AI is a unique approach to programming computers (Table 6.2) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Table 6.2: Artificial Intelligence vs. Conventional Programming

6.6 Does a Computer Really Think?
WHY? WHY NOT? Dreyfus and Dreyfus [1988] say NO! The Human Mind is Very Complex Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

AI Methods are Valuable
Models of how we think Methods to apply our intelligence Can make computers easier to use Can make more knowledge available to the masses Simulate parts of the human mind Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6.7 The Artificial Intelligence Field
Involves Many Different Sciences and Technologies Linguistics Psychology Philosophy Computer Science Electrical Engineering Hardware and Software Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Commercial, Government and Military Organizations Involved
(More) Mechanics Hydraulics Physics Optics Others Commercial, Government and Military Organizations Involved Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Lately Management and Organization Theory Chemistry Physics Statistics
Mathematics Management Science Management Information Systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Artificial Intelligence
AI is a Science and a Technology Growing Commercial Technologies Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major AI Areas Expert Systems Natural Language Processing
Speech Understanding Fuzzy Logic Robotics and Sensory Systems Computer Vision and Scene Recognition Intelligent Computer-Aided Instruction Machine Learning (Neural Computing) (Figure 6.3) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Expert Systems Attempt to Imitate Expert Reasoning Processes and Knowledge in Solving Specific Problems Most Popular Applied AI Technology Enhance Productivity Augment Work Forces Narrow Problem-Solving Areas or Tasks Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Human Expert Characteristics
Solve problems quickly and accurately Explain what (and how) they do Judge own conclusions Know when stumped Communicate with other experts Learn Transfer knowledge Use tools to support decisions Knowledge is a major resource Important to capture knowledge from a few experts Experts become unavailable -> knowledge not available Better than books and manuals Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Expert Systems Provide Direct Application of Expertise
Expert Systems Do Not Replace Experts, But Makes their Knowledge and Experience More Widely Available Permits Non Experts to Work Better Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Expert Systems Software Development Packages
Resolve (was EXSYS) K-Vision KnowledgePro Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Natural Language Processing
Can Communicate with the Computer in a Native Language Conversational Interface Limited Success Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Natural Language Processing (NLP)
Natural Language Understanding Natural Language Generation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Speech (Voice) Understanding
Recognition and Understanding by a Computer of Spoken Language Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Robotics and Sensory Systems
Vision Systems Tactile Systems Signal Processing Systems Plus AI = Robotics Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Robot An Electromechanical Device that Can be Programmed to Perform Manual Tasks Mars Rover: The AI Laboratory at MIT: Dante II: Robotics Institute, Carnegie Mellon University: Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Robot "a reprogrammable multifunctional manipulator designed to move materials, parts, tools, or specialized devices through variable programmed motions for the performance of a variety of tasks." Photo: Wheelesly, a Robotic Wheelchair: The AI Laboratory at MIT: Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Are Robots Part of AI? Not Always!

Computer Vision and Scene Recognition
Cheap Vision Machine: The AI Laboratory at MIT: Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Robotics and Computer Vision Web Sites
Carnegie Mellon University Robotics Institute: The AI Laboratory at MIT: Jet Propulsion Lab (NASA): List at the JPL:

Intelligent Computer-Aided Instruction (ICAI)
Machines that Can Tutor Humans Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Neural Computing Mathematical Model of the Way a Brain Functions
Other Applications Automatic Programming Summarizing News Language Translation Fuzzy Logic Genetic Algorithms Intelligent Agents Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6.8 Types of Knowledge-based Decision Support
Knowledge component extends the capabilities of computers well beyond data-based and model-based DSS Possible support for Qualitative aspects of the decision process Model management in a multiple model DSS Uncertainty analysis in applying AI tools The user interface (NLP and Voice Technology) Other Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6.9 Intelligent Decision Support Systems
Active (Symbiotic) DSS - Needed for Understanding the domain Problem formulation Relating a problem to a solver Interpreting results Explaining results and decisions Mili [1990] Need for an Intelligent Component(s) in the DSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Self-Evolving DSS - Extra Capabilities
Dynamic menu Dynamic user interface Intelligent model base management system Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Purposes of Self-Evolving DSS
Increasing the flexibility of the DSS Making the system more user friendly Enhancing control over the organization's information resource Encouraging system sharing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Structure of Self-Evolving DSS (Figure 6.4)
Major Components Data management, model management, and a user interface Usage record The user interface elements The central control mechanism Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Table 6.3: Problem Management

6.10 The Future of Artificial Intelligence
AI Research and Development Subfields Evolve and Improve New Software Techniques Improved Software Development Tools Improvement in ALL Decision Making Areas Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Hardware Advances Special search, pattern-matching, and symbolic processing chips New parallel computing and neural computing architectures Increased integration AI with other CBIS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

PLUS Natural language interfaces common
Intelligent databases economical Internet tools with intelligent agents and knowledge components Programs with knowledge-based subsystems for performance improvements Expert systems will become widely available Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Now Relatively few stand-alone AI application products (except ES)
Combinations of AI software and conventional algorithmic software / DSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

AI Transparent in Commercial Products
Anti-lock Braking Systems Video CAMcorders Kitchen Appliances Toasters Stoves Data Mining Software Help Desk Software Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary Artificial intelligence is an interdisciplinary field
The primary objective of AI is to build computer systems that perform intelligent tasks The major characteristics of AI are symbolic processing, heuristics and inferencing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

AI has several major advantages over people
Natural (human) intelligence has advantages over AI Knowledge is the key concept of AI Knowledge base Conventional computing vs. AI Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Techniques of reasoning: search and pattern matching
Digital computers are algorithmic but can be programmed for symbolic manipulation Techniques of reasoning: search and pattern matching AI computers may not think, but can be valuable Major application areas of AI Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Expert systems attempt to imitate experts
Effective expert systems are applied to a narrow knowledge domain and include qualitative factors Natural language processing Speech understanding Intelligent robots Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Intelligent Computer-Aided Instruction
Computer vision Fuzzy logic Genetic algorithms Intelligent agents Intelligent Computer-Aided Instruction AI technologies can be integrated together and with other CBIS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Intelligent DSS: Active
Intelligence is added to DSS by embedding knowledge bases Intelligence needed in problem management Active, symbiotic, and self-evolving DSS are different configurations of intelligent DSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Justify the need for a DSS Describe the role of the ES. Why was such a component needed? Review the role of the managers-users in this case What unique aspects in this case are related to the Chinese environment? What managerial lessons regarding DSS can be learned from this system? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Exercise 5. Commander Data, a member of the Enterprise starship crew was declared, legally, to be a sentient being, a culture of one, entitled to full rights as a citizen of the United Federation of Planets. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Answer the Following Describe the consequences of such a legal decision in today’s culture (recognizing that an artificial life form has equal stature to a human being). Do you think that such a court ruling will ever be possible? Why or why not? Should a sentient, artificial life form be entitled to “rights” in the human sense? Why or why not? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Exercise 1. Make a peanut butter and jelly sandwich in class.

Debates Do chess playing computer systems exhibit intelligence? Why or why not? Justify the position that computers cannot think. Then prepare arguments that show the opposite. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Bourbaki [1990] describes Searle's argument against the use of the Turing Test. Summarize all the important issues in this debate. The Soul: Proponents of AI claim that we cannot ever have machines that truly think because they cannot, by definition, have a soul. Supporters claim a soul is unnecessary. They cite the fact that originally humanity set out to create an artificial bird for flight. An airplane is not a bird, but yet it functionally acts as one. Debate the issue. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

APPENDIX 6-A: Human Problem Solving--An Information Processing Approach (The Newell-Simon Model)
Problem solving can be understood as information processing Based on a cognitive approach that uses a qualitative description of the ways in which people are similar, and of the manner in which people think Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Newell-Simon Model of Human Information Processing
Perceptual subsystem Cognitive subsystem Motor subsystem External memory (Figure 6-A.1) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Perceptual Subsystem External stimuli - inputs for human information processing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Cognitive Subsystem Selects appropriate information from sensory buffers and transfers it to the short-term memory Works in cycles Cognitive Subsystem Parts Elementary processor Short-term memory Interpreter Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Complex tasks - more elaborate processing
Cognitive processor draws on long-term memory Long-term memory - large number of stored symbols with a complex indexing system Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Simple Model of LTM Related symbols are associated with one another

Complex Model 1 of LTM Symbols are organized into temporal scripts
Memory consists of clusters of symbols called chunks Supports the decision-making process with external memory The long-term memory has essentially unlimited capacity The short-term memory is quite small Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major Limitations of Humans
The human operates in serial Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Motor Subsystem After scanning and searching memories, the processor sends information to the motor subsystem. Motor processors initiate actions of muscles and other internal human systems Results in some observable activity Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 7 User Interface and Decision Visualization Applications
Key to successful use of MSS is the user interface The simpler the better Many MSS applications have hard to use user interfaces Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.1 Opening Vignette: Geographic Information System at the Dallas Area Rapid Transit (DART)
Buses Vans Light Rail System Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

By the Mid-1980s Could Not DART had Respond to customer requests
Make changes rapidly Plan properly Manage security DART had 5,000 daily customer inquiries Over 200 bus routes Over 13,500 bus stops Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Geographic Information System (GIS) Solution
View and analyze data on digitized maps Now, DART Employees can Rapidly respond to customer inquiries (response time cut by 1/3) Provide more accurate information Plan services Perform environmental impact studies Cut bus schedule production costs Track bus locations via GPS Improve bus security Monitor subcontractors Analyze productivity and utilization Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Analysis time cut from days to less than an hour
Preparation of special maps: time cut from up to a week to five minutes (cost cut from $15,000 to pennies) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.2 User Interfaces: An Overview
Most computer users have limited computer experience Inexperienced users do not want to learn the computer-oriented details Most systems were developed for experienced users Need better user interfaces The design of an appropriate MSS user interface could be the most important determinant of success of the MSS implementation To many users, the user interface is the system! Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

User Interface Design is Influenced by User Characteristics
MSS execution time Learning time of the MSS Ease of recall System's versatility Errors made by end users Quality of help Adaptability to changes in the users' computer competency Concentration level required by end users Fatigue from using the system Command uniformity Fun the user derives Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

User Interface Human-computer interaction Surface
Physical aspects (see Figure 7.1) Input Devices Display (Output) Devices Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Cyclical Process (Figure 7.1)
1. Knowledge 2. Dialog 3. Action Language 4. Computer 5. Presentation Language 6. User's Reaction Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Important Issues in Building a User Interface
Choice of input and output devices Screen design Human-machine interaction sequence Use of colors and shading Information density Use of icons and symbols (especially for object-oriented) Information display format Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The User Interface Management System (UIMS)
Accommodates the various information representations Accommodates the action languages Provides an interface between the system user and the rest of the system Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.3 Interface Modes (Styles)
Interface (or interactive) Mode: the combination of presentation and action languages Determines how information is entered and displayed Determines the ease and simplicity of learning and using the system Menu interaction Command language Questions and answers Form interaction Natural language processing Graphical user interface (object manipulation) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Menu Interaction Includes Pull-down Menus (in GUI) Command Language
Questions and Answers Computer asks, user answers Form Interaction Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Natural Language Mainly with keyboard Some with voice input and output
Major limitation Inability of the computer to understand natural language AI advances are improving it Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Graphical User Interface (GUI)
Icons (or symbols) are directly manipulated by the user Most common PC GUI OS: Windows 95 Usability of four styles along four dimensions (Table 7.1) Hybrid Modes NLP + Hypermedia Command + Menu GUI + Menu Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

User Interface Importance
Interface cost can be 60 to 70 % of the total DSS cost Ideally, interface adaptable to different users’ needs and communicate consistent commands internally Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.4 Graphics Graphics Software
Purpose: to present visual images of information Integrated software packages: create graphic output directly from databases or spreadsheets Stand-alone graphics packages Integrated packages - often include 3-D graphic presentations and virtual reality Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Role of Computer Graphics
Help managers "visualize" data, relationships, and summaries (Figure 7.2) Graphics forms (Table 7.2) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.5 Multimedia and Hypermedia
Pool of human-machine communication media (Table 7.4) Sound Text Graphics Animation Video Voice Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Hypermedia Virtual reality via Virtual Reality Modeling Language (VRML) for Web delivery Hypermedia: multimedia documents linked by association Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Multiple Layers of Information
Menu-based natural language interface Object-oriented database A relational query interface A hypermedia abstract machine Media editors Change management virtual memory Especially effective in searching Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Hypermedia Characterizations
Explicitly linked different information structures Multimedia Linking information by association Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Classes of Hypermedia Presentation for knowledge and data navigation (Figure 7.3) Active participation in research to help record, organize, and integrate information and processes (Figure 7.4) Hypertext Nonlinear information access Follow a thread (drill) Internet browsing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Multimedia, Hypermedia, the Internet/Web and the Object-oriented Approach
GUI Icons Visual Programming Web Hooks Electronic Document Management (EDM) Problems with paper documents EDM systems Multimedia and Web access Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.6 Virtual Reality 3-D Presentations 3-D user interfaces
Manufacturing Marketing Virtual reality (VR) Decision making Advertising Data visualization Visual, spatial, and aural immersion VRML: Virtual Reality Markup Language for the Web Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.7 Geographic Information Systems (GIS)
Computer-based system for capturing, storing, checking, integrating, manipulating, and displaying data using digitized maps GIS Software GIS Data In-house or purchased GIS and Decision Making Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

GIS Applications Political campaign support
Consumer marketing and sales support Sales and territory analysis Site selection Fleet management Route planning Disaster planning Regulatory compliance Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

GIS and the Internet/Intranet GIS Servers Client GIS data
Emerging GIS Applications With GPS Intelligent GIS Virtual reality More Web hooks Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.8 Natural Language Processing (NLP)
Applied artificial intelligence technology Communicating with a computer in English (or other human) language Advantages: Disadvantages: Natural language understanding Natural language generation Versus speech recognition Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.9 Natural Language Processing: Methods
Natural language into the computer Example: English into Netscape Navigator Commands Natural language into another natural language - English to French Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major NLP Techniques Key word search (pattern matching)
Language processing (syntactic and semantic analysis) Neural computing (relatively new) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Key Word Analysis (Pattern Matching)
Pattern matching process: Search for selected key words or phrases Provide canned response Flow diagram (Figure 7.5) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Key Activities Parsing to determine word boundaries
Pattern matching to compare to prestored words and phrases OK for few key words Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Language Processing (Syntactic, Semantic, and Pragmatic Analysis)
Problems Many words with multiple meanings Many structures including those words in sentences Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Definitions Syntax analysis looks at the way a sentence is constructed; the arrangement of its components and their relationships Syntactic processes analyze and designate sentences to clarify the grammatical relationships between words in sentences Semantics assigns meaning to the syntactic constituents Pragmatic analysis relates individual sentences to each another and to the surrounding context Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Procedures How Language Processing Works
Simplified block diagram (Figure 7.6) Parser Lexicon Understander Knowledge base Generator Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Parser Syntactically Analyzes the Input Sentence
Each word is identified and its part of speech clarified The Parser maps the words into a structure called a parse tree The Parse tree shows the meanings of all of the words and how they are assembled The Lexicon is a dictionary The Parser is a pattern matcher and builds the parse tree The Understander works with the knowledge base to determine sentence meaning The Knowledge base is a repository of knowledge Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The understander uses the parse tree to reference the knowledge base
The understander can draw inferences from the input statement The generator can initiate additional action Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.10 Applications of Natural Language Processing and Software
Database interfaces Abstracting and summarizing text Grammar analysis Natural language translation Computer language to computer language translation Letter composition Speech understanding Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.11 Speech (Voice) Recognition and Understanding
The computer recognizes the normal human voice Advantages of Speech Recognition Ease of Access Speed Manual Freedom Remote Access Accuracy Good Morning Dave (2001) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Classifying Speech Recognizers
Word Recognizers Continuous Speech Recognizers Speaker Dependent Speaker Independent Voice Synthesis Computers speak Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

7.12 Research on User Interfaces in MSS
4 Independent Variables 1. Human user Demographics (age, education, experience) Psychological (cognitive style, intelligence, risk attitude). 2. Decision environment Decision structure Organizational level Others (stability, time pressure, uncertainty). Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4. Interface characteristics
3. Task Decision support (e.g., complexity level) Inquiry/information retrieval Data entry Word processing Computer-aided instruction. 4. Interface characteristics Input/output media Dialogue type Presentation format (tabular, graphical, colors, animation) Language characteristics (help facility, default options, other options). Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Dependent Variable: Human/Computer Effectiveness
Usefulness Perceived ease of use Performance (time, errors, profit) User attributes (satisfaction, confidence) Use of system option (high, low). Hwang and Wu [1990] Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Results of Some Experiments
1. Colors improve performance 2. Graphic versus tabular: inconclusive Research on Graphics and Modeling Metagraphs to represent system structure graphically for analysis New Interfaces Fish-eye View for GUI - Xerox Parc Research Center Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary Users want computer systems that are easy to use
The user interface represents the system to most users The user interface must be relatively friendly Graphics are crucial GIS Virtual reality Natural language processing and speech recognition Research on user interfaces continues Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Internet Exercise 10. Contact IBM ( to find information about their Voice Type Dictation, Merlin and other voice technology products. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Exercise Each group member will interview five computer users at school, work or home. For each user, identify the three interface modes preferred by the user, ranked in descending order. Also, the interviewer should discern the reasons why people prefer a particular interface mode. Then, the group will consolidate their findings and prepare a report to guide a novice computer user to the interface(s) with which he should become familiar. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. Why is a GIS considered a graphical user interface? 2. What are the advantages of GIS from a user interface point of view? 3. Which of the capabilities listed in the vignette support actual decision making? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Exercises 1. What is a natural language? Name two. What distinguishes a natural language from a computer language? Is Esperanto a natural language? Why or why not? 2. Obtain an NLP/DBMS software (e.g., Q&A). Try to use it on the database of Chapter 4, Exercise 5. Compare the use of a regular DBMS to the one supported by NLP. 3. Explain why icons in the Windows environment might be easier to use than typed commands. Demonstrate the two to verify your opinion. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

4. Why is it “easier” for a natural language to be translated into another by a human versus by a computer? 5. In the early days of language translation, the expression “The spirit is willing, but the flesh is weak” was translated to Russian and then back to English. The new English rendering was “The vodka is good, but the meat is rotten.” What happened? Why? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Questions for Case Application 7.1
1. Identify the voice recognition and voice synthesis portions of the system. 2. Identify all the tasks, which do not involve voice, carried out by a computer. 3. What paperwork can be eliminated by such a system? 4. What are the benefits to Nabisco? 5. What are the benefits to the employees? 6. What alternative communication technologies described in this chapter can be used instead of the system described here? Would you recommend any of these; why or why not? 7. Are there any disadvantages to the use of the technology? Explain. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 8 Constructing a Decision Support System and DSS Research
What must be done to acquire a DSS? DSS must be custom tailored Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.1 Opening Vignette: Hospital Healthcare Services Uses DSS
Jewish Hospital Healthcare Services (JHHS) Regional healthcare provider in Louisville, KY 7 facilities, 1,000 patient beds, 3,500 employees Total information management and computer services costs = 3 % of the operating budget SAS development tool 1991 JHHS managers can take clinical and financial files from the mainframe to perform analysis Nursing acuity system linked to the nurse staffing scheduling system Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1992 various DSS applications in
Productivity Cost accounting Patient mix Nurse staff scheduling Several different mainframe and PC software packages Early 1992, integrated mainframe-based DSS development tool MAPS Modeling Forecasting Planning Communications Database management systems Graphics Productivity DSS in MAPS Faster and easy to interpret Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.2 Introduction System Development Issues
Various commercial development software packages on different platforms Different software packages for different DSS applications Development packages for the mainframe applications PCs Diverse applications in different functional areas Vendors assisted in DSS construction Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

DSS Construction is Complicated
Technical Issues Behavioral Issues Many Different Approaches Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.3 Development Strategies
1. Customized DSS in general-purpose programming language 2. Fourth-generation language 3. DSS integrated development tool (generator or engine) 4. Domain-specific DSS generator 5. CASE methodology 6. Integrate several of the above approaches Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.4 The DSS Development Process
Prototyping Not all activities are performed for every DSS Process summary (Figure 8.1) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Phase A: Planning Phase B: Research
Need assessment and problem diagnosis Define objectives and goals of the DSS What are the key decisions? Phase B: Research Identification of a relevant approach for addressing user needs and available resources Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Phase C: System Analysis and Conceptual Design
Determination of the best construction approach and specific resources required to implement Includes Technical resources Staff resources Financial resources Organizational resources Conceptual design followed by a feasibility study Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Phase D: Design Phase E: Construction
Determine detailed specifications of system Components Structure Features Select appropriate software or write them Phase E: Construction Technical implementation of the design Tested and improve continuously Interface DSS with other systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Phase F: Implementation
Testing Evaluating Demonstration Orientation Training Deployment Phase G: Maintenance and Documentation Planning for ongoing system and user support Develop proper documentation Phase H: Adaptation Recycle through the earlier steps Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.5 The Development Process: Life Cycle versus Prototyping
Life-cycle approach Evolutionary prototyping approach (iterative process) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The System Development Life Cycle (SDLC) Approach and DSS
Inappropriate for Most DSS Users and Managers may not understand their information and modeling needs Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Evolutionary Prototyping Approach
Build a DSS in a series of short steps with immediate feedback from users 1.Select an important subproblem to be built first 2.Develop a small but usable system to assist the decision maker 3.Evaluate the system constantly 4.Refine, expand, and modify the system in cycles Repeat Stable and comprehensive system evolves Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Advantages of Prototyping
Short development time Short user reaction time (feedback from user) Improved users' understanding of the system, its information needs, and its capabilities. Low cost. Disadvantages and Limitations Gains might be lost through cycles Combining prototyping with the critical success method (Figure 8.3) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.6 Team-Developed Versus User-developed DSS
DSS 1970s and early 1980s Large-scale, complex systems Primarily provided organizational support Team efforts Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

User-Developed System Due to the Development of
Personal computers Computer communication networks PC-mainframe communication Friendly development software Reduced cost of software and hardware Increased capabilities of personal computers Enterprise-wide computing Easy accessibility to data and models Client/server architecture Balance Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.7 Team-Developed DSS Substantial effort
Extensive planning and organization Some generic activities Group of people to build and to manage it. Size depends on effort tools Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Organizational Placement of the DSS Development Group
1. In the information services (IS) department 2. Highly placed executive staff group 3. Finance or other functional area 4. Industrial engineering department 5. Management Science group 6. Information center group Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.8 End-user Computing and User-Developed DSS
End-user Computing (end-user development) the development and use of computer-based information systems by people outside the formal information systems areas End-users At any level of the organization In any functional area Levels of computer skill vary Growing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

User-Developed DSS Advantages
1. Short delivery time 2. Eliminate extensive and formal user requirements specifications 3. Reduce some DSS implementation problems 4. Low cost Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

User-Developed DSS Risks
1. Poor Quality 2. Quality Risks Substandard or inappropriate tools and facilities Development process risks Data management risks 3. Increased Security Risks 4. Problems from Lack of Documentation and Maintenance Procedures Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Issues in Reducing End-User Computing Risks
Error detection Use of auditing techniques Determine the proper amount of controls Investigate the reasons for the errors Solutions Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.9 DSS Technology Levels and Tools
Three Levels of DSS Technology Specific DSS [the application] DSS Integrated Tools (generators) [Excel] DSS Primary Tools [programming languages] Plus DSS Integrated Tools Now all with Web Hooks and easy GUI interfaces Relationships Among the Three Levels (Figure 8.6) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.10 Selection of DSS Development Tools
Questions a) Which tool(s) to use? b) Which hardware? c) Which operating system? d) Which network(s) to run it on? Options Mainframe DSS Software PC DSS Software (Unix) Workstation Software Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Complexity of the Software Selection Process
1. DSS information requirement and outputs are not completely known 2. Hundreds of software packages 3. Software packages evolve very rapidly 4. Frequent price changes 5. Several people involved Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6. One language for several DSS? Tool requirements may change
7. Dozens of criteria, some intangible, some conflict 8. Technical, functional, end-user, and managerial issues 9. Published evaluations are subjective and superficial 10.Trade off between open and closed environments Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

DSS Generator Selection
Some DSS generators are better for certain types of applications than others Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.11 Developing DSS Putting the System Together
Development tools and generators Use of highly automated tools Use of prefabricated pieces Both increase the builder’s productivity Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

DSS Development System Includes
Request (query) handler System analysis and design facility Dialog management system Report generator Graphics generator Source code manager Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Model base management system Knowledge management system
Object-oriented tools Standard statistical and management science tools Special modeling tools Programming languages Document imaging tools Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

DSS Development System Components
Some may be integrated into a DSS generator Others may be added as needed Components used to build a new DSS Core of the system includes a development language or a DSS generator Construction is done by combining programming modules Windows environment handles the interface Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.12 DSS Research Directions*
The DSS of the Future 1. Intelligent DSS can be proactive 2. Future DSS should be creative 3. DSS will become decision-paced 4. Larger role for management science, cognitive psychology, behavioral theory, information economics, computer science, and political science 5. Latest advances in computer technology improving DSS * Source: Based on J. J. Elam, J. C. Henderson, P. G. W. Keen and B. Konsynski, A Vision for Decision Support Systems, Special Report, University of Texas, Austin, TX, 1986. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6. Improved DSS apply to more unstructured problems
7. Must be able to create alternatives independently 8. Much longer-range perspective of DSS research 9. Research on interactions between individuals and groups 10.More examination of the human component of DSS: learning and empowerment. 11.The integration of DSS with other systems (ES, CBIS) 12.Expansion of the model management concept Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.Enhancement of DSS applications with values, ethics, and aesthetics
13.Enhancement of DSS theory (decision quality measurement, learning, and effectiveness) 14.New theories for of organizational decision making and group decision making 15.Enhancement of DSS applications with values, ethics, and aesthetics 16.Major research thrust in human-machine interfaces and their impacts on creativity and learning 17.Exploration to find the appropriate architectures for decision makers to use ES 18.Organizational impacts of DSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Extensive DSS Research
1. Broader view of decision making 2. Behavioral research 3. Research based on team theory 4. Stimulus-based DSS 5. Qualitative DSS 6. Usefulness of DSS 7. DSS and the Internet 8. Profile of DSS Research Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8.13 The DSS of the Future DSS Trends
1. PC-based DSS continues to grow 2. For institutionalized DSS: trend is toward distributed DSS 3. For pooled interdependent decision support, group DSS 4. Decision support system products are incorporating artificial intelligence: intelligent DSS 5. Focused versions of DSS toward specific sets of users or applications (EIS, GSS) 6. DSS groups moving into mainstream support 7. Continued development of user-friendly capabilities 8. The DSS software market continues to develop and mature Sprague and Watson [1996] Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Challenges of DSS Integrated Architecture Connectivity
Document Data Management More Intelligence Sprague and Watson [1996] Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Highlights / Summary DSS are complex and their construction can be
DSS Technologies Iterative (prototyping) approach DSS teams or individuals. End user computing allows decision makers to build their own DSS Most DSS are constructed with DSS development generators or with nonintegrated 4GL development tools Many DSS are also constructed in integrated software suites on personal computers. Tool and generator selection can be tricky. DSS research continues Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Debate Debate the issues (advantages and risks) in end-user DSS development. Use examples from the literature to back up your arguments. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Internet Assignments 1.Explore software vendors. Find vendors, download demos, identify user groups, and prepare a report. Group 1--Spreadsheet and modeling tools Group 2--Database related tools Group 3--Graphics and user interface tools. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. Describe the steps in Figure 8.1 that you can identify for the JHHS Vignette. 2. Why was a quantitative cost/benefit analysis not done? 3. Comment on the various DSS tools and generators. Can you classify them? 4. Why was the high level of trust and credibility in the integrity of the provided information important? 5. Discuss the benefits of the DSS. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

CASE APPLICATION 8.1: Wesleyan University--DSS for Student Financial Aid
Questions 1.Why was there a need for a DSS? 2.What kind of generators and tools were used during construction? 3.Identify some DSS capabilities that were used. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

APPENDIX 8-A: Prototyping
Process of building a "quick and dirty" version of an information system Throwaway Evolutionary Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Evolutionary Steps 1. Identify user's information and operating requirements in a "quick and dirty" manner. 2. Develop a working prototype that performs only the most important function (e.g., using a sample database). 3. Test and evaluate (done by user and builder). 4. Redefine information needs and improve the system. Repeat the last two steps several times Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Primary Features of Prototyping
1. Learning is explicitly integrated into the design process 2. Short intervals between iterations 3. User involvement is very important (joint application development (JAD) method) 4. Initial prototype must be low cost 5. Prototyping essentially bypasses the life-cycle stage of information requirements definition Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

PART 3: COLLABORATION, COMMUNICATION AND ENTERPRISE SUPPORT SYSTEMS
Decision support in more complex settings Potentially yielding large benefits Role of the Internet in decision support Use of computers to support collaborative work, especially decision making Specialized support given to executives and top managers, and organizational DSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

These are changing the way we
Chapter 9: Networked Decision Support--The Internet, Intranet And Collaborative Technologies These are changing the way we Manage Work Live Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.1 Opening Vignette: J.P. Morgan Combines Intranet and Notes
15,613 employees in 30 countries Web Browsers Lotus Notes to Simplify access to crucial corporate database resources on many different platforms and networks Consolidate the information quickly Intranet Applications on the Way Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.2 Networked Decision Support
Opening Vignette Major Characteristics Group decision making task Group members in different places Task must be accomplished very fast Impossible or expensive to bring all the team members to one place Some information in many sources (external and internal) Expertise by nonmembers of the team may be needed Use of information technology in a networked decision support environment Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Major Information Architecture
Internet Intranet Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Groupware Software Tools (Table 9.1)
Includes Group Decision Support Systems (GDSS), Groupware Examples Lotus Notes Netscape Communicator Electronic Commerce Execution of business via computer networks Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.3 The Internet: An Overview
A network of organizational internal computer networks Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Brief Internet History
Experimental project of the US Department of Defense (1969) to test the feasibility of a wide area computer network over which researchers, educators, military personnel and government agencies, could share data, exchange messages and transfer files 4 nodes in 1969 Over 50,000 nodes in 1996 (commercial orgs now) 50,000,000 individuals accessed in 1996 (est.) Backbone: main network that links the nodes Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Current Status Web Browser (Client) and Server Software: easy to use and more natural format for access World Wide Web (Web) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.4 The Intranet Internal Web is a network architecture designed to serve the internal informational needs of an organization, using Web (Internet) concepts and tools Operates within the company’s firewalls Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Intranet Changes Decision making processes
Organizational structure and procedures Helps reengineer corporations Helps to move IT to the end users Drives management consolidation Eli Lilly - U.S. Federal drug approval process done via an Intranet Geffen Records (Los Angeles, CA) for organizational support Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.5 Data Access and Information Retrieval
Use Internet and Intranets Intelligent Agents (Web Robot or Spider) to search and sift Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.6 Supporting Communication
Communication - Critical for Decision Support Decision Makers Communicate with Experts Government Agencies Customers Vendors Business Partners Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Groups of decision makers must
Decision makers access data and information stored in databases in several locations Groups of decision makers must Communicate Collaborate Negotiate Information technologies provide inexpensive, fast, and very capable means of providing communication Networked computer systems (Intranet/Internet) are the major enabling architectures Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Specific Communication Technologies
Electronic mail ( ) - Advantages Send and receive messages quickly over large distances Paperless communication Network connection from anywhere Mass mailing Trace correspondence Communicate with millions worldwide Collaborative computing Fast Information access Send and receive fully formatted documents Send and receive images Send and receive software Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

E-Mail Limitations Also No face-to-face communication
Typing knowledge required Security and confidentiality problems Also Enabled Messaging systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

* Electronic bulletin boards (EBBs)
*Chat Programs Webchat Internet Relay Chat (IRC) * Newsgroups UseNet News (News) * Mailing Lists * ListServe groups For details, to: * Electronic bulletin boards (EBBs) Some specialize in certain topics Others are general ALL can provide some level of decision support Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.7 Supporting Collaboration
A Time/Place Framework (Figure 9.2) Same Time/Same Place (Decision Room) Same Time/Different Place (Video Conference) Different Time/Same Place (via Internet) Different Time/Different Place Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Groupware Software products that support groups of people engaged in a common task or goal Provides a mechanism to share opinions and resources Term is ambiguous Thousands of software vendors and packages Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Groupware Applications and Development Tools
1. Idea Generation (Brainstorming) 2. Managing Sessions 3. Multi-criteria Making Products 4. Strategic Planning Decision Planner 5. Innovator (Meeting-Enhancement Keypad System) 6. One Touch (Supports Remote Teaching and Conferencing) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8. Higgins (Group scheduler) 9. Consensus Builder
7. OptionFinder (Brainstorm, weigh alternatives, identify priorities, vote, and work toward consensus in a non-decision room setting. Uses electronic keypads) 8. Higgins (Group scheduler) 9. Consensus Builder 10. Coordinator (Integrated , scheduling, and calendaring) 11. Meeting Maker (Groupware scheduler) 12. Vineyard (Repository of shared information ) 13. OptionLink (Real-time brainstorming for participants at different places and/or different times) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

14. LiveBoard (Large electronic document)
15. Iris (Distributed, multi-user collaborative multimedia writing environment) 16. Inforum (Meeting facilitating software) 17. The Meeting Room 18. InConcert (Workflow management system 19. Conference+ (Document sharing) 20. Team Expert Choice (Group version of Expert Choice) 21. GroupSystems for Windows (GroupSystems V for DOS) (Complete suite of electronic meeting room software) 22. TCBWorks (Web-based support for groups in electronic meetings over different times/different places ( Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

See Groupware Central at (The University of Georgia) The Unofficial Yellow Pages of CSCW Web site at (Technical University of Munich) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Workflow Systems Powerful business process automation tools
Have evolved into enterprise-wide computing

Three Types of Workflow Software
1. Administrative 2. Ad hoc 3. Production To provide workflow automation capabilities that provide end-users with tracking, routing, document imaging, and other facilities designed to improve business processes Web site references in text and on Web Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Screen Sharing Word processing Spreadsheets etc.
Distributed Interactive Desktop Groupware Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Enhanced Product Realization (EPR) System
Internet-based, state-of-the-art distributed system, by InfoTEST. Allow U.S. manufacturers to make product modifications anywhere in the world in as few as five days -- instead of several months EPR system to accelerate the time-to-market for new products and services Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

EPR integrates and leverages the Internet to enable collaborative manufacturing and electronic commerce applications - CAD/CAM, Product Data Management Systems, electronic white-boarding, and multi-point desktop videoconferencing Internet - non-proprietary, open standards environment: to bring collaborative computing to the entire supply chain Extranet Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.8 Electronic Teleconferencing
Telephone Conferencing. Videoconferencing (now on the desktop PC) (Video mail) Major Benefits of Videoconferencing Face-to-face communication in different location Simultaneous communication Several media types Voice Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Problems with Videoconferencing
Connectivity Problems Choppy Motion Lack of Standards Image Compression Problems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.9 Lotus Notes (Domino Server)
Popular Integrated Groupware Kit Distributed Document Database Wide Variety of Capabilities Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Characteristics and Benefits of Lotus Notes
1. Single, Consistent User Interface 2. Compound Documents (multiple data types from multiple sources) 3. Rapid Application Development Environment - allows for the rapid development of workgroup applications 4. Advanced Security 5. Use of Replication 6. Openness 7. Expanding Industry of value-added products and services 8. Scalability 9. Seamless Integration of client and server elements 10. Availability 11. Web-based interfaces Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Capabilities of Lotus Notes
Tracking Broadcasting References Team Discussion Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Specialized Application Categories
Things to do Contract library Corporate policy documents Notes - Tool for sharing information and improving communication Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Notes Database Types Discussion Databases Document Libraries
Information Services Notes as a Forms Generator Conventional Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Disadvantages of Notes
Cost Other Aspects Lotus Notes / Domino - Web functionality Other Notes-related Products Notes' capabilities (Figure 9.3) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.10 Netscape Communicator
Integrated client software product that allows users to communicate, share data, and access information on Intranets and the Internet Netscape Communicator is an integrated product for open Groupware Editing Calendaring Web browsing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Netscape Communicator Integrates Five Powerful Components
Netscape Navigator 4.0 Browser Software Netscape Composer HTML Authoring Software Netscape Messenger Electronic Mail Netscape Collabra Group Discussion Software Netscape Conference Real-time Collaboration Software Netscape Communicator Professional Edition adds Netscape Calendar scheduling software Netscape AutoAdmin Netscape Inbox Direct Service. Inbox Direct Customized, media-rich electronic news delivery service Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Advantages of Netscape Communicator
Inexpensive Web Browser access Interface is commonly available and known Automatically can access information from any Web site Provides most of the support needed by work groups Netscape SuiteSpot Enterprise server can provide access to its group generated information Most databases convert data and files automatically to HTML Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Disadvantages of Netscape Communicator
No database management capability Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.11 Electronic Commerce (EC)
Buying and selling of products, services and information via computer networks A modern business methodology that addresses the needs of organizations, merchants, and consumers to cut costs while improving the quality of goods and services and increasing the speed of delivery Forrester Research predicts that online shopping will be a $6.6 billion business in the year 2000, up from $518 million in 1996 In 1995, $45.5 billion worth of business-to-business transactions involved the exchange of documents, purchase orders, invoices, or shipping notices About 20% of those documents were transferred electronically, accounting for $10 billion worth of transactions Number of electronic transfers is growing at better than 15% per year, almost 3 times as fast as the growth rate for all business transactions Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Internet Advertisement Auctions Virtual reality experiences

EC Applications Electronic Data Interchange (later) Other (now!)

Consumer-Oriented Electronic Commerce
Enhancing the customer service life cycle phases Requirements: assisting the customer determine needs Acquisition: helping the customer acquire a product or service Ownership: supporting the customer on an ongoing basis Retirement: helping the client dispose of the service or product How? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Advertising on the Internet Marketing and Sales
Market and Product Research Customer Support Airline flight and fare information Package shipping and tracking software Electronic help desks Free software upgrades, add-ins, and printer drivers over the Web Investment analysis tools Trading Securities and Commodities The Job Market Electronic Malls--the Cyber Marketplace Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Business-to-Business Commerce
Via EDI (Electronic Data Interchange) [100’s of billions of dollars/year] Advertising Market research Auctions Trading commodities Venues for commerce: 1. Information distribution 2. EDI interface 3. High-bandwidth pipeline Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.12 Electronic Data Interchange (EDI)
Special type of Special Characteristics of EDI 1. Business Transactions Messages 2. Data Formatting Standards US and Canada - ANSI X.12; UN - EDIFACT 3. Data Formatting and EDI Translators (to standards - X.12) How EDI Works (Figure 9.4) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major Advantages of EDI
Send and receive large amounts of information globally in real time. Very few errors consistently and free information flow among trading partners Companies access partners' computers to retrieve and deposit standard transactions True (and strategic) partnership relationship Paperless transaction processing system Shorter cycle time for collecting payments Automated payments Cost savings Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

EDI Issues EDI and the Internet Third Party EDI For Small Businesses
GE Information Services (GE TradeWeb) Harbinger Corp. (TrustedLink Guardian) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.13 ETHICS AND LEGAL ISSUES ON THE NET
1. Privacy and Ethics in 2. Right of Free Speech 3. Copyright (Freeware, Shareware) 4. Privacy of Patients’ Information 5. Internet Manners (Netiquette, Network Etiquette, Flaming) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Legal Systems Eventually Catch Up with New Issues
Employers do own Bulletin board system owner/operators are responsible for content Copyrighted Web material is not considered a copy Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

9.14 Telecommuting (Working at Home)
Employees work at home on a computer More teams with members working at home Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Advantages of Telecommuting
Fewer interruptions Increased productivity. Less office and parking needed Gainful employment of housebound people Flexible hours Driving time and expense saved Reduced pollution, traffic and fossil fuel use People can be hired for specific tasks. Improvement in workforce quality Happier, more motivated workforce Traveling workers can keep in touch Freedom to operate anywhere Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major Disadvantages of Telecommuting
Supervision difficulties Lack of human interaction Increased isolation Some eliminated by coming to work periodically Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Telecommuting Support
Groupware Regular and Overnight Mail Special Messengers Fax Machines Scanners The Virtual Office is Gaining Ground Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary Groupware Distributed GSS over the Internet / Intranet
The same/different time/place framework People can work at home supported by GSS Using browsers and search engines, one can access enormous amount of information worldwide Electronic mail Workflow systems Chat programs, discussion lists and newsgroups Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Video-teleconferencing
Audio and video teleconferencing over the Internet. Demo software available on the Web for enterprise support applications Lotus Notes and Netscape Communicator Electronic commerce (EC) EDI is a special electronic mail EDI over the Internet. There are many open legal and ethical questions Telecommuting

1. What was the goal of the groupware applications at J.P. Morgan? 2. What is the potential impact of giving all employees access to the Web-based applications? 3. How was J.P. Morgan able to integrate Lotus Notes with its Web strategy? Why was this a good idea? 4. Relate the J.P. Morgan applications to the Price Waterhouse activities described in DSS in Action 9.1. 5. How can using Web browser technology bypass many hardware and software integration problems when setting up groupware? 6. How are the tools being used? Do these applications fit the mold for collaborative technology use? How or how not? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Case Application 9.1: Cushman And Wakefield Uses Intranet for Decision Support
Case Questions 1. Why was the Intranet the best solution for the company’s problems? 2. Why was it possible to install an Intranet so inexpensively? 3. How can a chat form improve communication? 4. How can the company brokers communicate with outside brokers using an Intranet/Internet mix? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Case Application 9.2: General Mills Uses EDI
Case Questions 1. Why was the EDI response time 24 hours? 2. Compare the use of EDI to that of Lotus Notes. 3. Why do they need two EDI codes? 4. Why do they need several VAN vendors? 5. Why was it important to select a translator that is VAN-independent? 6. Why was "enlisting trading partners" the last step in the process, and what difficulties may one encounter? 7. From what you learned about EDI, what kind of transactions can take place between a food manufacturer and its suppliers and customers? 8. Can this EDI be delivered on the Internet? Should it? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

APPENDIX 9-A: FUNDAMENTALS OF THE INTERNET
Accessing the Internet LAN at work or school via an Internet backbone Call from home via modem Commercial provider Web TV Special Internet terminals (Internet Lite) Internet Kiosks Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Internet Concepts Telnet The TCP/IP Protocol
Transmission Control Protocol / Internet Protocol Messages via Packets The World Wide Web (Web) Web Browser (client) - GUI Interface Netscape Navigator Microsoft InternetExplorer Text Web Browser - Lynx Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Browsing Like in a Bookstore Hyperlinks
Hypertext Markup Language (HTML) Home Page Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Plug-in Applications: Plug-and-play Applications
Search Engines Yahoo, Lycos, WebCrawler, Alta Vista, Infoseek and Excite Uniform Resource Locator (URL) - location (or address) of a Web site Hypertext Transport Protocol (HTTP) www = on the Web Oorganization’s Domain Name(s) Organization Type (com, edu, gov, org, ...) Country Code (au, fr, uk, cn) File or Subdirectory String Create Web Documents in the Hypertext Markup Language (HTML) or Use Web Authoring Tools (from Microsoft, Claris, Netscape) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Gopher Gopher can access any type of textual information on the Internet Menu-oriented Downloading Software Freeware Shareware ( FTP and Downloading By the Browser, Using File Transfer Protocol (FTP) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Specialized Search Tools
Veronica (Very Easy Rodent-Oriented Netwide Index to Computerized Archives) for Gopher resources Archie (short for archives) for FTP resources REMEMBER - It’s EASY!!! Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 10: Group Decision Support Systems
Most complex decisions made by groups Complexity of organizational decision making implies more need for meetings and groupwork Complex meeting preparation and activities Need for computerized support Group Decision Support Systems (GDSS) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.1 Opening Vignette: Quality Improvement Teams at the IRS of Manhattan
Internal Revenue Service (IRS), Manhattan, NY and the University of Minnesota Implemented a quality improvement program Supported by a Group Decision Support System (GDSS) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Problem Quality team participants Different functional areas
Different supervisory levels Different perspectives May induce process and task losses Domination by one or a few members Poor interpersonal communication Fear to express innovative ideas Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Solution: Group DSS
(Collaborative Computing) Technology Supports the Activities of A Decision Making Group Its Leader, and Its Facilitator (Table 10.1) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Implementation Used the GDSS facility at the University of Minnesota
SAMM, for Software-Aided Meeting Management Results Several Hundred Meetings Idea generation and evaluation Using sophisticated decision aid tools Creating and managing the agenda Group writing and record keeping Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

High level of satisfaction Comfortable with the technology
Improvement of teamwork GDSS was easy for the group to use GDSS played a major role in meetings Almost no negative effects reported Tremendous, positive impact Enhancements and the Future Different time / different location access Emotional aspects Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.2 Decision Making in Groups
Some fundamentals of group decision making 1. Groups 2. The Nature of Group Decision Making a) Meetings are a joint activity b) The outcome of the meeting depends on its participants. c) The outcome of the meeting depends on the composition of the groups d) The outcome of the meeting depends on the decision-making process e) Differences in opinion are settled either by the leader or negotiation or arbitration Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

5. Improving the Work of Groups
3. The Benefits and Limitations of Working in Groups (Table 10.2). But process Losses (Table 10.3) 4. Dispersed Groups 5. Improving the Work of Groups Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Nominal Group Technique (NGT)
(Delbec and Van de Ven) NGT Sequence of activities 1. Silent generation of ideas in writing 2. Round-robin listing of ideas on a flip chart 3. Serial discussion of ideas 4. Silent listing and ranking of priorities 5. Discussion of priorities 6. Silent re-ranking and rating of priorities Procedure is superior to conventional discussion groups in terms of generating higher quality, greater quantity, and improved distribution of information on fact-finding tasks Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

NGT does not solve several process losses
NGT success depends on Facilitator quality Participants’ Training NGT does not solve several process losses Fearing to speak Poor planning Poor meeting organization Compromises Lack of appropriate analysis Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Delphi Method (RAND Corporation)
Goal - To eliminate undesirable effects of interaction among group members Experts do not meet face-to-face Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Delphi Method Steps 1. Each expert provide an individually written assignment or opinion 2. Delphi coordinator edits, clarifies and summarizes the raw data 3. Provide anonymous feedback to all experts 4. Second round of issues or questions 5. Etc. Get more specific in each iteration, leading to consensus or deadlock Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Delphi Method Benefits
Anonymity Multiple Opinions Group Communication Plus, Avoids Dominant Behavior Groupthink Stubbornness Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Delphi Method Limitations
Slow Expensive Usually limited to one issue at a time Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.3 Group Decision Support Systems (GDSS)
Group Support Systems (GSS) Electronic Meeting Systems Collaborative Computing Evolved as information technology researchers recognized that technology could be developed for supporting meeting activities Idea generation Consensus building Anonymous ranking Voting, etc. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Two GDSS Schools of Thought
Social Sciences Approach Engineering Approach Now - Effective Merger Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

GDSS Definition Consists of a set of software, hardware, language components, and procedures that support a group of people engaged in a decision-related meeting (Huber [1984]) An interactive computer-based system that facilitates the solution of unstructured problems by a group of decision makers (DeSanctis and Gallupe [1987]) Components of a GDSS include hardware, software, people, and procedures Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Important Characteristics of a GDSS
Specially Designed IS Goal of Supporting Groups of Decision Makers Easy to Learn and Use May be designed for one type of problem or for many organizational decisions Designed to encourage group activities Attempts to minimize process losses Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

GDSS: Part of GSS or Electronic Meeting Systems (EMS)
“An information technology (IT)-based environment that supports group meetings, which may be distributed geographically and temporally. The IT environment includes, but is not limited to, distributed facilities, computer hardware and software, audio and video technology, procedures, methodologies, facilitation, and applicable group data. Group tasks include, but are not limited to, communication, planning, idea generation, problem solving, issue discussion, negotiation, conflict resolution, system analysis and design, and collaborative group activities such as document preparation and sharing (p. 593, Dennis et al., 1988). Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

GDSS Settings Single Location Multiple Locations
Common Group Activities Information Retrieval Information Sharing Information Use Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.4 The Goal of GDSS and Its Technology Levels
Goal - to improve the productivity and effectiveness of decision-making meetings, either by speeding up the decision-making process or by improving the quality of the resulting decisions GDSS attempts to Increase the benefits of group work (Tables 10.2 and 10.4) Decrease the losses (Table 10.3) By providing support to the group members (center column, Figure 10.1) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

GDSS Technology Levels
Level 1: Process Support Level 2: Decision-Making Support Level 3: Rules of Order Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Level 1: Process Support
Goal - to reduce or remove communication barriers Supports Electronic messaging Networks (Local) Public screen Anonymous input of ideas and votes Active solicitation of ideas or votes Summary and display of ideas and opinions and votes Agenda format Continuous display of the agenda, etc. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Level 2: Decision-Making Support
Adds modeling and decision analysis Goal - to reduce uncertainty and noise Provide task gains Features Planning and financial models Decision trees Probability assessment models Resource allocation models Social judgment models Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Level 3: Rules of Order Focus on decision making process
Controls its timing, content or message patterns Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.5 GDSS Technology GDSS Technology Options Components (Figure 10.2)
1. Special-purpose electronic meeting facility (decision room) 2. General purpose computer lab 3. Web (Internet) / Intranet or LAN-based software for any place / any time Components (Figure 10.2) Hardware Software People Procedures Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

GDSS Hardware 1. Single PC 2. PCs and Keypads 3. Decision Room
4. Distributed GDSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

GDSS Software Modules to support the individual, the group, the process and specific tasks Typical Group Features Numerical / graphical summarization of ideas, and votes Programs calculating weights for alternatives; anonymous idea recording; selection of a group leader; progressive rounds of voting; or elimination of redundant input Text and data transmission among the group members, between the group members and the facilitator, and between the members and a central data / document repository. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

People Group Members Facilitator (Chauffeur)
Procedures (that enable ease of operation and effective use of the technology) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.6 The Decision (Electronic Meeting) Room
12 to 30 networked personal computers Usually recessed Server PC Large-screen projection system Breakout rooms Example (Figure 10.3) See Cool Rooms at Need a Trained Facilitator for Success Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Cool Rooms Murraysville School District Bus
Source: Ventana Corp., Tuscon, AZ, Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Why Few Organizations Use Decision Rooms
High Cost Need for a Trained Facilitator Software Support for Conflict Issues, NOT Cooperative Tasks Infrequent Use Different Place / Different Time Needs May Need More Than One Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.7 GDSS Software Comprehensive GDSS Software Emerging Web-Based GDSS
.GroupSystems for Windows (Ventana Corp.) .VisionQuest (Collaborative Technologies Corp.) .TeamFocus (IBM Corp.) .SAMM (University of Minnesota) .Lotus Domino / Notes (Lotus Development Corp.) .Netscape Communicator (Netscape Communications Corp.) Emerging Web-Based GDSS .TCBWorks (The University of Georgia) ( Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

GroupSystems for Windows
Overview (Figure 10.4) Agenda is the Control Panel Standard Tools 1. Electronic Brainstorming (Figure 10.5) 2. Group Outliner 3. Topic Commenter 4. Categorizer 5. Vote Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Advanced Tools Other Resources 1. Alternative Analysis 2. Survey
3. Activity Modeler Other Resources People Whiteboard Handouts Opinion Meter Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Individual Resources Briefcase (Commonly Used Applications)
Personal Log Event Monitor Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Internet-Based GDSS Many new GDSS are Web-based
e.g., TCBWorks (October 10, 1995) Sample of Web-based GDSS TCBWorks Lotus Domino/Notes Netscape Communicator BrainWeb InterAction: A Web-based Collaboration Tool End of 1996, over 75 Web-based groupware systems See Groupware Central More developments on Web-based GDSS coming Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.8 Idea Generation Collaborative Effort Idea-chains
More Creativity Using GDSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.9 Negotiation Support Systems (NSS)
Conflicts among decision makers Negotiation - for Conflict Resolution Types of Disputes Negotiators' interests are fundamentally opposed Negotiators share basic objectives, differ in priorities GDSS Can Provide Structure to Negotiations Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Conflicts arise because of differences in interests among individuals
Conflicts also arise because of differences in problem understanding (misunderstanding) Requirements Negotiation 1. Conflict Detection 2. Resolution Generation 3. Resolution Choice Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Agent model - Electronic Brokers in Electronic Commerce
Oz - Prototype NSS Code Agent model - Electronic Brokers in Electronic Commerce NSS Can Provide Organizational Memory NSS Can be Deployed Over the Web Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.10 The GDSS Meeting Process
1. Group leader meets with facilitator to Plan the meeting Select the software tools Develop an agenda 2. Participants are gathered in the decision room and the leader poses a question or problem to the group. 3. Participants type their ideas or comments 4. Facilitator searches for common themes, topics, and ideas and organizes them into rough categories (key ideas) 5. Leader starts a discussion and participants prioritize the ideas 6. Top 5 to 10 topics are routed to idea generation software, after discussion 7. Repeat the process Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major Activities of a Typical GDSS Session (Figure 10.6)
Example (Appendix 10-A) Different Time/Different Place Meetings (via Intranets, the Internet/Web) Just Learning How to Do Distributed GDSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.11 Constructing a GDSS and the Determinants of Its Success
1. Construct (or Rent) a Decision Room 2. Acquire Software 3. Develop Procedures 4. Train a Facilitator 5. Put It All Together Can Use Someone Else's Facility Rent One Use a Dual Purpose Computer Lab Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Determinants of GDSS Success for a Decision Room Setting
Same Time/Same Place Meetings (DSS In Focus 10.4) Trained Facilitator Support Participants’ Training Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

More on Critical Success Factors for GDSS
1. Design a) Enhance the structuredness of unstructured decisions b) Anonymity c) Organizational involvement d) Ergonomic considerations Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

User involvement and participants’ behavior are also important factors
2. Implementation a) Extensive and proper user training b) Support of top management c) Qualified facilitator. d) Execute trial runs 3. Management a) Reliable system b) Incrementally improve system c) GDSS staff keeps up with technology User involvement and participants’ behavior are also important factors Building Decision Rooms Using Off-the-Shelf Software Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

10.12 GDSS Research Challenges
Differences in effectiveness between field studies and lab experiments Multi-methodological research programs to understand key GDSS issues Understand the effects that these differences have on the process and outcomes of group meetings Use this understanding to interpret and apply the conclusions of experiments to organizations’ GDSS use Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Research Models 1. GDSS Variables (Figure 10.7)
2. GDSS Research Topics (Table 10.5) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Additional Topics (Dickson [1991])
1. Tradeoffs of advanced features versus simplicity 2. Interface studies 3. Supporting groups with difficulties 4. Research on anytime/anyplace configurations 5. GDSS benefit identification 6. Level of support required 7. Embedded knowledge bases (Intelligent GDSS) 8. Training issues 9. Individuals’ characteristics Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Research Direction Categories
1. What groups do 2. Effects of GDSS on group work 3. Effects of GDSS on organizations 4. Effects of hardware on GDSS performance 5. Effects of software on GDSS performance 6. Cultural effects of GDSS 7. Training people to use GDSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

8. Cost-benefit analysis for GDSS
9. Critical success factors for implementation in industry 10. Robustness of research results 11. Innovative uses of GDSS 12. Theoretical foundations of GDSS 13. Barriers to research 14. Research methodologies 15. Anytime / anyplace meetings 16. Other ideas and research topics. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Six Major GSS Scenarios for Research
1. Anytime / Anyplace 2. Orchestrated Workflow 3. Virtual Team Rooms 4. Culture Bridging 5. Just-In-Time Learning 6. Window to Anywhere Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Research Sampler Er and Ng [1995]: Value of anonymity, group dynamics, organizational settings, social contexts and behavioral aspects Valacich and Schwenk [1995]: Value of a devil’s advocacy (a cognitive conflict technique) enhanced decision making performance Bryson et al. [1994]: Value of special voting approaches Anson et al.[1995]: Human facilitation impacts Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Lim et al. [1994]: Effect of lack of leadership in a GDSS meeting
Miranda and Bostrom [1997]: Impacts of process and content meeting facilitation across traditional and GDSS environments on meeting processes and outcomes Lim et al. [1994]: Effect of lack of leadership in a GDSS meeting Yellen et al. [1995]: Impact of individuals’ characteristics on GDSS outcome Dennis et al. [1996]: Effects of time and task decomposition on electronic brainstorming to produce more and more creative ideas Dennis et al. [1997]: Effects of multiple dialogues versus single dialogues for electronic brainstorming to reduce cognitive inertia Chidambaram [1996]: Individuals using GDSS technology over time tend to feel more cohesive as a group than groups not Groups whose members are dispersed and possibly communicate in different times Intranets, the Internet and Web-based GDSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary Many benefits to work groups, but many process losses
Delphi method and the Nominal Group Technique (NGT) Group Decision Support Systems (GDSS), Group Support Systems (GSS), electronic meeting systems (EMS), computer-supported cooperative work, collaborative computing, groupware, etc. - various computer support for groups GDSS attempts to reduce process losses and increase process gains GDSS over the Internet and Intranets, anytime/anyplace Group DSS over a LAN in a decision room environment Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

NSS can aid in resolving conflicts in groups GDSS can fail
Idea generation, idea organization, stakeholder identification, topic commentator, voting, policy formulation, enterprise analysis and negotiation support system NSS can aid in resolving conflicts in groups GDSS can fail GDSS research is very diverse Web-based group-ware for anytime/anyplace collaboration The Internet and Intranets - major role in distributed GDSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Exercises 1. Prepare a list of topics that are in your opinion most suitable for a decision room. List one each in the areas of: accounting, finance, marketing, manufacturing, government, an educational institution and a hospital. 2. Describe the advantages and disadvantages of using an existing computer lab for a dual-purpose to include GDSS technology. 3. Explain in detail, why it is preferable to build a GDSS facility with vendor software rather than to develop the software from scratch. 5. Group Brainstorming Exercise 1: (with paper) Standard brainstorming (standard meeting) Nonverbal brainstorming 1 (single dialogue meeting) Nonverbal brainstorming 2 (multiple dialogue meeting) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Case Application 10.1: GDSS In City Government
Case Questions 1. Why is this a "limited GDSS"? 2. The decision room accommodates the participants in 4 shifts, due to the small size of the GDSS facility. What could be the impact of such a restriction? How could new technology overcome this process loss? 3. Compare the activities in this case to those of the Delphi method. 4. What are some of the major advantages of this GDSS? 5. How would a GDSS for public policy compare with one in a for-profit organization? 6. Can the Internet be used instead of the decision room in this case? How? List the benefits inherent in this type of use. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Case Application 10.2: Chevron Pipe Line Evaluates Critical Business Processes with a GDSS
Case Questions 1. What happened to the team members at the end of the first day of meetings? Why? 2. What process and task gains do you think Group-Systems provided? Why? 3. What new technologies could be used for encouraging participation of team members not on site? How and why? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

APPENDIX 10-A: Example of a GDSS Meeting: The Queen Mary
The Topic: What to Do with the Queen Mary? March 6, 1992, Walt Disney Company Ended Its Lease of the Queen Mary What to Do? Use the Electronic Decision Support Facility at California State University, Long Beach Group-Systems Software Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Process Round one: Electronic Brainstorming
Round two: Ideas organized into 8 groups Round three: Groups ranked (Table 10-A.1) - very little agreement Round four: Discussion to reach consensus Reordering of key issues and increased level of agreement Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 11: Executive Information and Support Systems
DSS have been rarely used by top executives Why? What are the needs of top executives? What is needed in computer-based information systems for upper management?

Unique MSS Tools Executive Information Systems (EIS)
Executive Support Systems (ESS) and Organizational DSS (ODSS) Plus Client/Server Architecture (C/S) Enterprise Computing

11.1 Opening Vignette: The Executive Information System at Hertz Corporation
The Problem High competition Keys to Success - Marketing and flexible planning Instantaneous marketing decisions (decentralized) Based on information about cities, climates, holidays, business cycles, tourist activities, past promotions, and competitors' and customers' behavior Must know competitors’ pricing information The Problem - How to provide accessibility to this information and use it effectively

The Initial Solution: A Mainframe-Based DSS (1987)
Later: The Executive Information System (EIS) in 1988 PC-based front-end to the DSS Commander EIS (Comshare Inc.) Tools to analyze the mountains of stored information To make real-time decisions without help Extremely user-friendly Maintained by the marketing staff Continuous upgrades and improvements Conformed to how Hertz executives work Implementation and acceptance were no problem System allows Hertz to better use its information and IS resources

11.2 Executive Information Systems: Concepts and Definitions
Tool that can handle the executives’ many needs for timely and accurate information in a meaningful format (DSS In Focus 11.1) Most Popular EIS Uses Decision making (by providing data) Scheduling (to set agendas and schedule meetings) and electronic briefing (to browse data and monitor situations) (Table 11.1)

Majority of personal DSS support the work of professionals and middle-level managers
Organizational DSS support planners, analysts, and researchers Rarely do top executives directly use a DSS Executive Information Systems (EIS) (or Executive Support Systems (ESS) Technology emerged to meet executive information needs

EIS - Rapid growth Prime Tool for Gaining Competitive Advantage Many Companies - Sizable Increase in Profits with EIS Sometimes the Payback Period is Measured in Hours New Internet / World Wide Web and Corporate Intranets EIS Developments

EIS and ESS Definitions
Executive Information System (EIS) A computer-based system that serves the information needs of top executives Provides rapid access to timely information and direct access to management reports Very user-friendly, supported by graphics Provides exceptions reporting and "drill-down" capabilities Easily connected to the Internet Drill down

Executive Support System (ESS)
A Comprehensive Support System that Goes Beyond EIS to Include Communications Office automation Analysis support Intelligence (DSS In Action 11.2)

11.3 Executives’ Role and Their Information Needs
Decisional Executive Role (2 Phases) 1. Identification of problems and/or opportunities 2. The decision of what to do about them Flow Chart and Information Flow (Figure 11.1) Use Phases to Determine the Executives’ Information Needs

Methods for Finding Information Needs
Wetherbe's Approach [1991] (Figure 11.2) 1. Structured Interviews (Table 11.2) IBM's Business System Planning (BSP) Critical Success Factors (CSF) Ends/Means (E/M) Analysis 2. Prototyping Watson and Frolick's Approach [1992] .Asking (interview approach) .Deriving the needs from an existing information system .Synthesis from characteristics of the systems .Discovering (Prototyping) Ten methods (Table 11.3)

Volonino and Watson’s Strategic Business Objectives Approach [1991]
Attempts to address some potential problems of the other methods Ignoring soft information Identifying the information timeliness Independence of information and specific executives

Organization-wide view
Identify business objectives Link them to the information needs of individuals throughout the organization EIS evolves into an enterprise-wide system

SBO Method Determine the organization’s SBOs
Identify related business processes Prioritize the SBOs and their related business processes Determine the information critical to each business process Identify information linkages across the SBO business processes Plan for development, implementation and evolution SBO method meshes well with Business Process Reengineering Requires extensive coordination of communication between executive users and EIS developers

Other Approaches Information Success Factors Approach
Problem: Needs Change as Executives’ Tasks and Responsibilities Change EIS Evolves

11.4 Characteristics of EIS
Table 11.4 Important Terms Related to EIS Characteristics Drill Down Critical Success Factors (CSF) Monitored by five types of information 1. Key problem narratives 2. Highlight charts 3. Top-level financials 4. Key factors 5. Detailed responsibility reports

Navigation of Information Communication
Status Access Analysis by Built-in functions Integration with DSS products Intelligent agents Exception Reporting Use of Color Navigation of Information Communication

11.5 Comparing EIS and MIS Relationship between MIS and EIS (Figure 11.3) MIS is TPS based MIS typically lacks data integration across functional areas Differences (Table 11.5) MIS does not accommodate many users’ decision styles Often has slow response time Executive decision making is complex and multidimensional MIS usually designed to handle fairly structured, simpler configurations MIS do not usually combine data from multiple sources New advances are changing that

11.6 Comparing and Integrating EIS and DSS
Tables 11.6 and 11.7 compare the two systems Table Typical DSS definitions related to EIS Table Compares EIS and DSS EIS is part of decision support

Integrating EIS and DSS: An Executive Support System (ESS)
EIS output launches DSS applications Intelligent ESS Users' roles Commander Decision (Figure 11.4) Commander OLAP

Integrating EIS and Group Support Systems
EIS vendors - Easy interfaces with GDSS Some EIS built in Lotus Domino / Notes Comshare Inc. and Pilot Software, Inc. - Lotus Domino/Notes-based enhancements and Web/Internet/Intranet links

11.7 Hardware and Software EIS Hardware
Mainframe computers using graphics terminals Personal computers connected to a mainframe, a minicomputer, or a powerful RISC workstation Departmental LAN or a client/server architecture An enterprise-wide network, or on a client/server enterprise-wide system. Workstations perform high-speed graphics displays EIS (Enterprise Information System, Everybody’s Information System)

EIS Software Major Commercial EIS Software Vendors
Comshare Inc. (Ann Arbor, MI; Pilot Software Inc. (Cambridge, MA; Application Development Tools In-house components Comshare Commander tools Pilot Software’s Command Center Plus and Pilot Decision Support Suite

Trend for EIS Software Vendors with Third Party Vendors Producing Specialized EIS Applications
Comshare, Inc.’s Commander Series Commander FDC for consolidation, reporting, and analysis of financial information Commander Budget Plus for budget development and multidimensional planning Commander Prism for personal multidimensional analysis and modeling Arthur - a family of supply chain focused applications for retailing (planning, allocation and tracking) Boost Application Suite - a decision support solution for the consumer goods industry (Boost Sales and Margin Planning, Boost Sales Analysis)

More EIS Software Pilot Software, Inc.
Budget 2000 (with EPS, Inc.) is a budgeting application that includes the power of Pilot Decision Support Suite for budget preparation In Touch/2000 is a software agent that enables organizations to instantly create personal cubes (multidimensional databases), sales reports, budget forecasts and marketing plans Sales & Marketing Analysis Library of Pilot Decision Support Suite to perform detailed business reporting for sales and marketing professionals

Commercial EIS Software
Typically Includes Office Automation Electronic Mail Information Management Remote Information Access Information Analysis Representative List of EIS Software Products (Table W11.*)

11.8 EIS, Data Access, Data Warehousing, OLAP, Multidimensional Analysis, Presentation, and the Web
When data are delivered and viewed by an executive, by definition, the software is considered to be an EIS Data warehouses as data sources for EIS Advanced data visualization methods and hypermedia within EIS Comshare, Inc.’s Execu-View

Hypermedia over an Intranet via a Web Browser within the EIS
Comshare Commander DecisionWeb Internet Publishing module of the Pilot Decision Support Suite On-line Analytical Processing (OLAP) Tools Slice-and-dice multidimensional data cube

OLAP Packages DSS Web (MicroStrategy, Inc.)
Oracle Express Server (Oracle Corp.) Commander DecisionWeb (Comshare, Inc.) DataFountain (Dimensional Insight Inc.) Pilot Internet Publisher (Pilot Software, Inc.) WebOLAP (Information Advantage Inc.) Focus Fusion (Information Builders, Inc.) Business Objects Inc. (Business Objects) InfoBeaconWeb (Platinum Technology, Inc.) BrioQuery (Brio Technology Inc.) Data multidimensionality - In Touch/ Pilot personal cubes

Pilot Software, Inc.’s Decision Support Suite
Client/server, LAN-based, Windows-based software product (was Lightship) Pilot Desktop for ad hoc end-user data access Pilot Designer for development of executive information applications Pilot Analysis Server for access to multidimensional data models Pilot Discovery Server for data mining and predictive modeling Pilot Internet Publisher for publishing multidimensional data on the World Wide Web Pilot Sales & Marketing Library for a specific vertical market Excel Add-in - OLAP front end with Pilot Analysis Server

11.9 Enterprise EIS Tool for Enterprise Support
Executive-only EIS Enterprise-wide Information System Functional Management DSS Tools are Integrated with EIS EIS is Diffusing Lower into Organization Levels EIS = Enterprise Information System EIS = Everybody's Information System

11.10 EIS Implementation: Success or Failure

EIS Development Success Factors (Table 11.10)
Committed Executive Sponsor Correct Definition of Information Requirements Top Management Support

EIS Operational Success Factors (Table 11.11)
Deliver timely information Improve efficiency Provide accurate information Provide relevant information Ease of use Provide access to the status of the organization Provide improved communications An IS for upper management must fit with their decision styles

Motivations for Developing an EIS
Internal in nature Providing easier, faster access to information 80 % - Evolving approach Sequencing of the phases varies More successful development efforts include Initiation Definition of systems objectives Feasibility analysis

Determinates of EIS Acceptance
Rapid Development Time Staff Size EIS Age Not Ease of Use Not High Usage Not Many Features Not a Staff Close to Users

Factors Contributing to EIS Failures (Table 11.12)
Technology-related factors Support-related factors User-related factors Most EIS fail because they do not provide value for their high cost though EIS benefits are difficult to measure

Benefit and Cost Assessment Practices in EIS
Most Systems’ Realized Expected Benefits Were Lower than Expectation Greatest Problem - Information Contents, Not Information Delivery Issues

Unexpected EIS Benefits
Enhancements to the enterprise-wide information architecture Consolidation of data into warehouses Consolidation of analysis tools into OLAP methods Consistency of terminology across the enterprise

11.11 Including Soft Information in EIS
Soft information is fuzzy, unofficial, intuitive, subjective, nebulous, implied, and vague

Soft Information Used in Most EIS
Predictions, speculations, forecasts, and estimates (78.1%) Explanations, justifications, assessments, and interpretations (65.6%) News reports, industry trends, and external survey data (62.5%) Schedules and formal plans (50.0%) Opinions, feelings, and ideas (15.6%) Rumors, gossip, and hearsay (9.4%)

Soft Information Enhances EIS Value
More in the Future External news services Competitor information Ease of entering soft information

11.12 The Future of EIS and Research Issues
Toolbox for customized systems - Commander EIS LAN, Forest and Trees, and Pilot Decision Support Suite Multimedia support (databases, video and audio news feeds, GIS) Virtual Reality and 3-D Image Displays Merging of analytical systems with desktop publishing Client/server architecture Web-enabled EIS (Comshare Commander DecisionWeb, Pilot Decision Support Suite Internet Publishing module, SAS Institute Internet support enterprise software suite) Automated support and intelligent assistance Integration of EIS and Group Support Systems Global EIS

Research Issues Relationship between the executive sponsor’s organizational position and commitment level to EIS success Most important factors when selecting an operating sponsor? Prediction of EIS benefits in advance How EIS software affects the development process and system success Best staffing level and organizational structure for the builder/support staff

Most effective methods to identify executives' information requirements
Major EIS data management problems and their solutions Impact of soft data on EIS success Major problems associated with spread and evolution How to increase EIS functionality while maintaining ease of use Effective use of emerging technologies with EIS Most effective screen presentation formats

Current Trends in EIS More enterprise-wide EIS with greater decision support capabilities Integration with other software (Lotus Domino / Notes and World Wide Web) More intelligence - intelligent software agents

Other EIS Issues How to assess EIS benefits and costs
How to cluster EIS benefits depending on planned system uses How EIS diffuses throughout the organization How to perform screen management - creation, modification and elimination

Five Broad Categories of EIS Benefits
(Table W11.1) Help developers in design and development (Iyer and Aronson [1995])

11.13 Organizational DSS (ODSS)
Three Types of Decision Support Individual Group Organizational Hackathorn and Keen [1981]

Computer support is for
Organizational decision support focuses on an organizational task or activity involving a sequence of operations and actors Each individual's activities must mesh closely with other people's work Computer support is for Improving communication and coordination Problem solving

Definitions of ODSS A combination of computer and communication technology designed to coordinate and disseminate decision-making across functional areas and hierarchical layers in order that decisions are congruent with organizational goals and management's shared interpretation of the competitive environment (R. T. Watson [1990]) A DSS that is used by individuals or groups at several workstations in more than one organizational unit who make varied (interrelated but autonomous) decisions using a common set of tools (Carter et al. [1992])

A distributed decision support system (DDSS)
A distributed decision support system (DDSS). Not a manager's DSS, but supports the organization's division of labor in decision making (Swanson and Zmud [1990]) Apply the technologies of computers and communications to enhance the organizational decision-making process. Vision of technological support for group processes to the higher level of organizations (King and Star [1990])

Common Characteristics of ODSS (George [1991])
Focus is on an organizational task or activity or a decision that affects several organizational units or corporate problems Cuts across organizational functions or hierarchical layers Almost always involves computer-based technologies, and may involve communication technologies Can Integrate ODSS with Group DSS and Executive Information Systems Example: Egyptian Cabinet ODSS with EIS (DSS In Action 11.15)

11.14 The Architecture of ODSS
General Structure for ODSS (Figure 11.5) Major Differences ODSS Structure and Traditional DSS Case Management Component (CMS) Accessible by several users, in several locations, via LANs May have an intelligent component

Case Management Run a model many times Much output and many files
Helps the user manage the large numbers of similar runs Case = a specific run (scenario) of a computer model

CMS Main Functions 1. Record keeping of the model cases
2. Documenting the changes from one run to the next 3. Output comparison facilitation

11.15 Constructing an ODSS Formal, structured approach
Large, complex, system programming effort Combination of the SDLC and iterative approach

Phases 1. Getting started (a structured, organizational phase)
a) Needs assessment b) Getting management support c) Getting organized. Set up steering committee; identify project team members d) Getting a plan of action 2. Developing the conceptual design

4. Implementing and maintaining the system:
3. Developing the system a) Designing the physical system b) Developing the system's models and database 4. Implementing and maintaining the system: a) Installation b) Programming and updating system's modules (programs) c) Creating and updating the database d) Documenting the modules and database e) Training users

11.16 ODSS Example: The Enlisted Force Management System (EFMS)
Improve the effectiveness and efficiency Air Force staff managing the enlisted force in decision-making and information-processing Objective: to provide a group of airmen that is best able to support the missions and operational programs of the Air Force Iterative, continuous task Decisions about force structure, promotion policies, and the procurement, assignment, training, compensation, separation, and retirement of personnel Five major, independent organizational units (in three geographically locations) More than 125 person-years went into the EFMS development

The Elements of EFMS Model Base Screening and Impact Assessment Models
Authorization projection Grade allocation Aggregate planning, programming, and oversight Skills management Screening and Impact Assessment Models

Hardware and Databases
EFMS's mainframe computer DSS generator language, EXPRESS Access databases and models on PCs through EXPRESS

Databases from Output from another EFMS model
Data supplied by other branches of the Air Force External data The EFMS and other Air Force computer systems exchange data regularly

Important ODSS Implementation Issues
11.17 Implementing ODSS Important ODSS Implementation Issues 1. Steering committee for direction and control 2. Project team members join on an ad hoc basis 3. The System Management Office (SMO) 4. Conceptual design a) Design principles b) Functions to be supported c) Models d) Data requirements e) Hardware and software considerations f) Approach to implementation

Model Base Flexibility, adaptability and easy maintainability
Interlinked system of many small models

Database Coordination and integration
Specification of a common, consistent, easily accessed, centralized database All information from one module is automatically (instantaneously) available to others Internal and external data Many modules have their own databases

User Interface Common for all elements Menu driven Easy to learn
Easy to use Graphical User Interface (1993)

ODSS Data Understanding or defining the problem situation
Estimating the nature of the models Validating the models Running the models (input data) Database construction and data cleaning: 25 % - 30 % of effort

Integration and Networking
Many models and databases Integration of models, data, and knowledge can be complex Artificial Intelligence in ODSS Ideal - especially in CMS and machine learning (automatic rule induction)

Summary EIS serves the information needs of top executives and others
EIS provides rapid access to timely information at various levels of detail Very user friendly (user-seductive) ESS also has analysis capabilities Executives' work: finding problems (opportunities) and making decisions Finding the information needs of executives is very difficult

Methods: CSF (Critical Success Factors), BSP (Business System Planning), SBO (Strategic Business Objectives) and E/M (Ends/Means) Many EIS benefits are intangible Drill down Management by exception approach, centered on CSF, key performance indicators, and highlight charts In contrast to MIS, EIS has an overall organizational perspective and uses external data extensively Trend to integrate EIS and DSS tools EIS requires either a mainframe or a LAN

Constructing an EIS can be difficult. Vendors or consultants
EIS development tools Intranets to deliver information to executives Web-enabled EIS EIS success - many factors ranging from appropriate technology to managing organizational resistance The executive sponsor is crucial for the success of an EIS EIS failure - no value provided An EIS must fit the executives’ decision styles

Multidimensional analysis and presentation
Access to database information by end-users, enterprise-wide EIS technology and use diffusing to lower levels of management Data warehouses and client/server front end environments make an EIS a useful tool for end users EIS can provide valuable soft information Organizational DSS (ODSS) deals with decision making across functional areas and hierarchical organizational layers

ODSS includes a case management system (CMS)
ODSS is used by individuals and groups and operates in a distributed environment ODSS deals with organizational tasks ODSS for similar, repetitive situations involves a case management component ODSS is frequently integrated with EIS and/or GDSS ODSS built using both traditional SDLC and prototyping Data and databases are critical to the success of ODSS ODSS usually use several quantitative and qualitative models

1. Explain how Hertz added an EIS that is used as a front end to the DSS 2. Why did the new DSS not satisfy the executives’ information needs? 3. Why was it so important for the new system to provide information that conformed to the way executives at Hertz worked? Do you think that the system would have been acceptable otherwise? Why or why not? 4. What capabilities did the PCs bring to the EIS? 5. Why is it important for Hertz to be able to monitor competitors’ marketing strategies in real time?

Appendix W11-A: The Client/Server Architecture and Enterprise Computing
Approach to organizing PCs, local area networks, and possibly mainframes, into a flexible, effective, and efficient system

C/S Characteristics The clients are PCs or workstations, attached to a network. Clients access network resources The user interfaces directly with the client (via GUI) Servers provide shared resources to several clients A server provides clients with service capabilities (databases, large disk drives, or communications) Servers can be workstations, mainframes, minicomputers, and/or LAN PC devices

A client forms one or more queries or commands, in a predefined language such as SQL, for presentation to the server Clients can send queries or commands to the servers Server transmits results to client's screen Typical servers: database server, file server, print server, image-processing server, computing server, and communication server (Web server) Server only reacts to client's requests Servers can communicate with each other Tasks are split into two: front-end portion (client), and back-end portion (server(s))

Client / Server Computing
Changes the way people work People are empowered to access databases

Client/Server Applications Categories
Messaging applications, such as electronic mail Disseminating a database among several computer networks Offering file- or peripheral-sharing, or remote computer access Processing-intensive applications where jobs are divided into tasks, each of which is performed by a different computer

Enterprise-wide Client/Server Architecture
Computing systems that involve an entire organization Architecture for an integrated computer system to serve the business needs of the enterprise Technological framework that contains multiple applications, hardware, databases, networks, and management tools, usually from multiple vendors Requires a consensus on a set of standards ranging from operating systems to telecommunication protocols Requires a consensus on a common open management platform and a strong organizational commitment

Major Benefits of Enterprise Computing
Reliable and responsive service Smooth incorporation of new client/server solutions with existing approaches Frequent and rapid changes, and increasing complexity Greater optimization of network and system resources Automation of management processes Network and data security

Enterprise client/server architecture provides total integration of departmental and corporate IS resources Provides better control and security over data in a distributed environment IS organizations can maximize the value of information by increasing its availability. Enterprise client/server computing empowers organizations to Reengineer business processes Distribute transactions to streamline operations Provide better and newer services to customers

Chapter 12: Fundamentals of Expert Systems
12.1 Opening Vignette: CATS-1 at General Electric The Problem General Electric's (GE) Top Locomotive Field Service Engineer was Nearing Retirement

Traditional Solution: Apprenticeship
Good Short-term Solution BUT GE Wanted A more effective and dependable way to disseminate expertise To prevent valuable knowledge from retiring To minimize extensive travel or moving the locomotives

The Expert System Solution
To MODEL the way a human troubleshooter works Months of knowledge acquisition 3 years of prototyping A novice engineer or technician can perform at an expert’s level On a personal computer Installed at every railroad repair shop served by GE

12.2 (ES) Introduction Expert System: from the term knowledge-based expert system An Expert System is a system that employs human knowledge captured in a computer to solve problems that ordinarily require human expertise ES imitate the expert’s reasoning processes to solve specific problems

12.3 History of Expert Systems
1. Early to Mid-1960s One attempt: the General-purpose Problem Solver (GPS) General-purpose Problem Solver (GPS) A procedure developed by Newell and Simon [1973] from their Logic Theory Machine - Attempted to create an "intelligent" computer Predecessor to ES Not successful, but a good start

2. Mid-1960s: Special-purpose ES programs
DENDRAL MYCIN Researchers recognized that the problem-solving mechanism is only a small part of a complete, intelligent computer system General problem solvers cannot be used to build high performance ES Human problem solvers are good only if they operate in a very narrow domain Expert systems must be constantly updated with new information The complexity of problems requires a considerable amount of knowledge about the problem area

Limited Success Because
3. Mid 1970s Several Real Expert Systems Emerge Recognition of the Central Role of Knowledge AI Scientists Develop Comprehensive knowledge representation theories General-purpose, decision-making procedures and inferences Limited Success Because Knowledge is Too Broad and Diverse Efforts to Solve Fairly General Knowledge-Based Problems were Premature

BUT Key Insight Several knowledge representations worked
The power of an ES is derived from the specific knowledge it possesses, not from the particular formalisms and inference schemes it employs

4. Early 1980s ES Technology Starts to go Commercial
XCON XSEL CATS-1 Programming Tools and Shells Appear EMYCIN EXPERT META-DENDRAL EURISKO About 1/3 of These Systems Are Very Successful and Are Still in Use

Latest ES Developments
Many tools to expedite the construction of ES at a reduced cost Dissemination of ES in thousands of organizations Extensive integration of ES with other CBIS Increased use of expert systems in many tasks Use of ES technology to expedite IS construction

The object-oriented programming approach in knowledge representation
Complex systems with multiple knowledge sources, multiple lines of reasoning, and fuzzy information Use of multiple knowledge bases Improvements in knowledge acquisition Larger storage and faster processing computers The Internet to disseminate software and expertise.

12.4 Basic Concepts of Expert Systems
Expertise Experts Transferring Expertise Inferencing Rules Explanation Capability

Expertise Expertise is the extensive, task-specific knowledge acquired from training, reading and experience Theories about the problem area Hard-and-fast rules and procedures Rules (heuristics) Global strategies Meta-knowledge (knowledge about knowledge) Facts Enables experts to be better and faster than nonexperts

IS In Focus 12.2: Some Facts about Expertise
Expertise is usually associated with a high degree of intelligence, but not always with the smartest person Expertise is usually associated with a vast quantity of knowledge Experts learn from past successes and mistakes Expert knowledge is well-stored, organized and retrievable quickly from an expert Experts have excellent recall

Experts Degrees or levels of expertise
Nonexperts outnumber experts often by 100 to 1

Human Expert Behaviors
Recognizing and formulating the problem Solving the problem quickly and properly Explaining the solution Learning from experience Restructuring knowledge Breaking rules Determining relevance Degrading gracefully (awareness of limitations)

Transferring Expertise
Objective of an expert system To transfer expertise from an expert to a computer system and Then on to other humans (nonexperts) Activities Knowledge acquisition Knowledge representation Knowledge inferencing Knowledge transfer to the user Knowledge is stored in a knowledge base

Regarding the Problem Domain
Two Knowledge Types Facts Procedures (Usually Rules) Regarding the Problem Domain

Inferencing Reasoning (Thinking)
The computer is programmed so that it can make inferences Performed by the Inference Engine

Rules IF-THEN-ELSE Explanation Capability
By the justifier, or explanation subsystem ES versus Conventional Systems (Table 12.1)

12.5 Structure of Expert Systems
Development Environment Consultation (Runtime) Environment (Figure 12.2)

Three Major ES Components
Knowledge Base Inference Engine User Interface

All ES Components Knowledge Acquisition Subsystem Knowledge Base
Inference Engine User User Interface Blackboard (Workplace) Explanation Subsystem (Justifier) Knowledge Refining System Most ES do not have a Knowledge Refinement Component (See Figure 12.2)

Knowledge Acquisition Subsystem
Knowledge acquisition is the accumulation, transfer and transformation of problem-solving expertise from experts and/or documented knowledge sources to a computer program for constructing or expanding the knowledge base Requires a knowledge engineer

Knowledge Base The knowledge base contains the knowledge necessary for understanding, formulating, and solving problems Two Basic Knowledge Base Elements Facts Special heuristics, or rules that direct the use of knowledge Knowledge is the primary raw material of ES Incorporated knowledge representation

Inference Engine The brain of the ES
The control structure or the rule interpreter Provides a methodology for reasoning

Inference Engine Major Elements
Interpreter Scheduler Consistency Enforcer

User Interface Language processor for friendly, problem-oriented communication NLP, or menus and graphics

Blackboard (Workplace)
Area of working memory to Describe the current problem Record Intermediate results Records Intermediate Hypotheses and Decisions 1. Plan 2. Agenda 3. Solution

Explanation Subsystem (Justifier)
Traces responsibility and explains the ES behavior by interactively answering questions Why? How? What? (Where? When? Who?) Knowledge Refining System Learning for improving performance

12.6 The Human Element in Expert Systems
Builder and User Expert and Knowledge engineer. The Expert Has the special knowledge, judgment, experience and methods to give advice and solve problems Provides knowledge about task performance

The Knowledge Engineer
Helps the expert(s) structure the problem area by interpreting and integrating human answers to questions, drawing analogies, posing counterexamples, and bringing to light conceptual difficulties Usually also the System Builder

The User Possible Classes of Users
A non-expert client seeking direct advice - the ES acts as a Consultant or Advisor A student who wants to learn - an Instructor An ES builder improving or increasing the knowledge base - a Partner An expert - a Colleague or Assistant The Expert and the Knowledge Engineer Should Anticipate Users' Needs and Limitations When Designing ES

Other Participants System Builder Tool Builder Vendors Support Staff
Network Expert (Figure 12.3)

12.7 How Expert Systems Work
Major Activities of ES Construction and Use Development Consultation Improvement

ES Development Construction of the knowledge base
Knowledge separated into Declarative (factual) knowledge and Procedural knowledge Construction (or acquisition) of an inference engine, a blackboard, an explanation facility, and any other software Determine appropriate knowledge representations

Participants Domain Expert Knowledge Engineer and
(Possibly) Information System Analysts and Programmers

ES Shell Includes All Generic Components of an ES No Knowledge
EMYCIN from MYCIN RESOLVER (was EXSYS)

Consultation Deploy ES to Users (Typically Novices)
ES Must be Very Easy to Use ES Improvement By Rapid Prototyping

12.8 An Expert System at Work
See text or do a demo in Resolver (Exsys)

12.9 Problem Areas Addressed by Expert Systems
Generic Categories of Expert Systems (Table 12.2) Interpretation systems Prediction systems Diagnostic systems Design systems Planning systems Monitoring systems Debugging systems Repair systems Instruction systems Control systems Example in Human Resource Management (Table 12.3)

12.10 Benefits of Expert Systems
Major Potential ES Benefits Increased Output and Productivity Decreased Decision Making Time Increased Process(es) and Product Quality Reduced Downtime Capture of Scarce Expertise Flexibility Easier Equipment Operation Elimination of the Need for Expensive Equipment

Operation in Hazardous Environments
Accessibility to Knowledge and Help Desks Increased Capabilities of Other Computerized Systems Integration of Several Experts' Opinions Ability to Work with Incomplete or Uncertain Information Provide Training Enhancement of Problem Solving and Decision Making Improved Decision Making Processes Improved Decision Quality Ability to Solve Complex Problems Knowledge Transfer to Remote Locations Enhancement of Other CBIS (provide intelligent capabilities to large CBIS)

Lead to Improved decision making
Improved products and customer service A sustainable strategic advantage Some may even enhance the organization’s image

12.11 Problems and Limitations of Expert Systems
Knowledge is not always readily available Expertise can be hard to extract from humans Each expert’s approach may be different, yet correct Hard, even for a highly skilled expert, to work under time pressure Users of expert systems have natural cognitive limits ES work well only in a narrow domain of knowledge

Most experts have no independent means to validate their conclusions
The vocabulary of experts is often limited and highly technical Knowledge engineers are rare and expensive Lack of trust by end-users Knowledge transfer is subject to a host of perceptual and judgmental biases ES may not be able to arrive at conclusions ES sometimes produce incorrect recommendations

Longevity of Commercial Expert Systems
(Gill [1995]) Only about one-third survived five years Generally ES Failed Due to Managerial Issues Lack of system acceptance by users Inability to retain developers Problems in transitioning from development to maintenance Shifts in organizational priorities Proper management of ES development and deployment could resolve most

12.12 Expert System Success Factors
Two of the Most Critical Factors Champion in Management User Involvement and Training Plus The level of knowledge must be sufficiently high There must be (at least) one cooperative expert The problem to be solved must be qualitative (fuzzy) not quantitative The problem must be sufficiently narrow in scope The ES shell must be high quality, and naturally store and manipulate the knowledge

Need end-user training programs
A friendly user interface The problem must be important and difficult enough Need knowledgeable and high quality system developers with good people skills The impact of ES as a source of end-users’ job improvement must be favorable. End user attitudes and expectations must be considered Management support must be cultivated. Need end-user training programs The organizational environment should favor new technology adoption

For Success 1. Business applications justified by strategic impact (competitive advantage) 2. Well-defined and structured applications

12.13 Types of Expert Systems
Expert Systems Versus Knowledge-based Systems Rule-based Expert Systems Frame-based Systems Hybrid Systems Model-based Systems Ready-made (Off-the-Shelf) Systems Real-time Expert Systems

12.14 Expert Systems and the Internet/Intranets/Web
1. Use of ES on the Net 2. Support ES (and other AI methods)

Using ES on the Net To provide knowledge and advice to large numbers of users Help desks Knowledge acquisition Spread of multimedia-based expert systems (Intelimedia systems) Support ES and other AI technologies provide to the Internet/Intranet (Table 12.4)

Summary Expert systems imitate the reasoning process of experts
ES predecessor: the General-purpose Problem Solver (GPS). Failed - ignored the importance of specific knowledge The power of an ES is derived from its specific knowledge Not from its particular knowledge representation or inference scheme Expertise is a task-specific knowledge acquired from training, reading, and experience Experts can make fast and good decisions regarding complex situations

Most of the knowledge in organizations is possessed by a few experts
Expert system technology attempts to transfer knowledge from experts and documented sources to the computer and make it available to nonexperts Expert systems involve knowledge processing, not data processing Inference engine provides ES reasoning capability The knowledge in ES is separated from the inferencing Expert systems provide limited explanation capabilities A distinction is made between a development environment (building an ES) and a consultation environment (using an ES)

The major components of an ES are the knowledge acquisition subsystem, knowledge base, inference engine, blackboard, user interface and explanation subsystem The knowledge engineer captures the knowledge from the expert and programs it into the computer Although the major user of the ES is a nonexpert, there may be other users (such as students, ES builders, experts) Knowledge can be declarative (facts) or procedural Expert systems are improved in an iterative manner using a process called rapid prototyping

The ten generic categories of ES: interpretation, prediction, diagnosis, design, planning, monitoring, debugging, repair, instruction and control Expert systems can provide many benefits Most ES failures are due to non-technical problems (managerial support and end user training) Although there are several limitations to using expert systems, some will disappear with improved technology

ES success factors Expert systems can make mistakes Distinction between expert systems, and knowledge systems Some ES are ready-made Some expert systems provide advice in a real-time mode Expertise may be provided over the Internet / Intranets via ES ES and AI provide support to the Internet / Intranets

CASE APPLICATION 12.1: Gate Assignment Display System (GADS)
Case Questions 1. Why is the gate assignment task so complex? 2. Why is GADS considered a real-time ES? 3. What are the major benefits of the ES over the manual system? (Prepare a detailed list.) 4. What measures were taken to increase the reliability of the system and why were they needed?

CASE APPLICATION 12.2: Expert System in Building Construction (EXSOFS)

CASE APPLICATION W12.1: DustPro--Environmental Control in Mines

APPENDIX 12-B: Classic Expert Systems
MYCIN X-CON

Stanford Medical School in the 1970s by
1. MYCIN To aid physicians in diagnosing meningitis and other bacterial blood infections and to prescribe treatment To aid physicians during a critical hour period after the detection of symptoms, a time when much of the decision making is imprecise Early diagnosis and treatment can save a patient from brain damage or even death Stanford Medical School in the 1970s by Dr. Edward H. Shortliffe

MYCIN Features Rule-based knowledge representation Probabilistic rules
Backward chaining method Explanation User-friendly system

2. XCON (Expert VAX System Configuration and Mass Customization)
Digital Equipment Corp. (DEC) minicomputer system configuration Manually: Complex task, many errors, not cost effective Cost savings estimated at about $15 million / year Literature: Over $40 million / year later

Chapter 13: Knowledge Acquisition and Validation
13.1 Opening Vignette: American Express Improves Approval Selection with Machine Learning The Problem: Loan Approval 85 to 90 % Predicted Accurately 10 to 15 % in Gray Area Accuracy of Loan Officer’s Gray Area Decisions were at most 50 %

The Solution ES with Knowledge Acquisition Method of Machine Learning
Rule Induction Method Gray Area: Induced Decision Tree Correctly Predicted 70 % Induced Rules Explain Why Rejected

13.2 Knowledge Engineering
The art of bringing the principles and tools of AI research to bear on difficult applications problems requiring experts' knowledge for their solutions The technical issues of acquiring this knowledge, representing it and using it appropriately to construct and explain lines-of-reasoning are important problems in the design of knowledge-based systems The art of constructing intelligent agents is both part of and an extension of the programming art It is the art of building complex computer programs that represent and reason with knowledge of the world (Feigenbaum and McCorduck [1983])

We use the Narrow Definition
Narrow perspective: knowledge engineering deals with knowledge acquisition, representation, validation, inferencing, explanation and maintenance Wide perspective: KE describes the entire process of developing and maintaining AI systems We use the Narrow Definition Involves the cooperation of human experts Synergistic effect

Knowledge Engineering Process Activities
Knowledge Acquisition Knowledge Validation Knowledge Representation Inference Explanation and Justification (Figure 13.1)

13.3 Scope of Knowledge Knowledge acquisition is the extraction of knowledge from sources of expertise and its transfer to the knowledge base and sometimes to the inference engine Knowledge is a collection of specialized facts, procedures and judgment rules (Figure 13.2)

Knowledge Sources Documented (books, manuals, etc.)
Undocumented (in people's minds) From people, from machines Knowledge Acquisition from Databases Knowledge Acquisition Via the Internet

Knowledge Levels Shallow Knowledge (surface) Deep knowledge
Can implement a computerized representation that is deeper than shallow knowledge (Example: Figure 13.3) Special knowledge representation methods (semantic networks and frames) to allow the implementation of deeper-level reasoning (abstraction and analogy): important expert activity Represent objects and processes of the domain of expertise at this level The relationships among objects are important

Major Categories of Knowledge
Declarative Knowledge Procedural Knowledge Metaknowledge

Declarative Knowledge
Descriptive Representation of Knowledge Expressed in a factual statement Shallow Important in the initial stage of knowledge acquisition

Procedural Knowledge Considers the manner in which things work under different sets of circumstances Includes step-by-step sequences and how-to types of instructions May also include explanations Involves automatic response to stimuli May also tell how to use declarative knowledge and how to make inferences

Descriptive knowledge relates to a specific object
Descriptive knowledge relates to a specific object. Includes information about the meaning, roles, environment, resources, activities, associations and outcomes of the object Procedural knowledge relates to the procedures employed in the problem-solving process

Knowledge about Knowledge
Metaknowledge Knowledge about Knowledge In ES, Metaknowledge refers to knowledge about the operation of knowledge-based systems Its reasoning capabilities

13.4 Difficulties in Knowledge Acquisition
Problems in Transferring Knowledge Expressing the Knowledge Transfer to a Machine Number of Participants Structuring the Knowledge

Other Reasons Experts may lack time or not cooperate
Testing and refining knowledge is complicated Poorly defined methods for knowledge elicitation System builders may collect knowledge from one source, but the relevant knowledge may be scattered across several sources Collect documented knowledge rather than use experts The knowledge collected may be incomplete Difficult to recognize specific knowledge when mixed with irrelevant data Experts may change their behavior when observed and/or interviewed Problematic interpersonal communication between the knowledge engineer and the expert

Overcoming the Difficulties
Knowledge acquisition tools with ways to decrease the representation mismatch between the human expert and the program (“learning by being told”) Simplified rule syntax Natural language processor to translate knowledge to a specific representation Impacted by the role of the three major participants Knowledge Engineer Expert End user

Computer-aided knowledge acquisition tools
Critical The ability and personality of the knowledge engineer Must develop a positive relationship with the expert The knowledge engineer must create the right impression Computer-aided knowledge acquisition tools Extensive integration of the acquisition efforts

Required Skills and Characteristics of Knowledge Engineers
Computer skills Tolerance and ambivalence Effective communication abilities Broad educational background Advanced, socially sophisticated verbal skills Fast-learning capabilities (of different domains) Must understanding organizations and individuals Wide experience in knowledge engineering Intelligence Empathy and patience Persistence Logical thinking Versatility and inventiveness Self-confidence

13.5 Methods of Knowledge Acquisition: An Overview
Manual Semiautomatic Automatic (Computer Aided)

Manual Methods - Structured Around Interviews
Process (Figure 13.4) Interviewing Structured Semistructured Unstructured Tracking the Reasoning Process Observing Manual methods: slow, expensive and sometimes inaccurate

Semiautomatic Methods
Support Experts Directly (Figure 13.5) Help Knowledge Engineers

Automatic Methods Expert’s and/or the knowledge engineer’s roles are minimized (or eliminated) Induction Method (Figure 13.6)

Knowledge Modeling The knowledge model views knowledge acquisition as the construction of a model of problem-solving behavior-- a model in terms of knowledge instead of representations Can reuse models across applications

13.6 Interviews Most Common Knowledge Acquisition: Face-to-face Interview Analysis Types of Interviews Unstructured (Informal) Structured

Unstructured Interviews
Seldom provides complete or well-organized descriptions of cognitive processes because The domains are generally complex The experts usually find it very difficult to express some more important knowledge Domain experts may interpret the lack of structure as requiring little preparation Data acquired are often unrelated, exist at varying levels of complexity, and are difficult for the knowledge engineer to review, interpret and integrate Few knowledge engineers can conduct an efficient unstructured interview

The knowledge engineer slowly learns about the problem
Then can build a representation of the knowledge Knowledge acquisition involves Uncovering important problem attributes Making explicit the expert’s thought process

Unstructured Interviews
Most Common Variations Talkthrough Teachthrough Readthrough

Structured Interviews
Systematic goal-oriented process Forces an organized communication between the knowledge engineer and the expert Procedural Issues in Structuring an Interview (Table 13.1) Interpersonal communication and analytical skills are important

Interviews - Summary Are important techniques
Must be planned carefully Results must be verified and validated Are sometimes replaced by tracking methods Can supplement tracking or other knowledge acquisition methods

Before a knowledge engineer interviews the expert(s)
Recommendation Before a knowledge engineer interviews the expert(s) 1. Interview a less knowledgeable (minor) expert Helps the knowledge engineer Learn about the problem Learn its significance Learn about the expert(s) Learn who the users will be Understand the basic terminology Identify readable sources 2. Next read about the problem 3. Then, interview the expert(s) (much more effectively)

13.7 Tracking Methods Techniques that attempt to track the reasoning process of an expert From cognitive psychology Most common formal method: Protocol Analysis

Advantages and Limitations (Table 13.3)
Protocol Analysis Protocol: a record or documentation of the expert's step-by-step information processing and decision-making behavior The expert performs a real task and verbalizes his or her thought process (think aloud) Summary (Table 13.2) Advantages and Limitations (Table 13.3)

13.8 Observations and Other Manual Methods
Observe the Expert Work

Other Manual Methods Case analysis Critical incident analysis
Discussions with the users Commentaries Conceptual graphs and models Brainstorming Prototyping Multidimensional scaling Johnson's hierarchical clustering Performance review

13.9 Expert-driven Methods
Knowledge Engineers Typically Lack Knowledge About the Domain Are Expensive May Have Problems Communicating With Experts Knowledge Acquisition May be Slow, Expensive and Unreliable Can Experts Be Their Own Knowledge Engineers?

Approaches to Expert-Driven Systems
Manual Computer-Aided (Semiautomatic)

Manual Method: Expert's Self-reports
Problems with Experts’ Reports and Questionnaires 1. Requires the expert to act as knowledge engineer 2. Reports are biased 3. Experts often describe new and untested ideas and strategies 4. Experts lose interest rapidly 5. Experts must be proficient in flowcharting 6. Experts may forget certain knowledge 7. Experts are likely to be vague

Benefits May provide useful preliminary knowledge discovery and acquisition Computer support can eliminate some limitations

Computer-aided Approaches
To reduce or eliminate the potential problems REFINER+ - case-based system TIGON - to detect and diagnose faults in a gas turbine engine Other Visual modeling techniques New machine learning methods to induce decision trees and rules Tools based on repertory grid analysis

13.10 Repertory Grid Analysis (RGA)
Techniques, derived from psychology Use the classification interview Fairly structured Primary Method: Repertory Grid Analysis (RGA)

The Grid Based on Kelly's model of human thinking: Personal Construct Theory (PCT) Each person is a "personal scientist" seeking to predict and control events by Forming Theories Testing Hypotheses Analyzing Results of Experiments Knowledge and perceptions about the world (a domain or problem) are classified and categorized by each individual as a personal, perceptual model Each individual anticipates and then acts

How RGA Works 1. The expert identifies the important objects in the domain of expertise (interview) 2. The expert identifies the important attributes 3. For each attribute, the expert is asked to establish a bipolar scale with distinguishable characteristics (traits) and their opposites [see Table 13.4] 4. The interviewer picks any three of the objects and asks: What attributes and traits distinguish any two of these objects from the third? Translate answers on a scale of 1-3 (or 1-5)

Step 4 continues for several triplets of objects
Answers recorded in a Grid (Table 13.5) Expert may change the ratings inside box Can use the grid for recommendations

Use of RGA in Expert Systems - Tools
Expertise Transfer System (ETS) (Now in AQUINAS) AQUINAS KRITON

Other RGA Tools PCGRID (PC-based) WebGrid Circumgrids

13.11 Supporting the Knowledge Engineer
Knowledge Acquisition Aids Special Languages Editors and Interfaces Explanation Facility Revision of the Knowledge Base Pictorial Knowledge Acquisition (PIKA)

Integrated Knowledge Acquisition Aids
PROTÉGÉ-II KSM ACQUIRE KADS (Knowledge Acquisition and Documentation System) Front-end Tools Knowledge Analysis Tool (KAT) NEXTRA (in Nexpert Object)

13.12 Machine Learning: Rule Induction, Case-based Reasoning, Neural Computing, and Intelligent Agents Manual and semiautomatic elicitation methods: slow and expensive Other Deficiencies Frequently weak correlation between verbal reports and mental behavior Sometimes experts cannot describe their decision making process System quality depends too much on the quality of the expert and the knowledge engineer The expert does not understand ES technology The knowledge engineer may not understand the business problem Can be difficult to validate acquired knowledge

Computer-aided Knowledge Acquisition, or Automated Knowledge Acquisition Objectives
Increase the productivity of knowledge engineering Reduce the required knowledge engineer’s skill level Eliminate (mostly) the need for an expert Eliminate (mostly) the need for a knowledge engineer Increase the quality of the acquired knowledge

Automated Knowledge Acquisition (Machine Learning)
Rule Induction Case-based Reasoning Neural Computing Intelligent Agents

Machine Learning Knowledge Discovery and Data Mining
Include Methods for Reading Documents and Inducing Knowledge (Rules) Other Knowledge Sources (Databases) Tools KATE-Induction CN-2

Automated Rule Induction
Induction: Process of Reasoning from Specific to General In ES: Rules Generated by a Computer Program from Cases Interactive Induction

Case-based Reasoning (CBR)
For Building ES by Accessing Problem-solving Experiences for Inferring Solutions for Solving Future Problems Cases and Resolutions Constitutes a Knowledge Base

Neural Computing Fairly Narrow Domains with Pattern Recognition
Requires a Large Volume of Historical Cases

Intelligent Agents for Knowledge Acquisition
Led to KQML (Knowledge Query and Manipulation Language) for Knowledge Sharing KIF, Knowledge Interchange Format (Among Disparate Programs)

13.13 Selecting an Appropriate Knowledge Acquisition Method
Ideal Knowledge Acquisition System Objectives Direct interaction with the expert without a knowledge engineer Applicability to virtually unlimited problem domains Tutorial capabilities Ability to analyze work in progress to detect inconsistencies and gaps in knowledge Ability to incorporate multiple knowledge sources A user friendly interface Easy interface with different expert system tools Hybrid Acquisition - Another Approach

13.14 Knowledge Acquisition from Multiple Experts
Major Purposes of Using Multiple Experts Better understand the knowledge domain Improve knowledge base validity, consistency, completeness, accuracy and relevancy Provide better productivity Identify incorrect results more easily Address broader domains To handle more complex problems and combine the strengths of different reasoning approaches Benefits And Problems With Multiple Experts (Table 13.7)

Handling Multiple Expertise
Blend several lines of reasoning through consensus methods Use an analytical approach (group probability) Select one of several distinct lines of reasoning Automate the process Decompose the knowledge acquired into specialized knowledge sources

13.15 Validation and Verification of the Knowledge Base
Quality Control Evaluation Validation Verification

Evaluation Validation Verification
Assess an expert system's overall value Analyze whether the system would be usable, efficient and cost-effective Validation Deals with the performance of the system (compared to the expert's) Was the “right” system built (acceptable level of accuracy?) Verification Was the system built "right"? Was the system correctly implemented to specifications?

Dynamic Activities Repeated each prototype update
For the Knowledge Base Must have the right knowledge base Must be constructed properly (verification) Activities and Concepts In Performing These Quality Control Tasks (Table 13.8)

Method for Validating ES
Test 1. The extent to which the system and the expert decisions agree 2. The inputs and processes used by an expert compared to the machine 3. The difference between expert and novice decisions (Sturman and Milkovich [1995])

13.16 Analyzing, Coding, Documenting, and Diagramming
Method of Acquisition and Representation 1. Transcription 2. Phrase Indexing 3. Knowledge Coding 4. Documentation (Wolfram et al. [1987])

Knowledge Diagramming
Graphical, hierarchical, top-down description of the knowledge that describes facts and reasoning strategies in ES Types Objects Events Performance Metaknowledge Describes the linkages and interactions among knowledge types Supports the analysis and planning of subsequent acquisitions Called conceptual graphs (CG) Useful in analyzing acquired knowledge

13.17 Numeric and Documented Knowledge Acquisition
Acquisition of Numeric Knowledge Special approach needed to capture numeric knowledge Acquisition of Documented Knowledge Major Advantage: No Expert To Handle a Large or Complex Amount of Information New Field: New Methods That Interpret Meaning to Determine Rules Other Knowledge Forms (Frames for Case-Based Reasoning)

13.18 Knowledge Acquisition and the Internet/Intranet
Hypermedia (Web) to Represent Expertise Naturally Natural Links can be Created in the Knowledge CONCORDE: Hypertext-based Knowledge Acquisition System Hypertext links are created as knowledge objects are acquired

The Internet/Intranet for Knowledge Acquisition
Electronic Interviewing Experts can Validate and Maintain Knowledge Bases Documented Knowledge can be accessed The Problem: Identify relevant knowledge (intelligent agents) Many Web Search Engines have intelligent agents Data Fusion Agent for multiple Web searches and organizing Automated Collaborative Filtering (ACF) statistically matches peoples’ evaluations of a set of objects

New Developments WebGrid: Web-based Knowledge Elicitation Approaches
Plus Information Structuring in Distributed Hypermedia Systems

13.19 Induction Table Example
Induction tables (knowledge maps) focus the knowledge acquisition process Choosing a site for a hospital clinic facility (Section 13.6: Table 13.9)

Row 1: Factors Row 2: Valid Factor Values and Choices (last column) Table leads to the prototype ES Each row becomes a potential rule Induction tables can be used to encode chains of knowledge

Class Exercise: Animals
Knowledge Acquisition Create Induction Table I am thinking of an animal! Question: Does it have a long neck? If yes THEN Guess that it is a giraffe. IF not a giraffe, then ask for a question to distinguish between the two. Is it YES or NO for a giraffe? Fill in the new Factor, Values and Rule. IF no, THEN What is the animal? and fill in the new rule. Continue with all questions You will build a table very quickly

Summary Knowledge engineering: acquisition, representation, reasoning (inference) and explanation Knowledge available from many sources, some documented, some not Knowledge can be shallow or deep Knowledge acquisition is difficult Knowledge acquisition methods: manual, semi-automated and automated Primary manual approach is interviewing: completely unstructured to highly structured

Experts’ reasoning process can be tracked by several methods (protocol analysis)
Observation of experts in action is usually limited New manual and/or computerized tools for self-knowledge acquisition Repertory grid analysis (RGA) is the most applied method of semiautomated interviews Many productivity tools for knowledge acquisition Rule induction examines historical cases and generates rules

Rule induction can be used by a system engineer, an expert, other system builder
Benefits, limitations and problems in several experts Major methods of multiple experts: consensus methods, analytical approaches, selection of an appropriate line of reasoning, process automation and blackboard systems Knowledge base validation and verification - critical ES implementation success factors Many measures to determine knowledge validity

Automated knowledge acquisition methods are easier to validate and verify
Knowledge collected must be analyzed and coded prior to its representation Case-based reasoning, neural computing, intelligent agents and other machine learning tools can enhance the task of knowledge acquisition The Internet/intranet is expanding the methods for performing knowledge acquisition

1. Why was the system’s accuracy so much better than the human loan officers? 2. Why was an explanation facility so important here? 3. Why do you think the ES predicted much more accurately than the Loan Officers did? 4. For the ES, what are the implications when there are changes in the economic climate? Explain. 5. Why do you think so many test cases were needed? 6. Could the ES be used to train Loan Officers? Explain. 7. Because the rule induced decision tree is much more accurate, comb the literature and try to estimate how much money can be saved by denying predicted faulted loans.

Interview a Decision Maker
Group Exercises Interview a Decision Maker 1. Which interviewing technique? 2. What problems encountered? 3. What problems occurred because a group interview, not a dialog? 4. What personality traits helped and hindered the group, and the decision maker? Why? 5. If the decision maker actually makes a decision while you are interviewing him/her (or shortly after), how he/she reached the conclusion. 6. Report your findings in a report. Compare results to other groups’.

Tables for the Exercises
For Exercises 1, 2, 4 Table for Exercise 1 Table for Exercise 2 Table for Exercise 4

Chapter 14: Knowledge Representation
Once knowledge is acquired, it must be organized for later use

14.1 Opening Vignette: Pitney Bowes Expert System Diagnoses Repair Problems and Saves Millions
The Situation Postage meter repair Varying levels of expertise, and less consistency in repairs Many parts changed unnecessarily

The G2 Solution Expert system G2 (Gensym Corp.) provides consistent advice to operators diagnosing and repairing 24,000 postage meters a year Supports 30 repair personnel Reduces repair time and unnecessary parts replacement Knowledge server: captured and distributes expert knowledge Graphic format to portray knowledge

G2 Benefits Over $1 million savings in two years (projected)
Product cost reduced 23% Faster training and improved consistency Provides competitive advantage

14.2 Introduction A good knowledge representation ‘naturally’ represents the problem domain An unintelligible knowledge representation is wrong Most artificial intelligence systems consist of Knowledge Base Inference Mechanism (Engine)

Many knowledge representation schemes
Knowledge Base Forms the system's intelligence source Inference mechanism uses to reason and draw conclusions Inference mechanism: Set of procedures that are used to examine the knowledge base to answer questions, solve problems or make decisions within the domain Many knowledge representation schemes Can be programmed and stored in memory Are designed for use in reasoning Major knowledge representation schemas: Production rules Frames

Knowledge Representation and the Internet
Hypermedia documents to encode knowledge directly Hyperlinks Represent Relationships MIKE (Model-based and Incremental Knowledge Engineering Formal model of expertise: KARL Specification Language Semantic networks: Ideally suited for hypermedia representation Web-based Distributed Expert System (Ex-W-Pert System) for sharing knowledge-based systems and groupware development

14.3 Representation in Logic and Other Schemas
General form of any logical process (Figure 14.1) Inputs (Premises) Premises used by the logical process to create the output, consisting of conclusions (inferences) Facts known true can be used to derive new facts that also must be true

Two Basic Forms of Computational Logic
Symbolic logic: System of rules and procedures that permit the drawing of inferences from various premises Two Basic Forms of Computational Logic Propositional logic (or propositional calculus) Predicate logic (or predicate calculus)

Propositional Logic A proposition is a statement that is either true or false Once known, it becomes a premise that can be used to derive new propositions or inferences Rules are used to determine the truth (T) or falsity (F) of the new proposition

Symbols represent propositions, premises or conclusions
Statement: A = The mail carrier comes Monday through Friday. Statement: B = Today is Sunday. Conclusion: C = The mail carrier will not come today. Propositional logic: limited in representing real-world knowledge

(Note - the period “.” is part of the statement)
Predicate Calculus Predicate logic breaks a statement down into component parts, an object, object characteristic or some object assertion Predicate calculus uses variables and functions of variables in a symbolic logic statement Predicate calculus is the basis for Prolog (PROgramming in LOGic) Prolog Statement Examples comes_on(mail_carrier, monday). likes(jay, chocolate). (Note - the period “.” is part of the statement)

Knowledge Representation Scheme Describing a
Scripts Knowledge Representation Scheme Describing a Sequence of Events Elements include Entry Conditions Props Roles Tracks Scenes

Written Series of Related Items
Lists Written Series of Related Items Normally used to represent hierarchical knowledge where objects are grouped, categorized or graded according to Rank or Relationship

Decision Tables (Induction Table)
Knowledge Organized in a Spreadsheet Format Attribute List Conclusion List Different attribute configurations are matched against the conclusion

Decision Trees Related to tables
Similar to decision trees in decision theory Can simplify the knowledge acquisition process Knowledge diagramming is frequently more natural to experts than formal representation methods

O-A-V Triplet Objects, Attributes and Values O-A-V Triplet
Objects may be physical or conceptual Attributes are the characteristics of the objects Values are the specific measures of the attributes in a given situation O-A-V triplets (Table 14.1)

14.4 Semantic Networks Graphic Depiction of Knowledge
Nodes and Links Showing Hierarchical Relationships Between Objects Simple Semantic Network (Figure 14.2) Nodes: Objects Arcs: Relationships is-a has-a

Semantic networks can show inheritance
Semantic Nets - visual representation of relationships Can be combined with other representation methods

14.5 Production Rules Condition-Action Pairs
IF this condition (or premise or antecedent) occurs, THEN some action (or result, or conclusion, or consequence) will (or should) occur IF the stop light is red AND you have stopped, THEN a right turn is OK

Each production rule in a knowledge base represents an autonomous chunk of expertise
When combined and fed to the inference engine, the set of rules behaves synergistically Rules can be viewed as a simulation of the cognitive behavior of human experts Rules represent a model of actual human behavior

Forms of Rules IF premise, THEN conclusion
IF your income is high, THEN your chance of being audited by the IRS is high Conclusion, IF premise Your chance of being audited is high, IF your income is high

Inclusion of ELSE IF your income is high, OR your deductions are unusual, THEN your chance of being audited by the IRS is high, OR ELSE your chance of being audited is low More Complex Rules IF credit rating is high AND salary is more than $30,000, OR assets are more than $75,000, AND pay history is not "poor," THEN approve a loan up to $10,000, and list the loan in category "B.” Action part may have more information: THEN "approve the loan" and "refer to an agent"

Knowledge and Inference Rules
Common Types of Rules Knowledge rules, or declarative rules, state all the facts and relationships about a problem Inference rules, or procedural rules, advise on how to solve a problem, given that certain facts are known Inference rules contain rules about rules (metarules) Knowledge rules are stored in the knowledge base Inference rules become part of the inference engine

Major Advantages of Rules
Easy to understand (natural form of knowledge) Easy to derive inference and explanations Easy to modify and maintain Easy to combine with uncertainty Rules are frequently independent

Major Limitations of Rules
Complex knowledge requires many rules Builders like rules (hammer syndrome) Search limitations in systems with many rules Major Characteristics of Rules (Table 14.2)

Definitions and Overview
14.6 Frames Definitions and Overview Frame: Data structure that includes all the knowledge about a particular object Knowledge organized in a hierarchy for diagnosis of knowledge independence Form of object-oriented programming for AI and ES. Each Frame Describes One Object Special Terminology (Table 14.3)

Provide a concise, structural representation of knowledge in a natural manner
Frame encompasses complex objects, entire situations or a management problem as a single entity Frame knowledge is partitioned into slots Slot can describe declarative knowledge or procedural knowledge Major Capabilities of Frames (Table 14.4) Typical frame describing an automobile (Figure 14.3) Hierarchy of Frames: Inheritance

14.7 Multiple Knowledge Representation
Knowledge Representation Must Support Acquiring knowledge Retrieving knowledge Reasoning

Considerations for Evaluating a Knowledge Representation
Naturalness, uniformity and understandability Degree to which knowledge is explicit (declarative) or embedded in procedural code Modularity and flexibility of the knowledge base Efficiency of knowledge retrieval and the heuristic power of the inference procedure

Production Rules and Frames works well in practice
No single knowledge representation method is ideally suited by itself for all tasks (Table 14.5) Multiple knowledge representations: each tailored to a different subtask Production Rules and Frames works well in practice Object-Oriented Knowledge Representations Hypermedia

14.8 Experimental Knowledge Representations
Cyc NKRL Spec-Charts Language

The Cyc System Attempt to represent a substantial amount of common sense knowledge Bold assumptions: intelligence needs a large amount of knowledge Need a large knowledge base Cyc over time is developing as a repository of a consensus reality - the background knowledge possessed by a typical U.S. resident There are some commercial applications based on portions of Cyc

NKRL Narrative Knowledge Representational Language (NKRL)
Standard, language-independent description of the content of narrative textual documents Can translate natural language expressions directly into a meaningful set of templates that represent the knowledge

Knowledge Interchange Format (KIF)
To Share Knowledge and Interact

The Spec-Charts Language
Based on Conceptual Graphs: to Define Objects and Relationships Restricted Form of Semantic Networks Evolved into the Commercial Product - STATEMATE

14.9 Representing Uncertainty: An Overview
Dealing with Degrees of Truth, Degrees of Falseness in ES Uncertainty When a user cannot provide a definite answer Imprecise knowledge Incomplete information

Several Approaches Related to Mathematical and Statistical Theories
Uncertainty Several Approaches Related to Mathematical and Statistical Theories Bayesian Statistics Dempster and Shafer's Belief Functions Fuzzy Sets

Approximate Reasoning, Inexact Reasoning
Uncertainty in AI Approximate Reasoning, Inexact Reasoning

Relevant Information is Deficient in One or More
Information is partial Information is not fully reliable Representation language is inherently imprecise Information comes from multiple sources and it is conflicting Information is approximate Non-absolute cause-effect relationships exist Can include probability in the rules IF the interest rate is increasing, THEN the price of stocks will decline (80% probability)

Summary The two main parts of any AI system: knowledge base and an inferencing system The knowledge base is made up of facts, concepts, theories, procedures and relationships representing real-world knowledge about objects, places, events, people and so on The inference engine (thinking mechanism) uses the knowledge base, reasoning with it

To build the knowledge base, a variety of knowledge representation schemes are used: logic, lists, semantic networks, frames, scripts and production rules Propositional logic uses symbols to represent and manipulate premises, prove or disprove propositions and draw conclusions Predicate calculus: a type of logic to represent knowledge as statements that assert information about objects or events, and apply them in reasoning

Semantic networks: graphic depictions of knowledge that show relationships (arcs) between objects (nodes); common relationships: is-a, has-a, owns, made from Major property of networks: inheritance of properties through the hierarchy Scripts describe an anticipated sequence of events; indicate participants, actions, setting Decision trees and tables: used in conjunction with other representation methods. Help organize acquired knowledge before coding

Production rules: IF-THEN statement
Two rule types: declarative (describing facts) and procedural (inference) Rules: easy to understand; inferences can be easily derived from them Complex knowledge may require thousands of rules; may create problems in both search and maintenance. Some knowledge cannot be represented in rules

Frame: holistic data structure based on object-oriented programming technology
Frames: composed of slots that may contain different types of knowledge representation (rules, scripts, formulas) Frames: can show complex relationships, graphic information and inheritance concisely. Modular structure helps in inference and maintenance Integrating several knowledge representation methods is gaining popularity: decreasing software costs and increasing capabilities

Experimental knowledge representations focus on expressing general knowledge about the world, and specialized languages that incorporate graphs and logic Knowledge may be inexact and experts may be uncertain at a given time Uncertainty can be caused by several factors ranging from incomplete to unreliable information

1. What was the purpose of the Pitney Bowes ES? 2. Why was a rule-based knowledge representation appropriate? 3. Would a frame-based knowledge representation work? Why or why not? 4. What were the benefits of the ES? What potential disadvantages can you determine? 5. Check the literature for other ES for diagnosis and compare what you find to the description in the Opening Vignette.

Group Exercises 1. Have everyone in the group consider the fairly ‘easy’ task of doing laundry. Individually, write down all the facts that you use for sorting clothes, loading the washer and dryer, and folding the clothes. Compare notes. Are any members of the group better at the task than others. For simplicity, leave out details like ‘go to the laundromat.’ Code the doing laundry facts in a rule-base. How many exceptions to the rules did you find?

Chapter 15: Inferences, Explanations and Uncertainty 15
Chapter 15: Inferences, Explanations and Uncertainty Opening Vignette: Konica Automates a Help Desk with Case-based Reasoning The Problem Konica Business Machines wanted to fully automate its help desk for Internal Support External Support Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Solution Most Promising Approach
Case-based Reasoning Software Artistry’s Expert Advisor Could Run Multiple Problem Resolution Modes Decision Trees Adaptive Learning Tech Search More Used ‘standard cases initially Later used real cases to boost accuracy Includes digital photos Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Expert Advisor Used by Internal Tech Support (6750 people)
Used by Customers Tech Support handles unusual cases Early stage testing: 65% hit rate Adaptive learning being added Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.2 Reasoning in Artificial Intelligence
Once knowledge is acquired, it must be stored and processed (reasoned with) Need a computer program to access knowledge for making inferences This program is an algorithm that controls a reasoning process Inference engine or control program Rule interpreter (in rule-based systems) The inference engine directs the search through the knowledge base Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

How People Reason and Solve Problems
Sources of Power Formal methods (logical deduction) Heuristic reasoning (IF-THEN rules) Focus--common sense related toward more or less specific goals Divide and conquer Parallelism Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Representation Analogy Synergy Serendipity (Luck)
(Lenat [1982]) Sources of power translated to specific reasoning or inference methods (Table 15.1) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Reasoning with Logic Modus Ponens
If A, then B [A AND (A  B)]  B A and (A  B) are propositions in a knowledge base Modus Tollens: when B is known to be false Resolution: combines substitution, modus ponens and other logical syllogisms Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.3 Inferencing with Rules: Forward and Backward Chaining
Firing a rule: When all of the rule's hypotheses (the “if parts”) are satisfied Can check every rule in the knowledge base in a forward or backward direction Continues until no more rules can fire, or until a goal is achieved Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Forward and Backward Chaining
Chaining: Linking a set of pertinent rules Search process: directed by a rule interpreter approach: Forward chaining. If the premise clauses match the situation, then the process attempts to assert the conclusion Backward chaining. If the current goal is to determine the correct conclusion, then the process attempts to determine whether the premise clauses (facts) match the situation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Backward Chaining Goal-driven - Start from a potential conclusion (hypothesis), then seek evidence that supports (or contradicts) it Often involves formulating and testing intermediate hypotheses (or subhypotheses) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Forward Chaining Data-driven - Start from available information as it becomes available, then try to draw conclusions What to Use? If all facts available up front (as in auditing) - forward chaining Diagnostic problems - backward chaining Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

(Goal Tree or Logical Tree)
15.4 The Inference Tree (Goal Tree or Logical Tree) Schematic view of the inference process Similar to a decision tree (Figure 15.2) Inferencing: tree traversal Advantage: Guide for the Why and How Explanations Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.5 Inferencing with Frames
Much more complicated than reasoning with rules Slot provides for expectation-driven processing Empty slots can be filled with data that confirm expectations Look for confirmation of expectations Often involves filling in slot values Can Use Rules in Frames Hierarchical Reasoning Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.6 Model-based Reasoning
Based on knowledge of structure and behavior of the devices the system is designed to understand Especially useful in diagnosing difficult equipment problems Can overcome some of the difficulties of rule-based ES (AIS in Action 15.2) Systems include a (deep-knowledge) model of the device to be diagnosed that is then used to identify the cause(s) of the equipment's failure Reasons from "first principles" (Common Sense) Often combined with other representation and inferencing methods. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Model-based ES tend to be "transportable”
Simulates the structure and function of the machinery being diagnosed Models can be either mathematical or component Necessary condition is the creation of a complete and accurate model of the system under study Especially useful in real-time systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.7 Case-based Reasoning (CBR)
Adapt solutions used to solve old problems for new problems Variation - Rule-induction method (Chap. 13) But, CBR Finds cases that solved problems similar to the current one, and Adapts the previous solution or solutions to fit the current problem, while considering any difference between the two situations Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Finding Relevant Cases Involves
Characterizing the input problem, by assigning appropriate features to it Retrieving the cases with those features Picking the case(s) that best match the input best Extremely effective in complex cases Justification - Human thinking does not use logic (or reasoning from first principle) Process the right information retrieved at the right time Central problem - Identification of pertinent information whenever needed - Use Scripts Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

What is a Case? Case - Defines a problem in natural language descriptions and answers to questions, and associates with each situation a proper business action Scripts - Describe a well-known sequence of events Often “reasoning is applying scripts” More Scripts, Less (Real) Thinking Can be constructed from historical cases Case-based reasoning is the essence of how people reason from experience CBR - a more psychologically plausible expert reasoning model than a rule-based model (Table 15.2) Advantages of CBR (Table 15.3) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Case-based Reasoning Process (Figure 15.3)
Assign Indexes Retrieve Modify Test Assign and Store Explain, Repair and Test Types of Knowledge Structures (Ovals) Indexing Rules Case Memory Similarity Metrics Modification Rules Repair Rules Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

CBR Uses, Issues and Applications
Guidelines (Table 15.4) Target Application Domains Tactical planning Political analysis Situation assessment Legal planning Diagnosis Fraud detection Design/configuration Message classification (Cognitive Systems, Inc.) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

CBR Issues and Questions
What makes up a case? How can we represent case memory? Automatic case-adaptation rules can be very complex How is memory organized? What are the indexing rules? The quality of the results is heavily dependent on the indexes used How does memory function in retrieval of relevant information? How can we perform efficient search (knowledge navigation) of the cases? How can we organize (cluster) the cases? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

How can we design the distributed storage of cases?
How can we adapt old solutions to new problems? Can we simply adapt the memory for efficient query, depending on context? What are the similarity metrics and the modification rules? How can we factor errors out of the original cases? How can we learn from mistakes? i.e., how do we repair / update the case-base? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Are there alternative retrieval systems that match the CBR schema?
The case base may need to be expanded as the domain model evolves, yet much analysis of the domain may be postponed How can we integrate of CBR with other knowledge representations and inferencing mechanisms Are there better pattern matching methods than the ones we currently use? Are there alternative retrieval systems that match the CBR schema? Since 1995, More Real-World CBR Applications Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

A CBR Application Example Classify incoming telex messages to be Processed Faster
Step 1: Collection of messages Case library over 10,000 sample messages Step 2: Expert establishes a hierarchy of telex classifications based on content (109 types of messages) Step 3: An expert matches the messages in the case library against the 109 categories Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Step 4: Formulas are used to create abstract features, that can either be used to predict a classification or as the classification to be predicted Step 5: Lexical patterns, consisting of words, phrases, abbreviations and synonyms, are established for the domain. These patterns are used to tokenize each message Step 6: Each case in the library is then fully represented (classification, formulas and features summarized on one page) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Step 7: Using the CBR shell, the domain expert applies special techniques to identify possible features that may be important in determining the message category Step 8: An incoming message's classification is determined (automatically) by matching the incoming case with similar cases from the case library (explanations provided automatically)

CBR Construction - Special Tools - Examples
ART*Enterprise and CBR Express (Inference Corporation) KATE (Acknosoft) ReMind (Cognitive Systems Inc.) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.8 Explanation and Metaknowledge
Human experts justify and explain their actions ES should also do so Explanation: Attempt by an ES to clarify reasoning, recommendations, other actions (asking a question) Explanation facility (justifier) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Explanation Purposes Make the system more intelligible
Uncover shortcomings of the rules and knowledge base (debugging) Explain situations unanticipated Satisfy users’ psychological and/or social needs Clarify the assumptions underlying the system's operations Conduct sensitivity analyses Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Rule Tracing Technique
“Why” Provides a Chain of Reasoning Good Explanation Facility is critical in large ES Understanding depends on explanation Explanation is essential in ES Used for training Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Two Basic Explanations
Why Explanations - Typically why is a fact requested? How Explanations - Typically to determine how a certain conclusion or recommendation was reached Some simple systems - only at the final conclusion Most complex systems provide the chain of rules used to reach the conclusion Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Other Explanations Journalistic Explanation Facility (Wick and Slagle [1989]) Who, what, where, when, why and how (“5 Ws” plus How) Why not? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Metaknowledge Knowledge about how the system reasons
Knowledge about knowledge Inference rules are a special case Metaknowledge allows the system to examine the operation of the declarative and procedural knowledge in the knowledge base Explanation can be viewed as another aspect of metaknowledge Over time, metaknowledge will allow ES to create the rationale behind individual rules by reasoning from first principles Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Generating Explanations
Static Explanation: Preinsert pieces of English text (scripts) in the system Dynamic Explanation: Reconstruct explanation according to the execution pattern of the rules Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Typology of ES Explanations
Trace, or Line of Reasoning Justification - Explicit description of the causal argument or rationale behind each inferential step taken by the ES Strategy - high-level goal structure that determines how the ES uses its domain knowledge to accomplish a task (or metaknowledge). Ye and Johnson [1995] Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.9 Inferencing with Uncertainty
Uncertainty in AI - Three-step Process (Figure 15.4) 1. An expert provides inexact knowledge in terms of rules with likelihood values 2. The inexact knowledge of the basic set of events can be directly used to draw inferences in simple cases (Step 3) 3. Working with the inference engine, experts can adjust the Step 1 input after viewing the results in Steps 2 and 3. In Step 2: Often the various events are interrelated. Necessary to combine the information provided in Step 1 into a global value for the system Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Uncertainty is a serious problem
Major integration methods: Bayesian probabilities, theory of evidence, certainty factors and fuzzy sets Uncertainty is a serious problem Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.10 Representing Uncertainty
Numeric Graphic Symbolic Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Numeric Uncertainty Representation
Scale (0-1, 0-100) 0 = Complete uncertainty 1 or 100 = Complete certainty Problems with Cognitive Biases People May be Inconsistent at Different Times

Graphic and Influence Diagrams
Horizontal bars (Figure 15.5) Not as accurate as numbers Experts may not have experience in marking graphic scales Many experts prefer ranking over graphic or numeric methods Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Symbolic Representation of Uncertainty
Several Ways to Represent Uncertainty Likert Scale Approach Ranking Ordinal Cardinal Pair-wise Comparison (Analytical Hierarchy Process) Fuzzy logic includes a special symbolic representation combined with numbers Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.11 Probabilities and Related Approaches
The Probability Ratio P(X) = Number of outcomes favoring the occurrence of X / Total number of outcomes Multiple Probability Values in Many Systems Three-part antecedent (probabilities: 0.9, 0.7 and 0.65) The overall probability: P = (0.9)(0.7)(0.65) = Sometimes one rule references another - individual rule probabilities can propagate from one to another Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Several Approaches for Combining Probabilities
Probabilities can be Multiplied (joint probabilities) Averaged (simple or a weighted average) Highest value Lowest value Rules and events are considered independent of each other If Dependent - Use the Bayes extension theorem Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Bayesian Extension
Bayes' Theorem for combining new and existent evidence usually given as subjective probabilities To revise existing prior probabilities based on new information Based on subjective probabilities; a subjective probability is provided for each proposition Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Two Major Deficiencies
The single value does not tell us much about its precision The single value combines the evidence for and against a proposition without indicating how much there is individually in each The subjective probability expresses the "degree of belief," or how strongly a value or a situation is believed to be true The Bayesian approach, with or without new evidence, can be diagrammed as a network. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Dempster-Shafer Theory of Evidence
Distinguishes between uncertainty and ignorance by creating belief functions Especially appropriate for combining expert opinions, since experts do differ in their opinions with a certain degree of ignorance Assumes that the sources of information to be combined are statistically independent Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.12 Theory of Certainty (Certainty Factors)
Certainty Factors and Beliefs Uncertainty is represented as a Degree of Belief Express the Measure of Belief Manipulate degrees of belief while using knowledge-based systems Certainty Theory uses Certainty Factors Certainty Factors (CF) express belief in an event (or fact or hypothesis) based on evidence (or the expert's assessment) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

CFs are NOT probabilities CFs need not sum to 100
Several methods of using certainty factors in handling uncertainty in knowledge-based systems 1.0 or 100 = absolute truth (complete confidence) 0 = certain falsehood CFs are NOT probabilities CFs need not sum to 100 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Belief and Disbelief CF[P,E ] = MB[P,E] - MD[P,E] where
CF = certainty factor MB = measure of belief MD = measure of disbelief P = probability E = evidence or event Another assumption - the knowledge content of rules is much more important than the algebra of confidences that holds the system together Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Combining Certainty Factors
Must Know How CFs are Used (Appendix 15-A) Combining Several Certainty Factors in One Rule AND, OR Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

AND CF(A, B, and C) = Minimum[CF(A), CF(B), CF(C)]
IF inflation is high, CF = 50 percent, (A), AND IF unemployment rate is above 7 percent, CF = 70 percent, (B), AND IF bond prices decline, CF = 100 percent, (C) THEN stock prices decline CF(A, B, and C) = Minimum[CF(A), CF(B), CF(C)] The CF for “stock prices to decline” = 50 percent The chain is as strong as its weakest link

OR IF inflation is low, CF = 70 percent; OR
IF bond prices are high, CF = 85 percent; THEN stock prices will be high Only one IF need be true Conclusion has a CF with the maximum of the two CF (A or B) = Maximum [CF (A), CF (B)] CF = 85 percent for stock prices to be high Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Combining Two or More Rules
Example: R1: IF the inflation rate is less than 5 percent, THEN stock market prices go up (CF = 0.7) R2: IF unemployment level is less than 7 percent, THEN stock market prices go up (CF = 0.6) Inflation rate = 4 percent and the unemployment level = 6.5 percent Combined Effect CF(R1,R2) = CF(R1) + CF(R2)[1 - CF(R1)]; or CF(R1,R2) = CF(R1) + CF(R2) - CF(R1)  CF(R2) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Assume an independent relationship between the rules
Example: Given CF(R1) = 0.7 AND CF(R2) = 0.6, then: CF(R1,R2) = ( ) = (0.3) = 0.88 ES tells us that there is an 88 percent chance that stock prices will increase For a third rule to be added CF(R1,R2,R3) = CF(R1,R2) + CF(R3) [1 - CF(R1,R2)] Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

(Appendix 15-A - How different ES shells handle uncertainty)
Third Rule R3: IF bond price increases, THEN stock prices go up (CF = 0.85) Assuming all rules are true in their IF part, the chance that stock prices will go up is CF(R1,R2,R3) = ( ) = (.12) = 0.982 (Appendix 15-A - How different ES shells handle uncertainty) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

15.13 Qualitative Reasoning (QR)
Means of representing and making inferences using general, physical knowledge about the world QR is a model-based procedure that consequently incorporates deep knowledge about a problem domain Typical QR Logic “If you touch a kettle full of boiling water on a stove, you will burn yourself” “If you throw an object off a building, it will go down” Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

But No specific knowledge about boiling temperature, just that it is really hot! No specific information about the building or object, unless you are the object, or you are trying to catch it Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Main goal of QR: To represent commonsense knowledge about the physical world, and the underlying abstractions used in quantitative models (objects fall) Given such knowledge and appropriate reasoning methods, an ES could make predictions and diagnoses, and explain the behavior of physical systems qualitatively, even when exact quantitative descriptions are unavailable or intractable Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Qualitative Reasoning
Relevant behavior is modeled Temporal and spatial qualities in decision making are represented effectively Applies common sense mathematical rules to variables and functions There are structure rules and behavior rules Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Some Real-world QR Applications
Nuclear Plant Fault Diagnoses Business Processes Financial Markets Economic Systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary Several methods can direct search and reasoning: Chaining (backward and forward), model-based reasoning and case-based reasoning Analogical reasoning relates past experiences to a current case Modus ponens says that in an IF-THEN rule, if one part is true, so is the other Testing rules is based on a pattern-matching approach Backward chaining: Search starts from a specific goal Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chaining can be described by an inference tree
Forward chaining: Search starts from the data (evidence) and tries to arrive at one or more conclusions Chaining can be described by an inference tree Inferencing with frames is frequently done with rules In model-based reasoning, a model describes the system. Experimentations are conducted using a what-if approach Case-based Reasoning: Based on experience with similar situations Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Two Explanations in most ES: Why and How
In case-based reasoning, the attributes of an existing case are compared to critical attributes derived from cases stored in the case library Two Explanations in most ES: Why and How Metaknowledge is knowledge about knowledge - useful in generating explanations Static explanation Dynamic explanation AI treats uncertainty as : 1) uncertainty is represented, 2) combined, 3) inferences are drawn Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Disbelief expresses a feeling of what is not going to occur
Three basic methods can be used to represent uncertainty: numeric (probability-like), graphic and qualitative Disbelief expresses a feeling of what is not going to occur Certainty theory Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Certainty theory uses a special formula to combine two or more rules
Qualitative reasoning represents and reasons with knowledge about the physical world Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. Why did Konica want to automate its help desk? What was CBR selected as the technology of choice? 2. Who are the Expert Advisor end-users? Does this present any challenges for the implementation team? 3. Why did the hit rate go up when the knowledge engineers dropped the technical manuals in favor of real cases? 4. What will the adaptive learning feature of Expert Advisor do to the effectiveness of the system? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

CASE APPLICATION 15.1: Compaq QuickSource: Using Case-based Reasoning for Problem Determination
Case Questions 1. How can QuickSource empower Compaq's customers? 2. How is the system matching the problem description to the case base? 3. What are the benefits of the system to the customers? To Compaq? 4. How can QuickSolve provide a competitive advantage to Compaq? 5. When problems are really hard or QuickSolve cannot determine a recommendation, what should be done (e.g., what could the software do)? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

APPENDIX 15-A: How ES Shells Handle Uncertainty
1. EMYCIN (Classic ES shell) Given: -1  CF  + 1 i) IF, CF1  CF2  0 THEN, CFX = CF1 + CF2 - CF1  CF2 ii) IF, CF1  0, CF2  0 THEN, CFX = CF1 + CF2 + CF1  CF2 iii) IF, CF1 and CF2 have different signs THEN, CFX = (CF1 + CF2) / {1 - MIN(|CF1|, |CF2|)} Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2. EXSYS (popular, rule-based shell). Two options i)
2. EXSYS (popular, rule-based shell). Two options i) Scale of 0 through 10, 0 = False; 10 = True Given: CF = 0, 1, 2, , 10 IF, either CF1 or CF2 = 0 or 10 THEN, CFX is the first 0 or 10 found ELSE, CFX = AVG(CF1, CF2) ii) -100 to +100 scale. Three sub-options a) Average the certainty factors: Given: -100  CF  100 THEN, CFX = AVG(CF1, CF2) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

b) Multiply the certainty factors (similarly to a joint. probability):
b) Multiply the certainty factors (similarly to a joint probability): Given: -100  CF  IF, CF1  0 AND CF2  0 THEN, CFX = CF1  CF2/100 ELSE, UNDEFINED c) Certainty-factors-like approach Given: -100  CF  IF, CF1  0 AND CF2  0 THEN, CFX = (100 - CF1)  (100 - CF2)/100 ELSE, UNDEFINED Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3. VP Expert (small, popular shell)
Given: 0  CF  THEN, CFX = CF1 + CF2 - CF1  CF2/100

Chapter 16: Building Expert Systems: Process and Tools
Overview of the Expert System Building Process Performed Differently Depending on the Nature of the System Being Constructed Development Strategy Supporting Tools

16.1 Opening Vignette: The Logistics Management System (LMS) at IBM
June 1985, IBM Burlington, VT Plant Industrial Engineering Group (IEG) Proposed The Logistics Management System (LMS) Expert System to Improve line flow and utilization Reduce cycle time in semiconductor manufacturing

Obtained Top Management’s Full Support
Created a development team with IEG’s manager Several IEG building representatives LMS building representatives Two industrial engineers Internal software development consultant

GATEWAY and MAT provided many EIS and DSS features
Became quickly embedded into the business processes ALERT Component increased throughput by detecting conditions that lead to future bottlenecks

June 1987 - LMS status elevated from exploratory to tactical
Saved Tens of Millions $ System Enhanced and Expanded over Time Ported to a PC 1991: LMS became an IBM product 1994 LMS elevated to a strategic application LMS - A long-term, successful ES

Evolutionary Steps in LMS Development
Establishment of a funded project Establishment of a multidisciplinary development team Understanding and control of data Broad user involvement (later) Internal development of new expertise Expansion of the user base Personnel shifts and reassignment of tasks and responsibilities, and Emergence of new business opportunities (to a broader market)

Critical to LMS Implementation Success
Top management support Business benefits Talented multidisciplinary team

16.2 The Development Life Cycle
For building expert systems - six phases (Figure 16.1) Process is nonlinear

Phases 1. Project Initialization 2. Systems Analysis and Design
3. Rapid Prototyping 4. System Development 5. Implementation 6. Postimplementation

16.3 Phase I: Project Initialization
(Table 16.1) Problem Definition Need Assessment Evaluation of Alternative Solutions Verification of an Expert Systems Approach Consideration of Managerial Issues

16.4 Problem Definition and Need Assessment
Write a clear statement and provide as much supporting information as possible Conduct a formal needs assessment to understand the problem

16.5 Evaluation of Alternative Solutions
Using Experts Education and Training Packaged Knowledge Conventional Software Buying Knowledge on the Internet

16.6 Verification of an Expert System Approach
Framework to determine problem fit with an ES (Waterman [1985]) 1. Requirements for ES Development 2. Justification for ES Development 3. Appropriateness of ES

1. Requirements for ES Development (all are necessary)
1. Task does not require common sense 2. Task requires only cognitive, not physical, skills 3. At least one genuine expert, willing to cooperate 4. Experts involved can articulate their problem solving methods 5. Experts involved can agree on the knowledge and the solution approach

6. Task is not too difficult
7. Task well understood and defined clearly 8. Task definition fairly stable 9. Conventional (algorithmic) computer solution techniques not satisfactory 10. Incorrect or nonoptimal results generated by the ES can be tolerated 11. Data and test cases are available 12. Task's vocabulary has no more than a couple of hundred concepts

2. Justification for ES Development (Need at least one)
1. Solution to the problem has a high payoff 2. ES can preserve scarce human expertise, so it will not be lost 3. Expertise is needed in many locations 4. Expertise is needed in hostile or hazardous environments 5. The expertise improves performance and/or quality 6. System can be used for training 7. ES solution can be derived faster than a human 8. ES is more consistent and/or accurate than a human Of Course: Benefits Must Exceed System Costs

3. Appropriateness of the ES (Consider 3 Factors)
1. Nature of the problem: Symbolic structure and heuristics 2. Complexity of the task: Neither too easy nor too difficult for a human expert 3. Scope of the problem: Manageable size and practical value Problem Selection is a Critical Factor

16.7 Consideration of Managerial Issues
Selling the project Identifying a champion Level of top management support End user involvement, support and training Availability of financing Availability of other resources Legal and other potential constraints

16.8 Phase II: System Analysis and Design
(Table 16.2) Conceptual Design and Plan Development Strategy Sources of Knowledge Computing Resources Feasibility Study Cost-Benefit Analysis

16.9 Conceptual Design General Idea of the System
General capabilities of the system Interfaces with other CBIS Areas of risk Required resources Anticipated cash flow Composition of the team Other information for detailed design later Determine the development strategy after design is complete

16.10 Development Strategy and Methodology
General Classes of AI Development Strategies Do It Yourself 1. AI development can be part of end-user computing 2. AI projects can be completely centralized 3. AI development can be decentralized, but control can be centralized 4. High technology islands 5. Utilize information centers Hire an Outside Developer Enter into a Joint Venture Merge, Acquire or Become a Major Stockholder in an AI Company

16.11 Selecting an Expert Experts
Expertise is based on experience and can be expressed by heuristics (Table 16.3) Selection Issues Who selects the expert(s)? How to identify an expert? What to do if several experts are needed? How to motivate the expert to cooperate?

16.12 Software Classification: Technology Levels
(Figure 16.2) Specific Expert Systems Shells Support Tools Hybrid Systems (Environments) Programming Languages NEW Object-oriented Programming (OOP) Internet/Web/Intranet-based Tools

16.13 Building Expert Systems with Tools
1. The builder employs the tool's development engine to load the knowledge base 2. The knowledge base is tested on sample problems using the inference engine 3. The process is repeated until the system is operational 4. The development engine is removed and the specific expert system is ready for its users (using a separate runtime (executable) component of the tool)

16.14 Shells and Environments
Expert Systems Components 1. Knowledge acquisition subsystems 2. Inference engine 3. Explanation facility 4. Interface subsystem 5. Knowledge base management facility 6. Knowledge base Shell: Components 1-5 (Figure 16.3)

Rule-Based Shells EXSYS Guru NEXPERT OBJECT KEE 1stCLASS

Domain-Specific Tools
Designed to be used only in the development of a specific area Diagnostic systems Shells for configuration Shells for financial applications Shells for scheduling

Development Environments
Support several different knowledge representations and inference methods (Table 16.4) Examples KEE ART-IM Level5 Object KAPPA PC

Major Issues in Selecting ES Development Software (Table 16.5)
16.15 Software Selection Complex Problem Frequent Technology Changes Many criteria First Check Out “PC AI Buyer’s Guide” “Expert Systems Resource Guide” in AI Expert FAQs of newsgroups on ES Major Issues in Selecting ES Development Software (Table 16.5)

Software Choice Usually Depends
16.16 Hardware Support Software Choice Usually Depends on the Hardware AI Workstations Mainframes PCs Unix Workstations

16.17 Feasibility Study Outline in Table 16.6

16.18 Cost-Benefit Analysis
To Determine Project Viability Often Very Complicated Difficult to Predict Costs and Benefits Expert Systems Evolve Constantly

Cost-Benefit Analysis - Complicating Factors
Getting a Handle on Development Costs Consider (and Revise) the Scope of the System Estimate Time Requirements Evaluating the Benefits Some Intangible Hard to relate specifically to the ES Benefits result over time Not easy to assess quantity and quality Multiplicity of Consequences hard to evaluate: (goodwill, inconvenience, waiting time and pain) Key: Identify the Appropriate Benefits

When to Justify (Very Often!)
At the end of Phase I At the end of Phase II After the initial prototype is completed Once the full prototype is in operation Once field testing is completed (prior to deployment) Periodically after the system is in operation (e.g., every six or twelve months) Reality checks How to Justify?

16.19 Phase III: Rapid Prototyping and a Demonstration Prototype
Build a Small Prototype Test, Improve, Expand Demonstrate and Analyze Feasibility Complete Design

Rapid Prototyping Crucial to ES Development Small-scale System
Includes Knowledge Representation Small Number of Rules For Proof of Concept Rapid Prototyping Process (Figure 16.4) Possible Tasks and Participants in the Rapid Prototyping Process (Figure 16.5)

16.20 Phase IV: System Development
Complete the Knowledge Base Test, Evaluate, and Improve Knowledge Base Plan for Integration Following the Initial Prototype’s Acceptance System Development Begins

Use a System Development Approach
Continue with prototyping Use the structured life cycle approach Do both Tasks and Participants (Figure 16.6)

16.21 Building the Knowledge Base
Acquire and Represent the Knowledge in an Appropriate Form Define the Potential Solutions Define the Input Facts Develop an Outline Draw a Decision Tree Map a Matrix Create the Knowledge Base

16.22 Testing, Validating and Verifying, and Improving
Test and Evaluate the Prototype and Improved Versions of the System Both In the Lab In the Field Initially - Evaluate in a Simulated Environment

Iterative Process of Evaluation: Refine the ES in the Field
Modified Turing Test: Compare ES Performance to an Accepted Criterion (Human Expert's Decisions) Experimentation Iterative Process of Evaluation: Refine the ES in the Field Use New Cases to Expand the Knowledge Base Validation - Determination of Whether the Right System Was Built Whether the system does what it was meant to do and at an acceptable level of accuracy Verification confirms that the ES has been built correctly according to specifications

16.23 Phase V: Implementation
Acceptance by Users Installation, Demonstration, Deployment Orientation, Training Security Documentation Integration, Field Testing

ES Implementation Issues
Acceptance by the User Installation Approaches Demonstration Mode of Deployment Orientation and Training Security Documentation Integration and Field Testing

16.24 Phase VI: Postimplementation
Operations Maintenance and Upgrades (Expansion) Periodic Evaluation

Expansion (Upgrading)
The Environment Changes More Complex Situations Arise Additional Subsystems can be Added (e.g., LMS)

Evaluation (Periodically)
Maintenance Costs Versus Benefits? Is the Knowledge Up to Date? Is the System Accessible to All Users? Is User Acceptance Increasing? (Feedback)

16.25 Organizing the Development Team
Team Varies with the Phases Typical Development Team Expert Knowledge Engineer IS Person

Team May Also Include Vendor(s) User(s) System Integrator(s)
Cooperation and Communication Required!!! Possible Functions and Roles in an ES Team (Table 16.9)

Important Players Project Champion Project Leader

16.26 The Future of Expert Systems Development Processes
Expect Advances In Flexible toolkit capabilities, including inferencing hybrids Improved languages and development systems Better front ends to help the expert provide knowledge Improved GUIs via Windows-based environments Further use of intelligent agents in toolkits Better ways to handle multiple knowledge representations

Use of intelligent agents to assist developers
Use of blackboard architectures and intelligent agents in ES Advances in the object-oriented approach, for representing knowledge and ES programming Improved and customized CASE tools to manage ES development Increased hypermedia use and development (Web) Automated machine learning of databases and text

Summary Building an expert system is a complex process with six major phases: system initialization, system analysis and design, rapid prototyping, system development, implementation and postimplementation Defining the problem properly simplifies the remaining development tasks Sometimes conventional technologies outperform potential expert systems Like any other project, an ES needs to be justified Without the proper level of resource commitment, the ES will fail

Top management support is essential
A large ES needs a champion as a sponsor Expert systems can be developed in-house (internally) or subcontracted (several variations) Expert systems developed by end-users can be successful The Internet is changing the way we provide expertise in organizations adopting ES technology

Although ES can be developed with several tools, the trend is to develop the initial prototype with a simple (and inexpensive) integrated tool (either shell or hybrid environment) Currently, most ES are being developed and run on standard computers (PCs, workstations, mainframes) A feasibility study is essential for ES success Expert systems are difficult to justify because of many intangibles Do justification several times during the development process, and so is the go/no-go decision

Many ES are built by prototyping, testing, and improving and expanding
Many ES are built by prototyping, testing, and improving and expanding. Process advantageous Choosing the correct knowledge representation of the problem domain and the appropriate ES shell or tool are important Major system development aspects: building the knowledge base and evaluating and improving the system Evaluation of expert systems is difficult because of the many attributes that need to be considered and the difficulties in measuring some of them

Implementing an ES is similar to implementing any other CBIS
Implementing an ES is similar to implementing any other CBIS. Integration may be difficult Once the system is distributed to users, must perform: operation, maintenance, upgrade and post-implementation evaluation Developing a proper team for ES development can be challenging; Some Important Factors: size, composition and leadership New, powerful ES development methods include the object-oriented approach toward ES design and development, CASE tools and ES Web-server engines

1. How did the LMS project get started? Is it common in organizations for initial funding of a project like this to be made in the manner described? 2. The formal status of LMS kept being elevated within IBM. Comment on these actions. Is this important? Why or why not? 3. What was different about developing and maintaining LMS as a commercial product versus as an internally used system?

4.Explain the evolutionary steps taken for LMS and compare them to the steps outlined in the rest of the Chapter. Why are they called evolutionary, instead of simply steps? 5.Could the standard system development life cycle have been used for LMS? Why or why not? 6.Why were the three items listed at the end of the Vignette critical to LMS implementation success? Use the material in the chapter if you must.

CASE APPLICATION 16.1: State of Washington's Department of Labor
Case Questions 1. Why is it quicker to make massive changes in an application when ES is used? Why is it a lengthy process otherwise? 2. What is the advantage of rapid prototyping? 3. How one can retrain traditional computer programmers to work with ES technology?

APPENDIX 16-A: How to Build a Knowledge Base (Rule-based) System
Using Expert Systems for Wine and Food Pairing Choosing an appropriate wine match to a certain food is not simple Requires Expertise This ES can help

Building the ES Step 1. Specify the problem. Pairing wine and food at a restaurant Step 2. Name the system (Sommelier) Step 3. Write a starting text Step 4. Decide on an appropriate coding for uncertainty (0 to 10) Step 5. Decide on any other parameters as required by the shell Step 6. Make a list of the choices or alternatives (12 wines) Step 7. Build the what-if rules Step 8. Prepare any concluding note for the user

Rule Writing Standard format required by the shell
EXSYS uses qualifiers for the factors

Qualifier No. 1 One Rule The meat menu selection is
(1) prime rib (2) grilled steak (3) filet mignon One Rule IF the meat menu selection is prime rib THEN Pinot Noir, confidence = 9/10 AND Merlot, confidence = 8/10 AND aged cabernet sauvignon, confidence = 6/10.

Part 5: CUTTING EDGE DECISION SUPPORT TECHNOLOGIES
Neural Computing Genetic Algorithms Fuzzy Logic Integration Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 17: Neural Computing: The Basics
Artificial Neural Networks (ANN) Mimics How Our Brains Work Machine Learning Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17. 1 Opening Vignette: Maximizing the Value of the John Deere & Co
17.1 Opening Vignette: Maximizing the Value of the John Deere & Co. Pension Fund The Problem Managing the Pension Fund of John Deere & Co. About $1 Billion Managed Internally by Corporate Finance Want to Achieve a Better Return on Investment The Solution Neural Computing (Since 1993) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Process Build an individual, ANN for each of the 1000 largest U.S. Corporations Historical Data: 40 inputs Weekly, data updates into a stock prediction model (Figure 17.1) Ranks the stocks based on anticipated performance Selects a portfolio of the 100 top stocks Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Results Returns are well ahead of industry benchmarks

17.2 Machine Learning: An Overview
ANN to automate complex decision making Neural networks learn from past experience and improve their performance levels Machine learning: Methods that teach machines to solve problems, or to support problem solving, by applying historical cases Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Complications Many models of learning
Match the learning model with problem type Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

What is Learning? Accomplished by analogy, discovery and special procedures; by observing; or by analyzing examples Can improve the performance of AI methods Is a Support area of AI Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Learning as Related to AI
Learning systems demonstrate interesting learning behaviors No claims about learning as well as humans or in the same way Learning systems are not defined very formally; Implications are not well understood Learning in AI involves the manipulation of symbols (not numeric information) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Examples of Machine Learning Methods
Neural Computing Inductive Learning Case-based Reasoning and Analogical Reasoning Genetic Algorithms Statistical methods Explanation-based Learning Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.3 An Overview of Neural Computing
Constructing computers that mimic certain processing capabilities of the human brain Knowledge representations based on Massive parallel processing Fast retrieval of large amounts of information The ability to recognize patterns based on historical cases Neural Computing = Artificial Neural Networks (ANNs) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.4 The Biology Analogy Biological Neural Networks
Neurons: Brain Cells Nucleus (at the center) Dendrites provide inputs Axons send outputs Synapses increase or decrease connection strength and causes excitation or inhibition of subsequent neurons Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Artificial Neural Networks (ANN)
A model that emulates a biological neural network Software simulations of the massively parallel processes that involve processing elements interconnected in a network architecture Originally proposed as a model of the human brain’s activities The human brain is much more complex Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.5 Neural Network Fundamentals
Components and Structure Processing Elements Network Structure of the Network Processing Information in the Network Inputs Outputs Weights Summation Function Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Transformation (Transfer) Function
jth Neuron (Figure 17.6) Transformation (Transfer) Function Sigmoid Function (Logical Activation Function) where YT is the transformed (normalized) value of Y Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Learning Three Tasks 1. Compute Outputs
2. Compare Outputs with Desired Targets 3. Adjust Weights and Repeat the Process Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Set the weights by either some rules or randomly
Set Delta = Error = actual output minus desired output for a given set of inputs Objective is to Minimize the Delta (Error) Change the weights to reduce the Delta Information processing: pattern recognition Different learning algorithms Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.6 Neural Network Applications Development
Preliminary steps of system development are done ANN Application Development Process Collect Data Separate into Training and Test Sets Define a Network Structure Select a Learning Algorithm Set Parameters, values, Initialize Weights Transform Data to Network Inputs Start Training, and Determine and Revise Weights Stop and Test Implementation: Use the Network with New Cases Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.7 Data Collection and Preparation
Collect data and separate into a training set and a test set Use training cases to adjust the weights Use test cases for network validation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.8 Neural Network Architectures
Representative Architectures Associative Memory Systems Associative memory - ability to recall complete situations from partial information Systems correlate input data with stored information Hidden Layer Three, Sometimes Four or Five Layers Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Recurrent Structure Recurrent network (double layer) - each activity goes through the network more than once before the output response is produced Uses a feedforward and feedbackward approach to adjust parameters to establish arbitrary numbers of categories Example: Hopfield Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.9 Neural Network Preparation
(Non-numerical Input Data (text, pictures): preparation may involve simplification or decomposition) Choose the learning algorithm Determine several parameters Learning rate (high or low) Threshold value for the form of the output Initial weight values Other parameters Choose the network's structure (nodes and layers) Select initial conditions Transform training and test data to the required format Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.10 Training the Network Present the training data set to the network Adjust weights to produce the desired output for each of the inputs Several iterations of the complete training set to get a consistent set of weights that works for all the training data Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.11 Learning Algorithms Two Major Categories Based On Input Format
Binary-valued (0s and 1s) Continuous-valued Two Basic Learning Categories Supervised Learning Unsupervised Learning Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Supervised Learning For a set of inputs with known (desired) outputs
Examples Backpropagation Hopfield network Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Unsupervised Learning
Only input stimuli shown to the network Network is self-organizing Number of categories into which the network classifies the inputs can be controlled by varying certain parameters Examples Adaptive Resonance Theory (ART) Kohonen Self-organizing Feature Maps Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

How a Network Learns Single neuron - learning the inclusive OR operation Two input elements, X1 and X2 Inputs Case X1 X2 Desired Results (positive) (positive) (positive) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

A step function evaluates the summation of input values
Calculating outputs Measure the error (delta) between outputs and desired values Update weights, reinforcing correct results At any step in the process for a neuron, j, we get Delta = Zj - Yj where Z and Y are the desired and actual outputs, respectively Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The updated weights are
Wi (final) = Wi (initial) + alpha × delta × X1 where alpha is the learning rate parameter Weights initially random The learning rate parameter, alpha, is set low Delta is used to derive the final weights, which then become the initial weights in the next iteration (row) Threshold value parameter: sets Y to 1 in the next row if the weighted sum of inputs is greater than 0.5; otherwise, to 0 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.12 Backpropagation Backpropagation (back-error propagation)
Most widely used learning Relatively easy to implement Requires training data for conditioning the network before using it for processing other data Network includes one or more hidden layers Network is considered a feedforward approach Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Externally provided correct patterns are compared with the neural network output during training (supervised training) Feedback is used to adjust the weights until all training patterns are correctly categorized

Error is backpropogated through network layers
Some error is attributed to each layer Weights are adjusted A large network can take a very long time to train May not converge Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.13 Testing Test the network after training
Examine network performance: measure the network’s classification ability Black box testing Do the inputs produce the appropriate outputs? Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Not necessarily 100% accurate
But may be better than human decision makers Test plan should include Routine cases Potentially problematic situations May have to retrain

17.14 Implementation Frequently requires proper interfaces with other CBIS and user training Gain confidence of the users and management early Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Neural Computing Paradigms
Decisions the builder must make Size of training and test data Learning algorithms Topology: number of processing elements and their configurations Transformation (transfer) function Learning rate for each layer Diagnostic and validation tools Results in the Network's Paradigm Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.15 Programming Neural Networks
An ANN can be programmed with A programming language A programming tool Both Tools (shells) incorporate Training algorithms Transfer and summation functions May still need to Program the layout of the database Partition the data (test data, training data) Transfer the data to files suitable for input to an ANN tool Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Development Tool Examples
NNetSheet (spreadsheet-based) Neuralyst for Excel (front-ended by a spreadsheet) KnowledgeNet and NeuroSMARTS (work with expert systems) NeuralWorks, Explorer and Brainmaker (stand alone environments) Languages (C++) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.16 Neural Network Hardware
Advantages of massive parallel processing Greatly enhances performance Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Possible Hardware Systems for ANN Training
Faster general purpose computers General purpose parallel processors Neural chips Acceleration boards Pentium MMX and II (Graphics Processors) (1998)

17.17 Benefits of Neural Networks
Usefulness for pattern recognition, learning, classification, generalization and abstraction, and the interpretation of incomplete and noisy inputs Specifically - character, speech and visual recognition Potential to provide some of human problem solving characteristics Ability to tackle new kinds of problems Robustness Fast processing speed Flexibility and ease of maintenance Powerful hybrid systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.18 Limitations of Neural Networks
Do not do well at tasks that are not done well by people Lack explanation capabilities Limitations and expense of hardware technology restrict most applications to software simulations Training times can be excessive and tedious Usually requires large amounts of training and test data Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.19 Neural Networks and Expert Systems
ANN and ES technologies are so different that they can complement each other Expert systems represent a logical, symbolic approach Neural networks are model-based and use numeric and associative processing Main features of each (Table 17.2) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Expert Systems Especially good for closed-system applications (literal and precise inputs, logical outputs) Reason by using established facts and pre-established rules Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Major Limitations Experts do not always think in terms of rules
Experts may not be able to explain their line of reasoning Experts may explain it incorrectly Sometimes difficult or impossible to build the knowledge base Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Neural Networks in Knowledge Acquisition
Neural Computing Use Neural Networks in Knowledge Acquisition Fast identification of implicit knowledge by automatically analyzing cases of historical data ANN identifies patterns and relationships that may lead to rules for expert systems A trained neural network can rapidly process information to produce associated facts and consequences Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

17.20 Neural Networks For Decision Support
Inductive means for gathering, storing and using experiential knowledge Neural network-based DSS to appraise real estate in New York (90% accurate) Forecasting ANN in decision support: Easy sensitivity analysis and partial analysis of input factors Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The relationship between a combined expert system, ANN and a DSS (Figure 17.13)
ANN can expand the boundaries of DSS

Summary ANN conceptually similar to biological systems
Human brain composed of billions of neuron cells, grouped in interconnected clusters Neural systems are composed of interconnected processing elements called artificial neurons that form an artificial neural network Can organize an ANN many different ways Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Each neuron has an activation value
Weights express the relative strength (or importance) given to input data Each neuron has an activation value An activation value is translated to an output \through a transformation (transfer) function Artificial neural networks learn from historical cases Learning is done with algorithms Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Testing is done by using historical data
Supervised learning Testing is done by using historical data Neural computing is frequently integrated with traditional CBIS and ES Many ANNs are built with tools that include the learning algorithm(s) and other computational procedures ANNs lend themselves to parallel processing. However, most us standard computers Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Parallel processors can improve the training and running of neural networks
Acceleration boards expedite the computational work of neural computers Neural computing excels in pattern recognition, learning, classification, generalization and abstraction, and interpretation of incomplete input data ANN do well at tasks done well by people and not so well at tasks done well by traditional computer systems Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Machines learn differently from people; usually not as well
Machine learning describes techniques that enable computers to learn from experience Machine learning is used in knowledge acquisition and in inferencing and problem solving Machines learn differently from people; usually not as well Neural networks are useful in data mining Neural networks can enhance DSS Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. Describe why neural networks were used. 2. Why is historical data used? 3. Is the neural network really learning? Why or why not? 4. How could the neural network possibly outperform a human? 5. What related applications can you think of for using neural computing? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Exercise Self-organizing neural network class exercise. Everyone in the class should stand in the front of the room, away from the desks and chairs (a room free of furniture is even better). Without making a sound everyone should line up in order by height, tallest to shortest. Then, at a signal line up alphabetically. Notice that only minimal information is really needed by each person to identify their position. Discuss how this relates to how self-organizing neural networks learn. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 18: Neural Computing Applications, Genetic Algorithms, Fuzzy Logic and Hybrid Intelligent Systems Several Real-world Applications of ANN Technology Two Decision Support Technologies: Genetic Algorithms Fuzzy Logic Integration of these Cutting Edge Technologies Among Themselves With Expert Systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

18.1 Opening Vignette: Applying Neural Computing to Marketing
The Problem Veratex Corp. - Distributor of medical and dental products Deployed neural-network- and expert-system-based sales-support systems Veratex’s Unique Approach to Marketing Products Mail unsolicited catalogs to physicians and dentists When a customer buys from catalog, name added to customer database Telemarketers call the names regularly for reorders Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Company Had to Huge numbers of "dormants" accumulate
Creditability of older data questionable The Company Had to Verify older data Decide to which members of the dormant pool to assign limited telemarketing time Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Solution Back-propagation neural network
Helps identify dormants who are most likely to be "best" customers Weed out potentially bad customers Inputs: Statistical and demographic data Output: Customer rating Customers are rated and ordered Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Three Additional Applications
Expert System Resets Credit Limits Monthly Mimics credit manager’s decision making Scheduler prioritizes tasks: Staff handles delinquent accounts Veratex Benefits More specialized and personalized customer service Efficient credit department Increased sales Happy customers Happy customer service representatives Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

18.2 Areas of ANN Applications: An Overview
Representative Business ANN Applications Accounting Finance Human Resources Management Marketing Operations Airline Crew Scheduling Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Accounting Identify tax fraud
Enhance auditing by finding irregularities

Finance Signatures and bank note verifications Mortgage underwriting
Foreign exchange rate forecasting Country risk rating Bankruptcy prediction Customer credit scoring Credit card approval and fraud detection Corporate merger and take over predictions Currency trading Stock and commodity selection and trading Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Finance 2 Credit card profitability
Forecasting economic turning points Foreign exchange trading Bond rating and trading Pricing initial public offerings Load approvals Economic and financial forecasting Risk management Signature validation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Human Resources Predicting employees’ performance and behavior
Determining personnel resource requirements Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Management Corporate merger prediction Takeover target prediction
Country risk rating

Marketing Consumer spending pattern classification
New product analysis Customers’ characteristics Sales forecasts Data mining Airline fare management Direct mail optimization Targeted marketing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Operations Airline Crew Scheduling
Predicting airline seat demand Vehicle routing Assembly and packaged goods inspection Fruit and fish grading Matching jobs to candidates Production/job scheduling And Many More

18.3 Using ANNs for Credit Approval
Increases loan processor productivity by 25 to 35 % over other computerized tools Also detects credit card fraud Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The ANN Method Data from the application and into a database
Database definition (Figure 18.2) Preprocess applications manually Neural network trained in advance with many good and bad risk cases Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Neural Network Credit Authorizer Construction Process
Step 1: Collect data Step 2: Separate data into training and test sets Step 3: Transform data into network inputs Step 4: Select, train and test network Step 5: Deploy developed network application Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

18.4 Using ANNs for Bankruptcy Prediction
Concept Phase Paradigm: Three-layer network, back-propagation Training data: Small set of well-known financial ratios Data available on bankruptcy outcomes Supervised network Training time not to be a problem Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Application Design Five Input Nodes X1: Working capital/total assets
X2: Retained earnings/total assets X3: Earnings before interest and taxes/total assets X4: Market value of equity/total debt X5: Sales/total assets Single Output Node: Final classification for each firm Bankruptcy or Nonbankruptcy Development Tool: NeuroShell Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Development Three-layer network with backpropagation (Figure 18.5) Continuous valued input Single output node: 0 = bankrupt, 1 = not bankrupt Training Data Set: 129 firms Training Set: 74 firms; 38 bankrupt, 36 not Ratios computed and stored in input files for The neural network A conventional discriminant analysis program Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Parameters Testing Learning threshold Learning rate Momentum Two Ways
Test data set: 27 bankrupt firms, 28 nonbankrupt firms Comparison with discriminant analysis The neural network correctly predicted 81.5 percent bankrupt cases 82.1 percent nonbankrupt cases Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

ANN did better predicting 22 out of the 27 actual cases
Discriminant analysis predicted only 16 correctly Error Analysis Five bankrupt firms misclassified by both methods Similar for nonbankrupt firms Neural network at least as good as conventional Accuracy of about 80 percent is usually acceptable for neural network applications Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

18.5 Stock Market Prediction System with Modular Neural Networks
Accurate Stock Market Prediction - Complex Problem Several Mathematical Models - Disappointing Results Fujitsu and Nikko Securities: TOPIX Buying and Selling Prediction System Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Input: Several technical and economic indexes
Several modular neural networks relate past indexes, and buy / sell timing Prediction system Modular neural networks Very accurate Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Architecture (Figure 18.6)
Network Architecture Network Model (Figure 18.5): 3 layers, standard sigmoid function, continuous output [0, 1] High-speed Supplementary Learning Algorithm Training Data Data Selection Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Moving Simulation Prediction Method (Figure 18.7)
Preprocessing: Input Indexes - Converted into spatial patterns, preprocessed to regularize them Moving Simulation Prediction Method (Figure 18.7) Result of Simulations Simulation for Buying and Selling Stocks Example (Figure 18.8) Excellent Profit Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

18.6 Examples of Integrated ANNs and Expert Systems
1. Resource Requirements Advisor Advises users on database systems’ resource requirements Predict the time and effort to finish a database project ES shell AUBREY and neural network tool NeuroShell ES supported data collection ANN used for data evaluation ES final analysis Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

2. Personnel Resource Requirements Advisor
Project personnel resource requirements for maintaining networks or workstations at NASA Rule-based ES determines the final resource projections ANN provides project completion times for services requested (Figure 18.9) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3. Diagnostic System for an Airline
Singapore Airlines Assist technicians in diagnosing avionics equipment INSIDE (Inertial Navigation System Interactive Diagnostic Expert) Designed to reduce the diagnostic time (Figure 18.10) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

18.7 Genetic Algorithms Goal (evolutionary algorithms): Demonstrate Self-organization and Adaptation by Exposure to the Environment System learns to adapt to changes. Example 1: Vector Game Random Trial and Error Genetic Algorithm Solution Process (Figure 18.11) Example: The Game of MasterMind Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Definition and Process
Genetic algorithm: "an iterative procedure maintaining a population of structures that are candidate solutions to specific domain challenges (Grefenstette [1982]) Each candidate solution is called a chromosome Chromosomes can copy themselves, mate, mutate Use specific genetic operators - reproduction, crossover and mutation. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Primary Operators of Most Genetic Algorithms
Reproduction Crossover Mutation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Genetic Algorithms Applications and Software
Type of machine learning Set of efficient, domain-independent search heuristics for a broad spectrum of applications

General Areas of Genetic Algorithm Applications
Dynamic process control Induction of rule optimization Discovering new connectivity topologies Simulating biological models of behavior and evolution Complex design of engineering structures Pattern recognition Scheduling Transportation Layout and circuit design Telecommunication Graphs Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Documented Business Applications
Channel 4 Television (England) to schedule commercials Driver scheduling in a public transportation system Jobshop scheduling Assignment of destinations to sources Trading stocks Productivity in whisky making is increased Often Genetic Algorithm hybrids with other AI methods Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Representative Commercial Packages
Evolver (Excel spreadsheet addin) OOGA (object-oriented GA for industrial use) XperRule Genasys (ES shell with an embedded genetic algorithm) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

18.8 Optimization Algorithms
Via Neural Computing sometimes Genetic algorithms and their derivatives can optimize (or nearly optimize) complex problems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

18.9 Fuzzy Logic: Theory and Applications
Fuzzy logic deals with uncertainty Uses the mathematical theory of fuzzy sets Simulates the process of normal human reasoning Allows the computer to behave less precisely and logically Decision making involves gray areas and the term maybe Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Fuzzy Logic Advantages
Provides flexibility Provides options Frees the imagination More forgiving Allows for observation Shortens system development time Increases the system's maintainability Uses less expensive hardware Handles control or decision-making problems not easily defined by mathematical models Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Fuzzy Logic Example: What is Tall?
In-Class Exercise Proportion Height Voted for 5’10” 0.05 5'11" 0.10 6’ 6’1” 0.15 6’2” 0.10 Jack is 6 feet tall Probability theory - cumulative probability There is a 75 percent chance that Jack is tall Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

We are not completely sure whether he is tall or not
Fuzzy logic - Jack's degree of membership within the set of tall people is 0.75 We are not completely sure whether he is tall or not Fuzzy logic - We agree that Jack is more or less tall Membership Function < Jack, 0.75  Tall > Knowledge-based system approach: Jack is tall (CF = .75) Belief functions Can Use Fuzzy Logic in Rule-based Systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Fuzzy Logic Applications and Software
Difficult to Apply When People Provide Evidence Used in Consumer Products that have Sensors Air Conditioners Cameras Dishwashers Microwaves Toasters Special Software Packages like FuziCalc Spreadsheet Controls Applications Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Examples of Fuzzy Logic
Example 1: Strategic Planning STRATASSIST - fuzzy expert system that helps small- to medium-sized firms plan strategically for a single product Example 2: Fuzziness in Real Estate Example 3: A Fuzzy System for Bond Evaluation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

AIS In Focus 18.3: Fuzzy Logic Applications
Selecting stocks (on Japanese Nikkei Stock Exchange) Retrieving data (fuzzy logic can find data quickly) Regulating auto antilock braking systems Camera Autofocusing Automating laundry machine operation Building environmental controls Controlling video camcorders image position Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Controlling train motion Identifying killer whale dialects
Inspecting beverage cans for printing defects Keeping space shuttle vehicles in steady orbit Matching golf clubs to customer's swings Regulating shower head water temperature Controlling cement kiln oxygen levels Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Increasing industrial quality control application accuracy and speed
Sorting multidimensional space problems Enhancing queuing (waiting lines) models Decision making (see Glenn [1994]) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

18.10 Cross Fertilization Hybrids of Cutting Edge Technologies
Combine Neural Computing Expert Systems Genetic Algorithms Fuzzy Logic Example: International Investment Management--Stock Selection Fuzzy Logic and ANN (FuzzyNet) to Forecast the Expected Returns from Stocks, Cash, Bonds and Other Assets to Determine the Optimal Allocation of Assets Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Technologies Global markets
Integrated network architecture of the system (Figure 18.12) Technologies Expert system (rule-based) for country and stock selection Neural network for forecasting Fuzzy logic for assessing factors without reliable data Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

FuzzyNet Modules (Figure 18.13)
Membership Function Generator (MFG) Fuzzy Information Processor (FIP) Backpropagation Neural Network (BPN) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

18.11 Data Mining and Knowledge Discovery in Databases (KDD)
Hidden Value in Data Knowledge Discovery in Databases (KDD) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The KDD Process Start with Raw Data and Do
1. Selection to produce target the appropriate data which undergoes 2. Preprocessing to filter the data in preparation for 3. Transformation so that 4. Data Mining can identify patterns that go through 5. Interpretation and Evaluation resulting in knowledge.

Data Mining Find Kernels of Value in Raw Data Ore Theoretical Advances
Knowledge discovery in textual databases Methods based on statistics, cluster analysis, discriminant analysis, fuzzy logic, genetic algorithms, and neural networks Ideal for data mining Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Data Mining Applications Areas
Marketing Investment Fraud Detection Manufacturing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Information Overload Data mining methods can sift through soft information to identify relationships automatically Intelligent agents Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Important KDD and Data Mining Challenges
Dealing with larger databases Working with higher dimensionalities of data Overfitting--modeling noise rather than data patterns Assessing statistical significance of results Working with constantly changing data and knowledge Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Working through missing and noisy data
Determining complex relationships between fields Making patterns more understandable to humans Providing better user interaction and prior knowledge about the data Providing integration with other systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary ANN can be applied to several difficult problems in finance (credit authorization, stock market predictions) ANN can help interpret information in large databases Integrations seem to have many benefits: Expert systems and ANN Fuzzy logic and genetic algorithms Fuzzy logic and ANN Genetic algorithms can be used to solve complex optimization problems Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Fuzzy logic represents uncertainty by using fuzzy sets
Genetic algorithms use a three-step iterative process: Test a solution to see how good it is, Select the best "parents" and Generate offspring. Results improve as knowledge accumulates Fuzzy logic represents uncertainty by using fuzzy sets Fuzzy logic is based on: 1) People reason using vague terms. Classes boundaries are vague and subject to interpretation; 2) Human quantification is often fuzzy Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Data mining methods identify hidden relationships in databases
Fuzzy sets have well defined boundaries. Items have membership values to define the imprecise nature of belonging to a set Data mining methods identify hidden relationships in databases Data mining can boost an organization’s performance by targeting appropriate customers, etc. Intelligent systems and neural computing can overcome information overload by data mining Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. Describe Veratex's problem. 2. How can ANN identify the best potential customers? 3. How can ANN foster market analysis for Veratex? 4. Why is data collection such an important task in building this marketing ANN? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Exercises 2. Express the following statements in terms of fuzzy sets:
a) The chance for rain is 80 percent today (Rain? No rain?) b) Mr. Smith is 60 years old (Young?) c) The salary of the President of the United States is $250,000 per year (Low? High? Very high?) d) The latest survey of economists indicates that they believe that the recession will bottom out in April (20 percent), in May (30 percent) or in June (22 percent) Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3. You are trying to identify a specific number in the set of 1 to 16
3. You are trying to identify a specific number in the set of 1 to 16. You can ask questions such as, "Is this number in the set 1-8?" The answer can only be yes or no. In either case, you continue to ask more questions until you can identify the number. a) How many questions are needed, in the worst and the best possible cases, to identify such a number? b) Is the problem suitable for parallel processing? Why or why not? c) Can you relate this problem to a genetic algorithm? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Exercises 1. Fuzzy logic. Perform a written survey of the class and have everyone write down a height representing “tall” for men, and “tall” for women. Tally up the results and determine what is meant by tall in a fuzzy way. Treat the results like a probability density function. 3. Have the members of your group play (the manual version of) MasterMind for about 30 minutes to one hour. How do the better players in your group win? Write down the game concepts in terms of genetic algorithms and express winning strategies. Do you ever have to try random solutions to converge on a solution? Explain. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 19: Intelligent Agents and Creativity
19.1 Opening Vignettes: Examples of Intelligent Agents Vignette 1: Empowering Employees with Software Agents Nike and Signet Bank installed special software agents Employees access the human resources databases and Select and Change Benefits Make Charity Contributions Other Human Resources Activities Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Electronic Workforce (from Edify Corp.)
Delegates time-consuming and repetitive tasks of human resource (HR) employees to any employee Enables employees to enter, update and delete data; and interpret information Cuts error rate Enabling software is an intelligent agent Vignette 2: Software Agents Cooperating to Provide the Best Travel Plans Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

19.2 Intelligent Agents (IA): An Overview
Intelligent Agent (IA): Computer program that helps a user with routine computer tasks New Technology Other Names Software agents Wizards Knowbots Softbots Agent: Employing someone to act on your behalf Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Several Definitions of Intelligent Agent
“Intelligent agents are software entities that carry out some set of operations on behalf of a user or another program, with some degree of independence or autonomy and in so doing, employ some knowledge or representation of the user’s goals or desires.” (“The IBM Agent” [ An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through effectors (Russell and Norvig [1995], p. 33) Autonomous agents are computational systems that inhabit some complex dynamic environment, sense and act autonomously in this environment and by doing so realize a set of goals or tasks for which they are designed (Maes [1995], p. 108) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

More Definitions of Intelligent Agent
A persistent software entity dedicated to a specific purpose. “Persistent” distinguishes agents from subroutines; agents have their own ideas about how to accomplish tasks, e.g., their own agenda. “Special purpose” distinguishes them from entire multifunction applications; agents are typically much smaller” (Smith et al. [1994]) Intelligent agents continuously perform three functions: perception of dynamic conditions in the environment; action to affect conditions in the environment; and reasoning to interpret perceptions, solve problems, draw inferences and determine actions (Hayes-Roth [1995]) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Possible Components of an Agent
Owner Author Account Goal Subject Description Creation and Duration Background Intelligent System Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

19.3 Characteristics of Intelligent Agents
Autonomy Agent takes initiative, exercises control over its actions Goal-oriented Collaborative Flexible Self-starting Operates in the Background Mobile agents Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Single Task Communication Automates Repetitive Tasks
Supports Conditional Processing Learning Reactivity Proactiveness Temporal Continuity Personality Mobility Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

19.4 Why Intelligent Agents?
Information Overload Data Doubles Annually (in Large Enterprises [1998]) Can analyze only about 5% Most Efforts: Discover patterns, not meaning, not what to do Reduces decision making capabilities by 50% Much Caused by the Internet / Web How to filter data? How to identify relevant sources of data? Intelligent Agents Can Assist Searching Save Time: Agents Decide What is Relevant to the User Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Reasons for Intelligent Agent Technology Growth
Decision Support Repetitive Office Activity Mundane Personal Activity Search and Retrieval Domain Experts Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

19.5 Agent Classification and Types
Taxonomic Tree to Classify Autonomous Agents (Figure 19.1) Relevant to Managerial Decision Making Computational Agents Software Agents Task-specific Agents Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Intelligent Agent Classifications
Control Structure Computational Environment Programming Language Application Type Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Application Types Organizational and Personal Agents
Private Agents vs. Public Agents Software (Simple) Agents and Intelligent Agents Mobile Agents Classification by Characteristics Agency Intelligence Mobility Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Agency: Degree of Autonomy and Authority Vested in the Agent
Key value of agents More advanced agents can interact with other entities Intelligence: Degree of Reasoning and Learned Behavior Mobility: Degree to which agents themselves travel through the network Static Mobile Scripts Mobile with State Nonmobile Agents Defined in 2-D (Figure 19.4a) Mobile Agents Defined in 3-D (Figure 19.4b) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Classification by Intelligence Level and Power (Lee et al. [1997])
Level 0 (Lowest). Retrieve documents directly Level 1. Provide user-initiated searching facility for finding relevant Web pages Level 2. Maintain user’s profiles. Monitor Internet information; notify users when relevant information is found Level 3. Have a learning and deductive component of user profiles to help a user Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Classification by Application Area
Agents that assist in workflow and administrative management Agents that collaborate with other agents and individuals Agents that support electronic commerce Agents that support desktop applications Agents that assist in information access and management Agents that process mail and messages Agents that control and manage the network access Agents that manage systems and networks Agents that create user interfaces Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Agent Classification Internet Based Electronic Commerce Others

19.6 Internet-based Software Agents
Software Robots or Softbots Major Categories Agents (Figure 19.5) Web Browsing Assisting Agents Frequently Asked Questions (FAQ) Agents Intelligent Search (or Indexing) Agents Internet Softbot Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Network Management and Monitoring
Patrol Application Management Tabriz WatchGuard AlertView InterAp Mercury Center’s Newshound Infosage Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

19.7 Electronic Commerce Agents
Help users find information about products or services User provides information directly or indirectly Examples Bargain Finder Finding What Individuals Want: Firefly and Others Good Stuff Cheap (GSC) Other EC Agents Book Worms Bargainbot Eves Resume Robot Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

19.8 Other Agents, Including Data Mining
Representative Examples User Interface Intelligent Agents Monitor the user’s actions Develop models of user abilities Automatically help out Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Operating Systems Agents
Wizards in Microsoft Windows NT Operating Systems Add user accounts Group management Managing file and folder access Add printer Add/remove programs Network client administrator Licenses Install new modems Spreadsheet Agents: Makes software more friendly Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Workflow and Administrative Management Agents
Ascertain and automate user needs or business processes Example - FlowMark Software Development Many routine tasks can be done or supported by agents Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Data Mining One of the most important capabilities of information technology Can sift through large amounts of information Challenge: intelligent agents to sift and sort Categories Intelligent agents Query-and-reporting tools Multidimensional-analysis Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Web Mining Subsets (Etzioni [1996]) Resource Discovery
Information Extraction Generalization Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

NewsAlert Monitors data by personalized rules
Automatically delivers alerts to the user’s desktop into personalized newspapers Organizes alerts by user specified subject areas Provides smart tools so users can investigate the context of an alert and communicate findings to others Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Key Components of NewsAlert:
Software Agents Alert Objects Newspaper Client Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Electronic Newspapers
Combine Features of a Paper Newspaper Familiar Format Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Collaboration by Agents
Lotus Notes: Comprehensive collaborative software Includes Notes Agents: automates many Notes tasks Agents operate in the background performing routine tasks Agents can be created by designers within an application Agents can either be private or shared Collaboration: Natural area for agent-to-agent interaction and communication Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

19.9 Multiple Agents and Distributed AI
Software agents must communicate with each other Refine requests and queries through evolving dialogue Intelligent agents work together in multiple agent systems Agents can communicate, cooperate and/or negotiate Easy to build agents with small specialized knowledge But complex tasks require much knowledge Agents need to share their knowledge Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Figure 19.8 A Multiagent System for Travel Arrangements

Routing in Telecommunication Networks
Agents control a telecommunications network Can enter into agreements with other computers that control other networks about routing packets more efficiently Agent in a blackboard architecture Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

More Multiple Agents Personal digital assistants (PDA)
Shared (global) databases Agents (softbots) travel out on the Internet and collect information from shared databases Traffic control Coordination of vehicular traffic Air traffic control The University of Massachusetts CIG Searchbots Software agents make decisions based on communication and agreements with other agents Soon: Agents coordinating sellers and buyers Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Topics in Multiagent Systems
Negotiation in Electronic Commerce Coordination The Nature of the Agents Learning Agents Cooperation and Collaboration Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

19.10 Software-supported Creativity
Computer programs that Exhibit creative behavior Facilitate human creativity Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Creative Software Programs
Major characteristic of intelligent behavior is creativity Can computers be creative? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Three Creativity Software Tools (Rasmus [1995])
1. Copycat: Seeks analogies in patterns of letters (Identifying patterns is the essence of intelligence) 2. Tabletop Individual agents compete for correctness of fit Copycat’s internal architecture: Cohesive parallelism Converge upon an answer From partial ones While finding analogies. 3. AARON can draw real artwork But really need multiple agents to expand this area Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Computerized Support for Creativity and Idea Generation
There are many semistructured and unstructured situations for which not all the alternative courses of action are known Idea generation is frequently necessary in DSS Idea generation is part of problem solving To generate good ideas, people need to be creative Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

New idea generation methods are under development
Past conventional wisdom: An individual’s creative ability came from personality traits New studies indicate: Individual creativity can be learned and improved Innovative companies recognize that to foster creativity, create an idea-nurturing work environment New idea generation methods are under development Manual idea generation can be very successful But, may not be economically feasible nor possible Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Problems with Manual Idea Generation
1. Single decision maker 2. Poor (or no) facilitator 3. Not enough time for proper idea generation 4. Too expensive to conduct an idea generation session 5. Too sensitive a subject 6. Not enough participants, nonoptimal participant mix or not a positive climate for idea generation Then: induce idea generation electronically Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Idea Generation Software
Designed to help stimulate a single user or a group with new ideas, options and choices An electronic brainstorming tool based on synergy (and/or association) The user does all the work Software encourages and pushes like a personal trainer Several software packages on the market (See Book’s Web Site) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Idea Generation in GDSS
Allows participants to generate ideas simultaneously Recent GSS development: intelligent agent facilitator Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

19.11 Managerial Issues Cost Justification
Stand-alone agents perform complex tasks; but can be quite expensive (tremendous amount of R&D to create) Security (of Systems) Privacy Industrial Intelligence and Ethics Other Ethical Issues Agent Learning Agent Accuracy Heightened Expectations System Acceptance System Technology Agent Development Tool Kits (see Book’s Web Site) Strategic Information Systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Conclusions Agents can simplify our use of computers
Agents can provide friendly software assistance Agents promise to hide complexity Agents perform actions we do not do ourselves Agents could enhance human intelligence Agents provide support to Net users in handling the information overload problem Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

But: Danger! Agents are unlike other technological advances
Agents have some level of intelligence, some form of Self-initiated and Self-determined goals There is the potential for Social mischief Systems that run amok Loss of privacy Further alienation of society Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Can Eliminate Such Problems
Develop rules for well-behaving agents Determine the accuracy of information collected Respect restrictions of other servers Do only authorized work

Summary Several definitions of Intelligent Agents (IA): Software entities that perform tasks with some degree of autonomy IA can save time and are consistent. They have varying levels of autonomy Major characteristics of IA: Autonomy, operating in background, communication capabilities and reactivity More autonomous agents must be able to learn and improve their actions The major purpose of IA is to deal with information overload However, agents can improve productivity, quality and speed Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Agents can be classified in several ways, depending on their mission
Mobile agents can perform tasks in different locations. Other agents work in one place (e.g., a server, workstation). Agents can be classified into 3 major applications types: Internet, EC and others Multiagent systems can be used to execute more complex tasks than single agents, but they have not yet matured Intelligent agents play a major role in data mining, helping to finding appropriate data and knowledge quickly and answers queries Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. List the benefits that occurred at Signet Bank through intelligent agents. 2. List the benefits that occurred at Nike through intelligent agents. 3. How can the employees save time by working with intelligent agents? How would you feel about interacting with an intelligent agent instead of a human being? What if the cost of the service is 25% lower? What if the service cannot reasonably be offered otherwise? 4. How do you feel about making travel arrangements with a machine instead of with a travel agent? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Exercises 3. How are IA actually constructed? Investigate the literature and write a report. 4. What mundane tasks would you like an intelligent agent to perform for you? List them (you may want to include some tasks that people are handling for you), and describe how an IA could help. Compare your results to those of other members in the class. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Exercise Contact Microsoft and/or their vendors. Find what IA Wizards do for the improved use of operating systems, spreadsheets and other software products. Prepare a report on recent developments. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Part 6: Implementation, Integration and Impacts
Chapter 20: Implementing and Integrating MSS Chapter 21: Organizational and Societal Impacts of MSS (Sorry, No Part Initial Page, Check and click on Corrections) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 20: Implementing and Integrating MSS
Building MSS First phase: decision making support and problem solving Implementation Integration of MSS Technologies Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

20.1 Opening Vignette: INCA Expert Systems for The SWIFT Network
Society for Worldwide Interbank Financial Telecommunication (S.W.I.F.T.) Network International message-processing and message transmission services between financial institutions Want to automate the day to day control of the network Replace teams of operators working switches Two Control Centers: USA and Holland: 8 groups Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Real-time decision making system necessary for network control
The system could not fail Intelligent Network Controller Assistant (INCA) performs Filters incoming events and diagnoses problems Displays problems requiring attention Core business application Working prototypes not used Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Allow the flexibility required for AI development
Special software development methodology developed and followed for tight quality control Allow the flexibility required for AI development Incremental extraction of knowledge from experts not possible Hardware and software: standard workstations Future users and experts were involved in every implementation phase User training plans created and implemented simultaneously with system development Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Object-oriented paradigm Events trigger rules to fire
Temporal information was also considered INCA dynamically updates its model of the S.W.I.F.T. network on line. INCA developed by a core team of five; three from corporate research, two from operations Initially also two knowledge engineers from TI Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

INCA Implementation Timeline
Early 1989: Brief prototyping exercise April 1989: INCA started October 1989: First deployment in one control center Processing functions introduced in modular phases to minimize risks February 1990: INCA fully operational in the Holland Center May 1990: Fully operational in the USA Center Summer 1990: Maintenance gradually turned over to internal system support group INCA is working well and well accepted Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Criteria for Successful Implementation
INCA should achieve staff savings INCA should be accepted by the client group for Its functions Its quality Improve response times Limit network downtime Results INCA can automatically handle 97 % of all events Now only two INCA teams; each one can replace the original eight teams Estimated reduction in staff: 50 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The INCA development team: Victim of Success!
Managers want additional features and functions Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

20.2 Implementation: An Overview
INCA - Major Points About Systems Implementation Situation where standard methods do not work Custom implementation methods must be designed, tested and implemented Users must be involved in every phase of the development Management support is crucial (though not mentioned) Experts must be cooperative Criteria for success were clearly defined Large-scale, real-time ES can be developed on schedule and be very reliable Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Introduction MSS systems implementation is not always successful
Expert systems fail often Implementation is an ongoing process of preparing an organization for the new system And introducing the system to assure success. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

MSS implementation is complex
MSS are linked to tasks that may significantly change the manner in which organizations operate But, many implementation factors are common to any IS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

What Is Implementation?
There is "nothing more difficult to carry out, nor more doubtful of success, nor more dangerous to handle, than to initiate a new order of things" (Machiavelli) The introduction of change Implementation is a long, involved process with vague boundaries Implementation can be defined as getting a newly developed or significantly changed, system to be used by those for whom it was intended Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Ongoing process during the Entire Development
MSS Implementation Ongoing process during the Entire Development Original suggestion Feasibility study Systems analysis and design Programming Training Conversion Installation For MSS: Iterative Nature of Development Complicates Matters Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Can have Partial Implementation
Institutionalization: MSS implementation means commitment to routine and frequent system use Ad hoc decisions: MSS implementation means the one-time use of the system Can have Partial Implementation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Measuring Implementation Success
Indicators 1. Ratio of actual project execution time to the estimated time 2. Ratio of actual project development cost to budgeted cost 3. Managerial attitudes toward the system 4. How well managers' information needs are satisfied 5. Impact of the project on the computer operations of the firm Dickson and Powers [1973] Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Other MSS Success Measures
System Use User satisfaction Favorable attitudes Degree to which system accomplishes its original objectives Payoff to the organization Benefits to costs ratios Degree of institutionalization of MSS in the organization Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Additional Measures of ES Success
Degree to which the system agrees with a human expert Adequacy of the system’s explanations Percentage of cases submitted to the system for which advice was not given Improvement of the ES on the learning curve (speed to maturity) Guimaraes et al. [1992] and Sprague and Watson [1996]) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Contributing Factors to DSS Success
User involvement User training Top management support Information source Level of managerial activity being supported Characteristics of the tasks involved (structure, uncertainty, difficulty, interdependence) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

MSS Implementation Failures
Usually a closely held secret in many organizations Expected synergy of human and machine not developed Managers unwilling to use computers to solve problems Not much formal data on MSS failures Many informal reports on unsuccessful implementation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

20.3 The Major Issues of Implementation
Models of Implementation Many factors can determine the degree of success of any IS Factor or success factor - Important Generic Specific Determinants of successful implementation (next) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Success Factors of Implementation
(Figure 20.1) Technical Factors Behavioral Factors Change Management Process and Structure User Involvement Ethics Organizational Support External Environment Project Related Factors Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Technical Factors Relate to the mechanics of the implementation procedure (Table 20.1) Two Categories Technical Constraints Technical Problems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Behavioral Factors CBIS Implementation affected by the way people perceive systems and by how people behave Resistance to Change Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Process Factors Top Management Support (One of the most important)
Need for continuous financial support for maintenance Few studies on methods to increase top management MSS support Management and User Commitment Institutionalization Length of Time Users Have Been Using Computers and MSS Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

User Involvement Participation in the system development process by users or representatives of the user group Determining when user involvement should occur and how much is appropriate need more research In user-developed systems, the user obviously is very much involved With teams, involvement becomes fairly complex Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Joint Application Development (JAD) procedure strongly recommended
DSS Development: Heavy user involvement throughout the developmental process with a much direct management participation Joint Application Development (JAD) procedure strongly recommended Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Organizational Factors
Competence (Skills) and Organization of the MSS Team Responsibility for DSS Development and Implementation Adequacy of Resources Relationship with the Information Systems Department Organizational Politics Other Organizational Factors Role of the system advocate (sponsor) initiator Compatibility of the system with organizational and personal goals of the participants Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Management is Responsible
Values and Ethics Management is Responsible Project Goals Implementation Process Possible Impact on Other Systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

External Environment Factors Outside the Immediate Area of the Development Team, Including Legal Factors Social Factors Economic Factors Political Factors (e.g., Government Regulations) Other Factors (Positive or Negative) Up to Now - Implementation Climate Issues Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Project-related Factors
Evaluate each project on its own merits Relative importance to the organization Its members Cost-benefit criteria Other Project Evaluation Dimensions Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Other Project Evaluation Dimensions
Important or major problem needing resolution Real opportunity needing evaluation Urgency of solving the problem High-profit contribution of the problem area Contribution of the problem area to growth Substantial resources tied to the problem area Demonstrable payoff if problem is solved Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Expectations from a Specific System
Users have expectations as to how a system will Contribute to Their Performance Rewards Can Affect Which System is Used Over-expectations Dangerous Observed in AI Technologies Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Cost-benefit Analysis
View application as an alternative investment Application should show a payoff an advantage over other investment alternatives Since mid-1980s, IS justification pressures have increased Effective implementation depends on effective justification Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Other Items Project Selection Project Management
(Critical for ES) Project Management Availability of Financing and Other Resources Timing and Priority Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

20.4 Implementation Strategies
Many implementation strategies Many are generic Can be used as guidelines in implementing DSS ES Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Implementation Strategies for DSS
Major Categories Divide the project into manageable pieces Keep the solution simple Develop a satisfactory support base Meet user needs and institutionalize the system (Table 20.4) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Expert System Implementation
Especially important in ES implementation Quality of the system Cooperation of the expert(s) Conditions justifying the need for a particular ES Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Quality of the Expert System
1. The ES should be developed to fulfill a recognized need 2. The ES should be easy to use (even by a novice) 3. The ES should increase the expertise of the user 4. The ES should have exploration capabilities 5. The program should respond to simple questions 6. The system should be capable of learning new knowledge 7. The knowledge should be easily modified Necessary, but not sufficient features for success Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Some Questions About Experts' Cooperation
Should the experts be compensated for their contribution? How can one tell if the experts are truthful? How can the experts be assured that they will not lose their jobs, or that their jobs will not be de-emphasized? Are the experts concerned about other people whose jobs may suffer, and if so, what can management do? Use incentives to influence the experts to ensure cooperation Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Some Conditions That Justify an ES
An expert is not always available or is expensive Decisions must be made under pressure, and/or missing even a single factor could be disastrous Rapid employee turnover resulting in a constant need to train new people (costly and time-consuming) Huge amount of data to be sifted through Shortage of experts is holding back development and profitability Expertise is needed to augment the knowledge of junior personnel Too many factors--or possible solutions--for a human to juggle Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

More Conditions Problem requires a knowledge-based approach and cannot be handled by conventional computing Consistency and reliability, not creativity, are paramount Factors are constantly changing Specialized expertise must be made available to people in different fields Commitment on the part of management User involvement Characteristics of the knowledge engineer Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

20.5 What Is Systems Integration and Why Integrate?
Not separate hardware, software and communications for each independent system At development tools level or application system level Two General Types of Integration Functional Physical Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Integration Types Functional Integration Physical Integration
(Our primary focus) Different support functions are provided as a single system Physical Integration Packaging hardware, software and communication features required together for functional integration Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

for MSS Software Integration
Why Integrate? Two Major Objectives for MSS Software Integration Enhancements of Basic Tools Increasing the Applications’ Capabilities Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Integrating DSS and ES Mutual Benefits Each Technology Provides (Table 20.5) Integrating DSS, ES and EIS (health care industry) Integrating medical expert systems, patient databases and user interfaces using conventional tools: PACE, a comprehensive expert consulting system for nursing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Two General Types of Integration
Different Systems (e.g., ES and DSS) Same Type Systems (e.g., Multiple ES) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

20.6 Models of ES and DSS Integration
Names ranging from expert support systems to intelligent DSS Models ES Attached to DSS components ES as a Separate DSS Component ES Generating Alternative Solutions for DSS Unified Approach Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Expert Systems Attached to DSS Components
Five ES (Figure 20.2) 1: Intelligent database component 2: Intelligent agent for the model base and its management 3: System for improving the user interface 4: Consultant to DSS builders 5: Consultant to users Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

ES as a Separate DSS Component
Architecture for ES and DSS integration (Figure 20.3) ES is between the data and the models to integrate them Integration is Tight But can be over Communications Channels like the Internet Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

3 Possible Integration Configurations
ES Output as Input to a DSS DSS Output as Input to ES Feedback (both ways) (Figure 20.4) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Sharing in the Decision-making Process
ES can complement DSS in the decision-making process (8-step process) 1. Specification of objectives, parameters, probabilities 2. Retrieval and management of data 3. Generation of decision alternatives 4. Inference of consequences of decision alternatives 5. Assimilation of verbal, numerical and graphical information 6. Evaluation of sets of consequences 7. Explanation and implementation of decisions 8. Strategy formulation Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1-7: Typical DSS functions
8: Requires judgment and creativity - can be done by ES ES supplements the DSS with associative memory with business knowledge and inferential rules. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

20.7 Integrating EIS, DSS, and ES, and Global Integration
EIS and DSS EIS is commonly used as a data source for PC-based modeling How? EIS-generated information as DSS input DSS feedback to the EIS and possible interpretation (AIS In Action 20.3) and ES explanation capability Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Global Integration May include several MSS technologies
Comprehensive system conceptual architecture (Figure 20.5) Inputs Processing Outputs Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Outputs. User Can Generate
1. Visually attractive tabular graphic status reports that describe the decision environment, track meaningful trends and display important patterns 2. Uncontrollable event and policy simulation forecasts 3. Recommended decision actions and policies System graphically depicts the reasoning explanations and supporting knowledge that leads to suggested actions Feedback loops to provide additional data, knowledge, and enhanced decision models Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Global Integrated System Example
To connect the MSS to other organizations - EDI and Internet (Figure 20.6) Corporate MSS includes DSS and ES Internet-based videoconferencing system for group-work EDI for transaction processing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

20.8 Intelligent Modeling and Model Management
Add intelligence to Modeling and Management Tasks require considerable expertise Potential benefits could be substantial Integration implementation is difficult and slow Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Issues in Model Management
Problem Diagnosis and Model Selection Model Construction (Formulation) Models Use (Analysis) Interpretation of Models' Output Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Quantitative Models Proposed architecture for quantitative intelligent model management (Figure 20.7) Human experts often use quantitative models to support their experience and expertise Many models are used by experts in almost all aspects of engineering Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

ES Contributions in Quantitative Models and Model Management
Demonstrate by examining the work of a consultant 1. Discussing the nature of the problem with the client 2. Identifying and classifying the problem 3. Constructing a mathematical model of the problem 4. Solving the model 5. Conducting sensitivity analyses with the model 6. Recommending a specific solution 7. Assisting in implementing the solution Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Some ES research is moving in this direction
System involves a decision maker (client), a consultant and a computer. If we can codify the knowledge of the consultant in an ES, we can build an intelligent computer-based information system capable of the same process But - Hard to do Some ES research is moving in this direction ES can be used as an intelligent interface between the user and quantitative models There are several commercial systems to assist with statistical analysis Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

20.9 Examples of Integrated Systems
Manufacturing Marketing Engineering Software Engineering Financial Services Retailing Commodities Trading Property-casualty Insurance Industry Decision Making Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Manufacturing Integrated Manufacturing System
Logistics Management System (LMS) - IBM Combines expert systems, simulation and decision support systems And computer-aided manufacturing and distributed data processing subsystems Provides plant manufacturing management a tool to assist in resolving crises and help in planning Similar system at IBM by financial analysts to simulate long-range financial planning Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Embedded Intelligent Systems
Combination of several, complex expert systems (implemented as intelligent agents) with a scheduling system and a simulation-based DSS for rescheduling production lines when problems occur Embedded Intelligent Systems Data mining systems Others Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

DSS/Decision Simulation (DSIM - IBM). Integration provides:
Ease of communication Assistance in finding appropriate model, computational algorithm or data set Solution to a problem where the computational algorithm(s) alone is not sufficient to solve the problem, a computational algorithm is not appropriate or applicable and/or the AI creates the computational algorithm Intelligent Computer Integrated Manufacturing Error recovery in an automated factory MSS in CAD/CAM Systems Comprehensive CIM System (Table 20.6) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Marketing Promoter TeleStream

Engineering STRUDL

Software Engineering CREATOR2: CASE Tools with ES CREATOR3

Financial Services Integrated system to match services with customers' needs Credit evaluation Strategic planning FINEXPERT American Express Inference Corp. System (Figure 20.8)

Retailing Buyer's Workbench Deloitte and Touche for Associated Grocers
(Figure 20.9) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Commodities Trading Intelligent Commodities Trading System (ICTS)
(Figure 20.10) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Property-casualty Insurance Industry Decision Making
Decision making for insurance industry based on forecasting Major decisions involve Determining what products to offer Pricing of products Determining territories to operate Deciding how to invest premium money collected Integrated ES-ANN system combined with a DSS (Figure 20.11) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Flow Chart Shows the Roles of Each Major Component
1. DSS provides statistical analysis and graphical display 2. ANN analyzes historical data and recognizes patterns 3. Results generated by the DSS and by the ANN to ES for interpretation and recommendation Recommendations are tested by the DSS using "what-if" Condensed from Benjamin and Bannis [1990] Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

20.10 Problems and Issues in Integration
Need for Integration Justification and Cost-benefit Analysis Architecture of Integration People Problems Finding Appropriate Builders Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Attitudes of Employees of the IS Department
Part of the problem is cultural Development Process Organizational Impacts Data Structure Issues Data Issues Connectivity Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary Many MSS projects fail or are not completed
Many factors determine successful implementation Implementation is an ongoing process Implementation means introducing change Partial success of implementation is possible; usually measured by several criteria Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Organizational climate and politics can be detrimental to success
Technical success is related to system's complexity, reliability and responsiveness; hardware, network and software compatibilities; and technical skills of builders Organizational climate and politics can be detrimental to success Many dimensions to change and to its resistance; overcoming resistance is complex Importance of user involvement varies depending on the MSS technology Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Lack of adequate resources means failure
Several organizational factors are important to successful implementation Lack of adequate resources means failure Medium and large MSS projects must go through a rigorous cost-benefit analysis. Many benefits are intangible Functionality of conventional CBIS may be increased when MSS technology is integrated with them Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Functional integration differs from the physical integration
Intelligent databases are a major integration area of databases (and DBMS) with ES and NLP ES can simplify databases accessibility Major area of integration: ES to interpret results of data generated by models ES can enhance knowledge management and model management Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

ES are being successfully integrated with DSS
There are several conceptual models of integration of ES and DSS MSS technologies are being integrated with many CBIS There are many problems with integrating AI technology: technical, behavioral and managerial factors Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1.Why was INCA not allowed to fail? That is, what were the consequences of failure? 2.Why couldn’t rapid prototyping be used as an implementation method? 3.Describe some of the unique aspects of INCA that required a modified development approach. 4.Describe some difficulties in developing a system design methodology while developing a system. 5.Why did the object-oriented approach make sense here? Do you think that INCA used a variant of forward chaining or backward chaining for its inference engine? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Exercise Administer the questionnaire to users of DSS in your class!
Comment on the results Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Exercise 3. Meet and discuss ways in which intelligence could be integrated into your university’s advising and registration system. Are there any concrete ways in which it could be accomplished at low cost and relatively fast? Explain. Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

CASE APPLICATION 20.1: Urban Traffic Management
Case Questions 1. Why is it necessary to employ a DSS in this case? 2. Why is it necessary to include knowledge bases? 3. Why are there different databases, knowledge bases and model bases? 4. How can GIS be incorporated into such a system? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Chapter 21: Organizational and Societal Impacts of Management Support Systems

21.1 Opening Vignette: Police Department Uses Neural Networks to Assess Employees
Chicago Police Internal Affairs used neural network software to predict whether an officer might act improperly Model trained comparing characteristics of all officers to 200 officers who did or might have acted improperly Matches classified as heading for trouble Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Application - Fairly accurate Software created major debate
The Chicago Police Dept. says - Can’t manually check all officers Early detection helps Leads to a counseling program to fix the situation Computer program can screen many people periodically Cannot be biased Neural networks are helpful in similar tasks Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

BUT Labor union unhappy “System is unethical
It was a tactic to avoid managing the officers Neural networks are a black box: How do they work? It’s not fair! Software Developer: “Users don’t need to know what the software is doing--they only need to know whether it works.” The Chicago Police Department believes that the computer program works very well What do you think??? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Opening Vignette Demonstrates that an MSS
21.2 Introduction Opening Vignette Demonstrates that an MSS 1. Can radically change the decision making process 2. There is resistance to new technology 3. The value of technology is debatable 4. Introduction of an MSS application may have multiple impacts Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

MSS MSS are important enablers of the Information Revolution
Unlike slower revolutions (Industrial Revolution) Much faster Affecting our entire lives Many managerial and social problems Impact on organizational structure Resistance to change Possible rapid increased unemployment levels etc. MSS - Maybe 30% of the IT market by 2000 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Hard to separate the impact of MSS from other computerized system
Trend to integrate MSS with other CBIS Little published information about MSS impacts Techniques are so new E.g., First: The Internet Now: The World Wide Web What Next ??? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

MSS Impacts MSS can have both micro- and macro-implications
MSS can affect Particular individuals and jobs The work structure of departments Units within the organization MSS can have significant long-term effects on Total organizational structures Entire industries Communities Society as a whole. Complete Management System Framework (Figure 21.1) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Movements of Major Changes
Organization Transformation Business Process Reengineering (BPR) The support of IT to BPR was voted as the most important issue of information management in 1994/1995 ((Brancheau et al. [1996]) Information technology is an enabler of BPR (Hammer and Champy [1993]) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.3 Overview of Impacts General Categories: Organizational and Societal Organizational Impacts (Table 21.1) Social Impacts (Table 21.2) Computer technology has already changed our world Much more change is anticipated Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.4 Organization Structure and Related Areas
Degree of Centralization of Authority Distribution of Power and Status New Organizational Units Organizational Culture Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Structure Flatter Organizational Hierarchies
Staff-to-Line Ratio Increasing Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Centralization of Authority
Difficult to establish a clear pattern of IT influence on authority and power IT can support either centralization or decentralization (DSS In Focus 21.1) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Power and Status Knowledge is power
Developments in IS are changing the power structure within organizations Who will control the computers and information resources? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

New Organizational Units
DSS Department Management Support Department AI Department (AIS In Action 21.3) Knowledge Management Department (Headed by a Chief Knowledge Officer (CKO)) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Organizational Culture
Can impact the diffusion rate of technology Can be influenced by it Some dissolution of organizational structure due to technology Virtual teams can meet anytime / anyplace Individuals can join a virtual team as needed Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.5 MSS Support to Business Process Reengineering
Business Process Reengineering (BPR) Major Innovation Changing the way organizations conduct business Involves Changes in Structure Organizational Culture Processes Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

BPR greatly changes organizational structure
BPR Creates Management realignments Mergers Consolidations Operational integrations Reoriented distribution practices BPR greatly changes organizational structure Team-based organizations Mass customization Empowerment Telecommuting MSS is an enabler Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

(Especially ES, DSS and EIS)
MSS (Especially ES, DSS and EIS) Allows business to be conducted in different locations Provides flexibility in manufacturing Permits quicker delivery to customers Supports rapid and paperless transactions ES enable organizational changes by providing expertise to nonexperts (Figure 21.2) Simulation Modeling and BPR (Case Applic. 21.1) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.6 Personnel Management Issues
Role of Employees and Managers Many role definitions will be changed New jobs (knowledge engineers) Some jobs will disappear Top management support staff moving to information specialists Interesting changes in the jobs of experts supported by ES (AIS In Action 21.5) Job Content Role Ambiguity and Conflict Employee Career Ladders Changes in Supervision Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Other Considerations Impacts of MSS
On job qualifications? On training requirements? On worker satisfaction? How can jobs be designed to be a challenge? How might MSS be used to personalize or enrich jobs? What can be done so MSS does not demean jobs or has other negative impacts? How to allocate functions to people and machines? Should cost or efficiency be the major criterion for such allocation? What is the role of the Human Resources Department in a Virtual Organization? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.7 Impact on Individuals Job Satisfaction
Inflexibility and Dehumanization Cooperation of Experts Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.8 Productivity, Quality, and Competitiveness
Major MSS Benefits Leading to Competitive Advantage Increased productivity Increase in quality Cost reduction Timely production Faster time to market Fast training of employees Increased production (service) capacity Unique services Enable BPR and organization transformation Enhance other computer systems Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.9 Decision Making and the Manager's Job
Impact on the manager's job since the 1960s Until now mainly at lower- and middle-levels Now MSS impact at top manager's job MSS can change how managers make decisions So, MSS can change managers' jobs Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Impacts of MSS on Decision Making
Automation of routine decisions or decision making phases Less expertise (experience) required for many decisions Faster-made decisions Less reliance on experts to provide support to top executives Power redistribution among managers Support to complex decisions, making them faster and of better quality Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Provide information for high-level decision making
MSS frees managers from routine tasks and decision making AI technologies can improve environmental scanning of information Change in leadership requirements Methods that managers use to do their jobs will change Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.10 Institutional Information Bases, Knowledge Bases, and Knowledge Management
Possible development of very large and complex information and knowledge bases that require trained expertise to maintain and use Organizational intelligence will become a critical issue Knowledge stored in knowledge bases is accumulating fast Intelligent agents could help Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Some Questions For Successful Implementation
How will the availability of knowledge affect strategic plans? How will the communication stream be affected? Will results of decisions be as readily communicated to peers, subordinates, and superiors by managers, who may assume that these people have and take advantage of the access to information bases? How will managers be trained to make effective use of these new tools? What needs to be done to assess the current competency of managers and to match the tools of these competencies? Implementation called knowledge management Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Knowledge Management Companies collect large amounts of knowledge about problem solving, treating customers, working with the government, etc. This knowledge, if properly stored and organized, can be shared for the benefit of the organization and its members Can view an organization as a human community - collective wisdom gives a distinctive edge against competitors Best way is called knowledge management Technologies that support it are the Intranet, Internet, data warehousing, groupware, and data access and mining tools Technology tools that support knowledge management: know-ware tools: DecisionSuite, WINCITE, KnowledgeShare, SolutionBuilder and grapeVINE. Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

KM related to THE LEARNING ORGANIZATION
To help the knowledge worker be more effective, companies creating formal knowledge bases that contain Lists of experts Information maps Corporate yellow pages Custom desktop applications Other systems KM related to THE LEARNING ORGANIZATION Reuters: global knowledge development and reuse project in developing and deploying a global knowledge organization Also, see DSS In Action 21.6 Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

The Learning Organization
Implies an organizational memory and a means to save, represent and share it Organizational learning is the development of new knowledge and insights that have the potential to influence behavior Intelligent agents and interfaces, and hypermedia, especially via the Web, have the capacity to transform organizations Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

For Organization Learning to be Successful, Important Critical Success Factors
Orientation of the developers toward organizational issues Development focus that favors process over product Development paradigm based on kernels from social science theories View of expertise in its organizational context by the developers Good developer expert interaction IT is playing an ever important role in organizational learning See and Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Organizational Computing
Emerging field of study Could have profound socioeconomic implications for organizations and individuals Knowledge Warehouse: Like data warehouse with OLAP Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.11 Issues of Legality, Privacy, and Ethics Legality
Liability for the actions of intelligent machines are just A computer as a form of unfair competition in business (airline reservation systems) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Some Legal Questions Who is liable if an enterprise finds itself bankrupt as a result of using the advice of ES? Will the enterprise itself be held responsible for not testing such systems adequately before entrusting them with sensitive issues? Will auditing and accounting firms, share the liability for failing to apply adequate auditing tests? Will the manufacturers of intelligent systems be jointly liable? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Specific Legal Issues What is the value of an expert opinion in court when the expertise is encoded in a computer? Who is liable for wrong advice (or information) provided by an ES? What happens if a manager enters an incorrect judgment value into an MSS and the result is damage or a disaster? Who owns the knowledge in a knowledge base? Should royalties be paid to experts who provide the knowledge to ES, and if so how much? Can management force experts to contribute their expertise? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Representative Issues in Ethics
Computer abuse and misuse Electronic surveillance Software piracy Invasion of individuals' privacy Use of proprietary databases Exposure of employees to unsafe environments related to computers Computer accessibility for workers with disabilities Accuracy Accessibility to information Liability of programmers and other IS employees Use of corporate computers for private purposes How much decision making to delegate to computers Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Four Topics of Ethics (Figure 21.3)
Personal Values Major factor in ethical decision making Ethical issues in MSS is complex (multidimensionality) Four Topics of Ethics (Figure 21.3) Accuracy Property Accessibility Privacy Mason et al. [1995] Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Privacy New computer systems can affect privacy rights
Confidential information can be misused Can result in invasion of privacy and other injustices Cookies - New Issue (DSS In Focus 21.7) Law Enforcement - Use of AI technologies Other AI implications Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.12 Intelligent Systems and Employment Levels
Intelligent systems / MSS can affect productivity and employment AI (and ES and ANN) will increase the productivity of knowledge workers Impact on the aggregate employment level? Massive unemployment? (Wassily Leontief) Increased employment? (Herbert Simon) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Massive Unemployment 1. The need for human labor will be reduced significantly 2. The skill levels of people performing jobs with the help of AI will be low 3. AI will affect both blue- and white-collar employees in all sectors 4. In the past few years (in 1991) several industries have laid off many employees 5. Industry, government and services already have a lot of hidden unemployment 6. Unemployment levels have grown steadily in the past decade in spite of increased computerization 7. The per capita amount of goods and services that people can consume is limited - may stop growing Unemployment (DSS In Focus 21.8) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Increased Employment Levels
1. Historically, automation has always resulted in increased employment, by creating new occupations (AIS In Focus 21.9) 2. Unemployment is worse in unindustrialized countries. 3. Work, especially professional and managerial, can always be expanded 4. The task of converting to automated factories and offices is complex - may take several generations 5. Many tasks cannot be fully automated Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

6. Machines and people can be fully employed, each where appropriate
7. Real wages may be reduced, however, because people will have income from other sources; people will have enough money to spend to create more jobs 8. The cost of goods and services will be so low that demand will increase significantly (automation will never catch up with increased demand) Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Other Questions Is some unemployment really socially desirable?
Should the government intervene more in the distribution of income and in the determination of the employment level? Can the "invisible hand" in the economy continue to be successful in the future? Will AI make most of us idle but wealthy? (AIS In Action 21.10) Should the income issue be completely separate from employment? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.13 Other (Potential) Societal Impacts Positive Effects
Work in Hazardous Environments Opportunities for the Disabled Changing Role of Women. Working at Home (Telecommuting) Improvements in Health Aids for the Consumer Quality of Life Law Enforcement Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Negative Effects Unemployment Creation of large economic gaps
Other negative situations Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Computer Crime: Fraud and Embezzlement
Losses as much as US $ 45 billion / year ES can deliberately provide bad advice DSS, ES and neural computing to detect and prevent computer crimes Neural computing : Detect stolen credit cards and cellular phones almost instantaneously Too Much Power Blaming the Computer Phenomenon Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

21.14 Managerial Implications and Social Responsibilities
What can management do? How to anticipate the broad societal effects of MSS? What to do to ensure that people's attitudes toward MSS are well founded and that their expectations are reasonable? How to determine potential positive and negative beforehand? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Key Issues Social Responsibility Public Pressure
Computer and Staff Resources Planning Electronic Community Related to electronic commerce Electronic communities will change the nature of corporate strategy and how business is done Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

MSS Summary and Conclusions
MSS are having far reaching and dramatic impacts on society and organizations Impacts Providing rapid information access Instantaneous communication Artificial intelligence assisting and replacing human effort Revolution of Technology Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Summary MSS can affect organizations in many ways
Flatter organizational hierarchies The impact of MSS on the degree of centralization of power and authority is inconclusive. MSS could cause a power redistribution Special intelligent systems units and departments may appear Many jobs will require fewer skills Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

MSS could reduce the need for supervision
The job of the surviving expert will become more important: custodian of the ES and the knowledge base MSS could reduce the need for supervision The impact of MSS on individuals is unclear; may be either positive or negative Organizational data and knowledge bases will be critical issues Serious legal issues may develop with AI: liability and privacy One view, intelligent systems will cause massive unemployment Continue Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Another view, intelligent systems will increase employment levels
Many positive social implications can be expected Quality of life, both at work and at home, is likely to improve Electronic (virtual) communities are evolving Managers need to plan for the MSS of the future to be ready to make the most of them Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

1. Is this another example where the needs of the society are in conflict with the rights of the individual? 2. Is using the neural network computer program ethical? What if a statistical approach were used? Is that ethical? 3. If you were an officer being evaluated, would you object to such a program? Why? 4. Is it fair for the Police Department to use the neural network program to screen new applicants? If so, what is the problem with using such a program retroactively? 5. As a citizen, would you want your police department to use the program? Why or why not? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Group Exercise Knowledge Management Exercise
Debate Effectiveness of Information Systems and Ethical Issues Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

Case Application 21.1: Xerox Re-engineers its $3 Billion Purchasing Process with Graphical Modeling and Simulation Case Questions 1. Why is it so important to reduce cycle time in organizational processes? 2. What were the benefits of using simulation to model and test the proposed plan? 3. How did the visual nature of the simulation help the decision makers? 4. What other MSS tools might be combined with the simulator to make the decision making more effective? Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson Copyright 1998, Prentice Hall, Upper Saddle River, NJ

EFRAIM TURBAN and JAY E. ARONSON

Similar presentations

Presentation on theme: "EFRAIM TURBAN and JAY E. ARONSON"— Presentation transcript:

Similar presentations

About project

Feedback

Log in

Auth with social network:

EFRAIM TURBAN and JAY E. ARONSON

Similar presentations

Presentation on theme: "EFRAIM TURBAN and JAY E. ARONSON"— Presentation transcript:

Similar presentations

About project

Feedback