Decision Support System

5/12/2010
Data & Information
The Business School
University of Kashmir
Definitions
 Fact – statement of some element of truth about a subject matter or a domain.
Example: milk is white, sun rises in east
 Intelligence – capacity to acquire, store, improve and apply knowledge
 Experience – what we have done and what has happened in past in a specific
area of work
 Common sense – natural ability to sense, judge or perceive situations ; grows

stronger over time
 Memory – ability to store and retrieve relevant experience at will, is part of

intelligence
 Learning – is knowledge or skill that is acquired by instruction or study
5/12/2010 10:21:02 PM Rafi A Khan 2

Definitions
 Knowledge. Information once analyzed, understood, and explained is
 knowledge or foreknowledge (predictions or forecasts).
Rafi A Khan
3
Data, Information and Systems
 Data vs. Information
– Data
• A ―given,‖ or fact; a number, a statement, or a picture
• Represents something in the real world
• The raw materials in the production of information
– Information
• Data that have meaning within a context
• Data in relationships
• Data after manipulation
Rafi A Khan
4
Data, Information and Systems
 Data Manipulation
– Example: customer survey
• Reading through data collected from a customer survey with

questions in various categories would be time-consuming and
not very helpful.
• When manipulated, the surveys may provide useful

information.
Rafi A Khan
5
Why Information Systems
Rafi A Khan
6
What Is an Information System?
Rafi A Khan
7
Systems
 Generating Information
– Computer-based ISs take data as raw material, process it,

and produce information as output.
Figure 1.1 Input-process-output
Rafi A Khan
8
Characteristics of Information
Rafi A Khan
Figure 1.2 Characteristics of useful information 9
Activities in an Information System
INPUT PROCESS OUTPUT
FEEDBACK
Rafi A Khan
10
Rafi A Khan
11
Information Needs of a Shopkeeper
Daily sales account
List of low stock items to be re-ordered
List of overstock items
Long overdue payments
Profit and loss account
Used to streamline day to day operations called Operational

information
Rafi A Khan
12
 Slow or fast moving items
 Reliable supplier of items
 Sales trends
Used to improve profitability of shop called Tactical information
Rafi A Khan
13
Whether to stock different varieties of items
Whether to diversify
Whether to start a new branch in a different locality
Whether to start an e-shop
Information to expand business and explore new opportunities

Known as Strategic Information
Rafi A Khan
14
Types of Information
Strategic : Needed for long range planning and directions.
• This is less/unstructured.
Tactical : Needed to take short range decisions to improve
• Profitability and Performance.
Operational : Needed for day to day operations of the

organization.
• Eg: Daily Sales, Billing.
Rafi A Khan
15
SYSTEM
System as a group of interrelated components working together toward a

common goal by accepting inputs and producing outputs in an organized
transformation process.
Such a system has three basic interacting components or functions:
 Input: Involves capturing and assembling elements that enter the system to be
processed. For example, raw materials, energy, data, and human effort must be
secured and organized for processing.
Rafi A Khan
16
SYSTEM
 Processing: Involves transformation process that converts input into output.

Examples to these are manufacturing process, the human breathing process,
etc.
 Output: Involves transferring elements that have been produced by

transformation process to their ultimate destination, Examples to these are
finished products, human services and management information that must be
transmitted to their human users.
Rafi A Khan
17
System
System
Input Process Output
Feedback and Control
Environment
Fig. Showing Elements of a System
Rafi A Khan
18
System
A system with feedback and control components is sometimes called a

cybernetic system, that is, a self-monitoring, self-regulating system.
 Feedback: It is data about the performance of a system. It is actually

measured in terms of the outcome to that of the predefined objectives
set out at the beginning of the process.
 Control: It involves monitoring and evaluating feedback to determine

whether a system is moving toward the achievement of its goal.
Rafi A Khan
19
System Characteristics
 A system does not exist in a Vacuum; rather, it exists and function in an

environment containing other systems
 If a system is one of the components of a larger system, it is then

referred to as a subsystem, and the larger system is its environment.
 The system that has the ability to change itself or its environment in
order to survive is an adaptive system
Rafi A Khan
20
Types of System
 A large system can be split or decomposed into smaller subsystems

up to a certain level
 The decomposition of a system into subsystems can be in a serial

form or it could be in a matrix form
 In a serial system processing, the entire output of a subsystem is the

input to the next sub-system and so on.
 In the matrix arrangement the different outputs go to different sub-

systems. A subsystem receives more than one input from other
subsystems.
Rafi A Khan
21
Types of System
 If the process of input transformation is not visible and

understandable then we say that the system is a black box and
the process is not transparent
 Most of the systems can be viewed in a hierarchical structure.

Breaking the system in a hierarchical manner provides a way to
structured systems analysis. It gives a clear understanding of the
contribution of each subsystem in terms of data flow and
decisions, and its interface to the other subsystems.
Rafi A Khan
22
Types of System
The systems can be classified in different categories based on the

predictability of its output and the degree of information exchange
with the environment.
Deterministic- when the inputs, the process and the outputs of a

system are known with certainty. In a deterministic system, you can
predict the output with certainty.
Probabilistic- when the output can only be predicted in

probabilistic terms. The accounting system is deterministic while
the demand forecasting system is a probabilistic one.
Rafi A Khan
23
Types of System
 If a system is functioning in isolation from the environment, then the
system does not have any exchange with the environment nor is it
influenced by the environmental changes. Such a system is called a
closed system.
 If the system has exchange with the environment and is influenced by

the environment then it is called an open system.
 All kinds of accounting systems, viz., cash, stocks, attendance of

employees are closed systems. Most of the systems based on rules
and principles are closed systems.
 The systems which are required to respond to changes in the

environment, such as marketing, communication and forecasting are
open systems
Rafi A Khan
24
Types of System
Specify in the inputs, processes, and outputs of the following

systems. Determine what is required for each system to be
efficient and effective.
 Post Office
 Elementary school
 Grocery store
 Farm
Rafi A Khan
25
Types of System
Organization Inputs Processes Outputs
Post Office Letters mailed Delivery of mail Mail delivered
School Students Teaching Graduating

students
Grocery Store Food sold to
Food products Stocking, selling
customers
Farm Feedstock, seeds, Animals and Food delivered to
fertilizer plants market
growing
Rafi A Khan
26
System
List possible kinds of feedback for the systems in the previous question.
 Post Office: Customers' complaints, average days for a delivery, cost,

percent of lost mail
 School: Students' complaints, achievement on national tests, success

in job placement
 Grocery store: Customer feedback on quality, quantity, percent of

theft and waste, etc.
 Farm: Quality of output sold to market
Rafi A Khan
27
Information Systems
•Information system consists of physical and non-physical components

working together
•A computer combines with a software program may constitute an

information system, but only if the program is designed to produce
information that helps an organization or person to achieve a specific
goal.
Rafi A Khan
28
Information Systems
Management Information System (MIS) Computer-based or manual system

- transforms data into information to support the decision making.
MIS can be classified as performing three functions:
(1) To generate reports - for example, financial statements, inventory status

reports, or performance reports needed for routine or non-routine
purposes.
(2) To answer what-if questions asked by management. For example,
questions such as "What would happen to deposits if the bank increases
interest rates?" can be answered by MIS.
(3) To support decision making. This type of MIS is appropriately called
Decision Support System (DSS).
-DSS attempts to integrate the decision maker, the data base, and the
quantitative models being used.
Rafi A Khan
29
Information Systems?
Rafi A Khan
30
WHY INFORMATION SYSTEMS?
Major Business Functions
 Sales and marketing
 Manufacturing
 Finance
 Accounting
 Human resources
Rafi A Khan
31
MIS in Marketing
Marketing Management Information Systems:
 It supports managerial activity in the area of product

development, distribution, pricing decisions, promotional
effectiveness, and sales forecasting.
 It mainly relies on external sources of data like competitors

and customers.
Rafi A Khan
32
MAJOR TYPES OF SYSTEMS IN ORGANIZATIONS
Sales and Marketing Systems
Major functions of systems:
 Sales management, market research, promotion, pricing, new

products
Major application systems:
 Sales order info system, market research system, pricing

system
Rafi A Khan
33
Sales and Marketing

Systems
Rafi A Khan
34
MIS in Manufacturing
Manufacturing Management Information Systems:
 Inventories are provided just in time to reduce costs of

warehousing huge inventories .
Rafi A Khan
35
Manufacturing and Production Systems

 Scheduling, purchasing, shipping, receiving, engineering,
operations

 Materials resource planning systems, purchase order
control systems, engineering systems, quality control
systems
Rafi A Khan
36
Rafi A Khan
37
Rafi A Khan
38
MIS in Finance
Financial Management Information Systems:
 It provides financial information to all financial managers

within an organization including the chief financial officer.
 The chief financial officer analyzes historical and current

financial activity, future financial needs, and monitors and
controls the use of funds over time using the MIS
Rafi A Khan
39
Financing and Accounting Systems
 Budgeting, general ledger, billing, cost accounting
 General ledger, accounts receivable, accounts payable,

budgeting, funds management systems
Rafi A Khan
40
Rafi A Khan
41
MIS in HR
Human Resources Management Information Systems:
 These systems are concerned with activities related to

workers, managers, and other individuals employed by the
organization.
 It includes, work-force analysis and planning, hiring,

training, and job assignments.
Rafi A Khan
42
Human Resource Systems
 Personnel records, benefits, compensation, labor relations,

training
 Payroll, employee records, benefit systems, career path

systems, personnel training systems
Rafi A Khan
43
Rafi A Khan
44
Rafi A Khan
45
Key Elements of An Organization
 People: Managers, knowledge workers, data workers,

production or service workers
 Structure: Organization chart, products, geography
 Operating procedures: Standard operating procedures (SOP,

rules for action)
Rafi A Khan
46
IT/Tools for Managers
 Hardware: Physical equipment
 Software: Detailed preprogrammed instructions
 Storage: Physical media for storing data and the software
 Communications Technology: transfers data from one

physical location to another
 Networks: link computers to share data or resources
Rafi A Khan
47
IS & Organizations
Rafi A Khan
48
TOWARD THE DIGITAL FIRM
Rafi A Khan
49
Rafi A Khan
50
Information Sytems
Rafi A Khan
51
Major Types of Systems
• Executive Support Systems (ESS)

• Decision Support Systems (DSS)
• Management Information Systems (MIS)
• Knowledge Work Systems (KWS)
• Office Systems
• Transaction Processing Systems (TPS)
Rafi A Khan
52
Rafi A Khan
53
Transaction Processing Systems (TPS)

Operational Level :
• Basic business systems that serve the operational level
• A computerized system that performs and records the daily

routine transactions necessary to the conduct of the business
Rafi A Khan
54
Rafi A Khan
55
Rafi A Khan
56
Knowledge Work Systems (KWS)
Knowledge level
 Inputs : Design specs
 Processing : Modeling
 Outputs : Designs, graphics
 Users : Technical staff and professionals
Example: Engineering work station
Rafi A Khan
57
Management Information System (MIS)
Management level
 Inputs : High-volume data
 Processing : Simple models
 Outputs : Summary reports
 Users : Middle managers
Example: Annual budgeting
Rafi A Khan
58
Rafi A Khan
59
Rafi A Khan
60
Decision Support System (DSS)
Management level
 Inputs : Low/High volume data
 Processing : Interactive
 Outputs : Decision analysis
 Users : Professionals, Staff
Example: Forecasting
Rafi A Khan
61
Rafi A Khan
62
Executive Support System (ESS)
Strategic level
 Inputs : Aggregate data
 Processing : Interactive
 Outputs : Projections
 Users : Senior managers
Example: 5-year operating plan
Rafi A Khan
63
Rafi A Khan
64
Rafi A Khan
65
Management Information Systems 8/e
Chapter 2 Information Systems in the Enterprise
ENTERPRISE APPLICATIONS
Business Processes and Information Systems

Business processes
 Manner in which work is organized, coordinated, and focused to produce

a valuable product or service
 Concrete work flows of material, information, and knowledge—sets of

activities
Rafi A Khan
66

 Unique ways to coordinate work,
information, and knowledge
 Ways in which management chooses
to coordinate work
Rafi A Khan
67

Information systems help organizations
 Achieve great efficiencies by automating parts of processes
 Rethink and streamline processes
Rafi A Khan
68
Examples of Business Processes

 Manufacturing and production: Assembling product, checking
quality, producing bills of materials
 Sales and marketing: Identifying customers, creating customer

awareness, selling
Rafi A Khan
69
Examples of Business Processes

 Finance and accounting: Paying creditors, creating financial
statements, managing cash accounts
 Human Resources: Hiring employees, evaluating performance,

enrolling employees in benefits plans
Rafi A Khan
70
Cross-Functional Business Processes
 Transcend boundary between sales, marketing, manufacturing, and

research and development
 Group employees from different functional specialties to a complete

piece of work
Example: Order Fulfillment Process
Rafi A Khan
71
Rafi A Khan
72
Enterprise Applications
 Enterprise systems
 Supply chain management systems
 Customer relationship management systems
 Knowledge management systems
Rafi A Khan
73
Traditional View of the Systems
 Within the business: There are functions, each having its uses of
information systems
 Outside the organization‘s boundaries: There are customers and

vendors
Functions tend to work in isolation
Rafi A Khan
74
Figure 2-13
Rafi A Khan
75
Rafi A Khan
76
Benefits of Enterprise Systems

 Firm structure and organization: One organization
 Management: Firm-wide knowledge-based management processes
 Technology: Unified platform
 Business: More efficient operations and customer-driven business

processes
Rafi A Khan
77
Challenges of Enterprise Systems

 Difficult to build: Require fundamental changes in the way the
business operates
 Technology: Require complex pieces of software and large

investments of time, money, and expertise
 Centralized organizational coordination and decision

making: Not the best way for the firms to operate
Rafi A Khan
78
Supply Chain Management (SCM)

 Close linkage and coordination of activities involved in buying,
making, and moving a product
 Integrates supplier, manufacturer, distributor, and customer logistics

time
 Reduces time, redundant effort, and inventory costs
Rafi A Khan
79

Supply Chain
 Network of organizations and business processes
 Helps in procurement of materials, transformation of raw materials

into intermediate and finished products
Rafi A Khan
80

Limitations:
 Inefficiencies can waste as much as 25% of company‘s operating costs
 Bullwhip Effect: Information about the demand for the product gets
distorted as it passes from one entity to next
Rafi A Khan
81
Rafi A Khan
82

 Helps in distribution of the finished products to customers
 Includes reverse logistics - returned items flow in the reverse direction

from the buyer back to the seller
Rafi A Khan
83
How Information Systems Facilitate Supply Chain Management

 Decide when, what to produce, store, move
 Rapidly communicate orders
 Communicate orders, track order status
 Check inventory availability, monitor levels
 Track shipments
 Plan production based on actual demand
 Rapidly communicate product design change
 Provide product specifications
 Share information about defect rates, returns
Rafi A Khan
84

 Supply chain planning system: Enables firm to generate forecasts for
a product and to develop sourcing and a manufacturing plan for the
product
 Supply chain execution system: Manages flow of products through

distribution centers and warehouses
Rafi A Khan
85
Collaborative Commerce
 Uses digital technologies to enable multiple organizations to
collaboratively design, develop, build, move, and manage products
 Increases efficiencies in reducing product design life cycles, minimizing

excess inventory, forecasting demand, and keeping partners and
customers informed
Rafi A Khan
86
Rafi A Khan
87
Industrial Networks
Private Industrial Networks
 Web-enabled networks
 Link systems of multiple firms in an industry
 Coordinate transorganizational business processes
Rafi A Khan
88
Customer Relationship Management (CRM)

 Manages all ways used by firms to deal with existing and potential new
customers
 Business and Technology discipline
 Uses information system to coordinate entire business processes of a firm
Rafi A Khan
89

 Provides end-to-end customer care
 Provides a unified view of customer across the company
 Consolidates customer data from multiple sources and provides

analytical tools for answering questions
Rafi A Khan
90
Rafi A Khan
91
Knowledge Management Systems
 Creating knowledge
 Discovering and codifying knowledge
 Sharing knowledge
 Distributing knowledge
Rafi A Khan
92
The End
Rafi A Khan
93
5/12/2010 94
Management Information Systems

Feasibility Study
Three types of feasibility :
 Technical Feasibility
 Economical Feasibility
 Operational Feasibility
Rafi A Khan
95
Technical Feasibility
 H/W - I/P, O/P, Communication, Storage
 S/W - Database, OS, Languages
 Application - System Packages, Management Science Models
Rafi A Khan
96
Economical Feasibility
 Costs - Sytems/Programmes, Operations, H/W, S/W
 Savings - Operating Expenses, Clerical Personnel,

Equipment
 Benefits - Tangible ---- Reduction in Production Cost
Intangible ---- Customer Satisfaction
Rafi A Khan
97
Operational Feasibility
 Management - Operating Management
Middle Management
Top Management
Rafi A Khan
98
Reports of MIS
 Periodic Scheduled Reports.
 Exception Reports.
 Demand Reports and Responses.
 Push Reporting.
99
Information Technology and MIS
 Information Technology is defined as that branch of computer science

that includes:
Hardware.
Software.
Communication Technology.
Storage systems and
Other Information processing technologies.
10
Computer Hardware-------
The physical equipment
Secondary storage
Communication •Magnetic disk
Devices
•Optical disk
•Magnetic tape
Central Buses
Processing Unit Primary Storage
Input Devices Output Devices

•Keyboard •Printers
•Computer mouse •VDT
•Touch screen •Plotters
•Source data automation •Audio output
10
Computer Software
Computer
Software
Application System
Software Software
General purpose
Application-Specific System Management System Development
Application
Programs Programs Programs
Programs
10
Communication Technology
 Communications technology allows systems to transfer data from one location to

another for the transmission of voice, data, images, sound and even video. It can
take the form of :-
• Wired transmission: The transmission media can be
• Twisted pair cable.
• Coaxial cable.
• Fiber-optic cable.
• Wireless transmission: This includes:-
• Microwave Transmission.
• Satellite Transmission.
10
Using Communication Technology for Business
Solutions
 The Internet is revolutionizing communications by providing a
worldwide network linking business, government, and scientific and
educational organizations to individuals. Internet use falls into several
major areas, including:
• Electronic mail/ Voice mail.
• World Wide Web.
• Chat.
• Electronic Data Interchange.
• Electronic Commerce.
• Mobile Commerce.
10
Using Communication Technology for Business Solutions
 Intranets
These help organizations in creating richer, more responsive information

environments in which members of an organization can exchange
ideas, share information and work together on common projects and
assignments regardless of their physical location.
10
Using Communication Technology for Business Solutions
 Extranets
These are privately owned networks that are extended to
authorized users outside the company e.g. authorized buyers,
retailers, distributors, customers. They are often used for
collaborating with other companies for:
1. Supply Chain Management.
2. Customer Relationship Management.
3. Product design and development.
4. Training efforts.
10
Enterprise application architecture presenting an overview of the major
cross-functional enterprise applications and their interrelationships.
Suppliers
Supply Chain Management.

Collaboration. Decision Support.
Knowledge Management.
Selling. Distribution.
Partner Relationship Management.
Sourcing. Procurement
Employees
Enterprise Resource Planning.
Partners
Internal Business Processes.
Customer Relationship
Management.
Marketing. Sales. Service
Customers
10
Supply Chain Management
 It is a cross-functional inter-enterprise system that uses information

technology to help support and manage the links between some of
company‘s key business processes and those of its suppliers, customers
and business partners.
 The goal of SCM is to create a fast, efficient and low-cost network of

business relationships, or supply chain, to get a company‘s products
from concept to market.
10
Enterprise Resource Planning
 Integrated cross-functional software that re-engineers manufacturing,

distribution finance, human resources and other basic business process
of a company to improve its efficiency, agility, and profitability.
 It focuses on the company‘s internal aspects giving them an integrated

real-time view of its core business processes.
 Simply the technological backbone of e-business.
10
Customer Relationship Management
 A cross-functional e-business application that integrates and automates

many customer serving processes in sales, direct marketing, account
and order management, and customer service and support.
 CRM systems create an IT framework of web-enabled software and

databases that integrates these processes with the rest of a company‘s
business operations.
Rafi A Khan
11
Knowledge Management
 Organizing and sharing the diverse forms of business information

created within an organization. Includes managing project and
enterprise document libraries, discussion databases, intranet website
databases, and other types of knowledge bases.
 Different phases of a knowledge management system (KMS).
• Capturing/Acquisition of data/information
• Transformation of Info. into Knowledge
• Knowledge Storage
• Disseminating/Sharing of Knowledge
Figure
Rafi A Khan
11
Capturing/Acquisition of Data/Information
 Various technologies that can help in capturing of information

are:
• Document Management System:
Document management system keeps track of masses

of data and information, which is stored in a secure file
vault where its integrity is guaranteed and all changes to it,
is monitored, controlled, and recorded providing far easy
and faster access to all the documents. It takes care of
creating, storing, editing, and distributing documents.
Rafi A Khan
11
 Database
Database is a collection of data organized to serve many

applications efficiently by centralizing the data and minimizing
redundant data. It is a computerized record keeping system that stores,
maintains, and provides access to information.
Database Management System (DBMS) is simply the software

that permits an organization to centralize data, manage them
efficiently, and provide access to the stored data by applications
programs. The DBMS acts as an interface between application
programs and the physical data files.
Rafi A Khan
11
 Data Warehouse
An integrated collection of data extracted from operational,

historical and external databases and cleaned, transformed and
cataloged for retrieval and analysis to provide business intelligence for
business decision making.
 Search Engines
These are huge databases of web page files that have been
assembled automatically by machine.
Rafi A Khan
11
Transformation of Info. into Knowledge
Useful technologies for this phase of the knowledge management process

include:
 Multidimensional Data Analysis: Another term for multidimensional

data analysis is Online Analytical Processing (OLAP), which is a function
of business intelligence software that enables a user to easily and
selectively extract and view data from different points of view.
OLAP tools structure data hierarchically – the way managers think of

their enterprises, and also allows business analysts to rotate that data,
changing the relationships to get more detailed insight into corporate
information.
Rafi A Khan
11
Transformation of Info. into Knowledge
 Data mining or Knowledge Discovery in Databases (KDD) provides an

organization with highly tangible benefits in the area of analysis. Data
mining is the nontrivial extraction of implicit, previously unknown,
and potentially useful information from data. This encompasses a
number of different technical approaches such as clustering, data
summarization, learning classification rules, finding dependency net
works, analyzing changes, and detecting anomalies.
 Data mining software tools find hidden patterns and relationships in

large pools of data and infer rules from them that can be used to
predict future behavior and guide decision-making.
Rafi A Khan
11
Transformation of Info. into Knowledge contd.
 Decision Support Systems (DSS)
These are a specific class of computerized information system that

supports business and organizational decision-making activities.
 Artificial Intelligence (AI)
Rafi A Khan
11
Info/Knowledge Storage
 Knowledge repositories are widely recognized as key components of

most knowledge management systems. Once knowledge is captured, it
must be stored in a knowledge repository. A knowledge repository is a
collection of both internal and external knowledge.
Rafi A Khan
11
Info/Knowledge Dissemination
 The final phase is effectively communicating the captured "knowledge."

In fact, knowledge is not truly captured. Instead, what is captured is
information that is more easily transformed into knowledge by the
recipient. The key technologies that can be used for dissemination are:
• E-mail
• Teleconferencing, Data-conferencing
• Videoconferencing
• Groupware, and
• Intranets.
Rafi A Khan
11
5/12/2010 120
Decision Making
2009
Rafi A Khan
12
DECISION MAKING
System is a collection of objects such as people, resources,

concepts, and procedures intended to perform a function or
to serve a goal.
• Closed systems are totally independent.

• Open systems dependent on their environment.
• System effectiveness is the degree to which goals are achieved.

• System efficiency is a measure of the use of inputs (or resources)
to achieve outputs.
Rafi A Khan
5/12/2010 12 121
Decision making is a process of choosing among alternative
courses of action for the purpose of attaining a goal or goals.
Simon’s 4 Phases of Decision Making
(1) intelligence
decision making (2) design problem solving
(3) choice decision making
problem solving (4) implementation
Rafi A Khan
5/12/2010 12 122
INTELLIGENCE PHASE
Organizational objectives
Search and scanning
Simplification/Assumption
Data collection
Reality Problem identification
Problem ownership
Problem classification
Problem statement
Problem statement
DESIGN PHASE
Validation of the model Formulate a model
Set criteria for choice
Search for alternatives
Predict and measure outcomes
Alternatives
CHOICE PHASE
Success Verification, testing of proposed solution
Solution to the model
Sensitivity analysis
Selection of best alternative (s)
Plan for implementation
Implementation
of solution Solution
Failure
Rafi A Khan
5/12/2010 12 123
1. Intelligence phase
 Scan the environment

 Analyze organizational goals (e.g. Inventory Management, Job
Selection, lack or an incorrect web presence)
 Collect data (Monitoring & analyzing)
 Identify problem
 Categorize problem
– Programmed (repetitive & routine) ---Scheduling of employees,
inventory level etc
– Non-programmed (Unstructured) --- Merger & Acquisitions
– Decomposed into smaller parts
 Assess ownership and responsibility for problem resolution
Rafi A Khan
5/12/2010 12 124
2. Design phase
• Formulate a model
• Set criteria for choice (Are we willing to take High risk

or we prefer low risk approach)
• Search for alternatives
• Predict and measure outcomes (E.g. Profit

Maximization)
Rafi A Khan
5/12/2010 12 125
3. Choice phase
•Each alternative must be evaluated
•Sensitivity analysis (determines robustness of any given alternative)
•Selection of best alternative (s)
•Plan for implementation
solution - set of values for the decision variables in a selected alternative
Rafi A Khan
5/12/2010 12 126
4. Implementation phase
•Putting a recommended solution to work
• Vague boundaries which include:

–Dealing with resistance to change
–User training
–Upper management support
•The problem is considered solved after the recommended solution to the

model is successfully implemented.
Rafi A Khan
5/12/2010 12 127
Source: Based on Sprague, R.H., Jr., “A Framework for the Development of DSS.” MIS Quarterly, Dec. 1980, Fig. 5, p. 13.
Rafi A Khan
5/12/2010 12 128
Decision Support Systems
 Intelligence Phase
– Automatic
• Data Mining
– Expert systems, CRM, neural networks
– Manual
• OLAP
• KMS
– Reporting
• Routine and ad hoc
Rafi A Khan
5/12/2010 12 129
 Design Phase
– Financial and forecasting models
– Generation of alternatives by expert system
– Relationship identification through OLAP and data mining
– Use of KMS
– Business process models from CRM, RMS, ERP, and SCM
Rafi A Khan
5/12/2010 13 130
 Choice Phase
– Identification of best alternative
– Identification of good enough alternative
– What-if analysis
– Goal-seeking analysis
– May use KMS, GSS, CRM, ERP, and SCM systems
Rafi A Khan
5/12/2010 13 131
 Implementation Phase
– Improved communications
– Collaboration
– Training
– Supported by KMS, expert systems, GSS
Rafi A Khan
5/12/2010 13 132
5/12/2010 133
TYPES OF DECISIONS
Rafi A Khan
13
TYPES OF DECISIONS
Decisions are categorized along two dimensions:-
 The nature of the decision to be made
 The scope of the decision itself
Rafi A Khan
5/12/2010 13 134
TYPES OF DECISION
On the basis of the nature of the decision:-
1)Structured decision:-It‘s the one for which a well defined decision making
procedure exists.
2)Unstructured decision:- it is the one for which all the three decision phases
are unstructured.
3)Semi structured decision:- In this type one or two phases are structured and
the others are not.
Rafi A Khan
5/12/2010 13 135
On the basis of scope of the decision itself.
1. Strategic Decision:- It is the one which effects the entire organization or a

major part of it for a long period of time
2. Tactical Decision:- It effects how a part of the organization does business

for a limited time in the future.
3. Operational Decision:- It is the one which effects a particular activity

currently taking place in an organization but either has a little impact on
the future.
Rafi A Khan
5/12/2010 13 136
Combination of various types of Decisions
 Structured /operational
 Structured / tactical
 Structured/ strategic
 Semi-structured/ operational.
 Semi-structured/ tactical
 Semi-structured / strategic
 Unstructured/ operational
 Unstructured/ tactical
 Unstructured/ strategic
Rafi A Khan
5/12/2010 13 137
 Structured/Operational: Decide how to cut a log into boards in order
to minimize wastage.
The intelligence phase is trivial; if a log arrives at mill, it must be cut .
The design phases likewise fixed; the products that the mill produces
and hence the acceptable types of cuts.
The choice phase can be optimized mathematically because the value of

each potential board is known from business consideration and the
number of boards that can be operated via each communication of cuts
is a problem.
Rafi A Khan
5/12/2010 13 138
 Structured /Tactical: Choosing the way in which to depreciate
corporate assets.
Resource allocation problems that can be solved by linear

programming methods are also in this category.
 Structured /Strategic: Deciding weather or not to proceed with an

R&D project on the bases of projected ROI
A plant location decision could be in this category if the only factors in

decisions are quantifiable, such as transportation costs of known raw
materials from known locations and of known products to known
markets.
Rafi A Khan
5/12/2010 13 139
 Semistructured/Operational: Deciding to accept or reject an applicant
to a selective collage.
 Semitructured /Tactical: Choosing an insurance company for an

employee health program. Cost per employee is an important and
objective factor in this decision. Intangible factors include acceptability
of a company to the employee population and the relative importance
of different benefits: is 100 percent hospitalization coverage with Rs.
500 deductible amount better or worse than 80 percent coverage with
no deductible?
Rafi A Khan
5/12/2010 14 140
 Semitructured /Strategic: Deciding whether or not to enter a new
market. Sales projections, marketplace growth data, development cost
estimates and marketing expenses forecasts can combine to provide a
profit-and-loss forecast. However there are countless factors that could
make it totally worthless. Judgment of experienced managers is
needed for the final step.
 Unstructured/Operational: Dealing with a machine breakdown. There

is no set procedure what to do while awaiting repairs. The decision is
operational because the way a company deals with one machine failure
need not set a precedent for the next.
Rafi A Khan
5/12/2010 14 141
 Unstructured /Tactical: Hiring decisions typically fall into this area,
especially if the job to be filled is above level where aptitude and
ability tests can be relied on as performance indicators.
 Unstructured/Strategic: Deciding how to respond to an unfriendly

takeover proposal made by a competitor. The action can have a long
term impact on the entire firm.
Rafi A Khan
5/12/2010 14 142
Decision Support Frameworks
Type of Control
Type of Decision: Operational Control Managerial Control Strategic Planning
Structured Accounts receivable, Budget analysis, short- Investments, warehouse
(Programmed) accounts payable, order term forecasting, locations, distribution

entry personnel reports centers
Semistructured Production scheduling, Credit evaluation, Mergers and

inventory control budget preparation, acquisitions, new
project scheduling, product planning,
rewards systems compensation, QA, HR
policy planning
Unstructured Buying software, Negotiations, R&D planning,
(Unprogrammed) approving loans, help recruitment, hardware technology

desk purchasing development, social
responsibility plans
Rafi A Khan
5/12/2010 14 143
The components of the quantitative model
– result variable indicate how well the system performs
– decision variables describe the alternative course of action
– uncontrollable variables or parameters are not under the
control of the decision maker
Uncontrollable
variables
Mathematical
Decision variables Result variables
relationships
– intermediate result variables reflect intermediate outcomes
Rafi A Khan
5/12/2010 14 144
Examples of the Components of Models.
Decision Result Uncontrollable
Area Variables Variables Variables and Parameters
Financial investment Investment alternatives and Total profit Inflation rate
amounts Rate of return (ROI) Prime rate
How long to invest Earnings per share Competition
When to invest Liquidity level
Marketing Advertising budget Market share Customers' income
Where to advertise Customer satisfaction Competitors' actions
Manufacturing What and how much to Total cost Machine capacity
produce Quality level Technology
Inventory levels Employee satisfaction Materials prices
Compensation programs
Accounting Use of computers Data processing cost Computer technology
Audit schedule Error rate Tax rates
Legal requirements
Transportation Shipments schedule Total transport cost Delivery distance

Regulations
Services Staffing levels Customer satisfaction Demand for services
Rafi A Khan
5/12/2010 14 145
Example
Company makes special purpose computers.

Decision to be made: how many computers should be produced next month?
Two types of computers are considered: T1, T2.
They require different days of labour, different costs for material.
Uncontrollable
variables
constraints on labour
and budget
Decision variables Mathematical

Result variables
relationships
X1 = NofT1 Maximise profit
X2 = NofT2 Total profit
subject to constraints
Rafi A Khan
5/12/2010 14 146
Principle of choice is a decision regarding the acceptability of a
solution approach.
• Normative models
– chosen alternative is the best of all possible alternatives
– suboptimisation
– optimisation models
• Descriptive models describe things as they are, or as they are

believed to be.
– no guarantee a solution is optimal
– simulation
Generating alternatives
– automatically by the model
– by using heuristics
Rafi A Khan
5/12/2010 14 147
Predicting the outcomes of alternatives
1. Decision making under certainty
Decision maker knows exactly what the outcome of
each course of action will be - deterministic environment.
2. Decision making under risk
Each alternative has several possible outcomes,
each with a given probability of occurrence
- probabilistic or stochastic decision situation.
3. Decision making under uncertainty
Several outcomes are possible for each course of action,
their probabilities are not known.
Rafi A Khan
5/12/2010 14 148
Measuring outcomes
The value of the an alternative is judged in terms of
goal attainment.
Scenario describes the decision and uncontrollable variables

and parameters for a specific modelling situation.
Of special interest are:
– the worst possible scenario
– the best possible scenario
– the most likely scenario
Rafi A Khan
5/12/2010 14 149
Search
• Analytical techniques
– mathematical formulae
– algorithm: step-by-step search process
• Blind search
– complete enumeration
– incomplete search
• Heuristic search (derived from the Greek word for discovery)
rules guide the search process
Normative models:
– analytical techniques
– complete, exhaustive enumeration
Descriptive models:
– blind search
– using heuristics
Rafi A Khan
5/12/2010 15 150
Evaluation
• Multiple goals
– Today's management systems want to achieve
multiple goals simultaneously.
– Goals are usually partially or totally conflicting.
• Sensitivity analysis
Checks the impact of a change in the input data or parameters

on the proposed solution (the result variable)
1. Automatic sensitivity analysis
tells the range within which an input variable or parameter
can vary without impact on the proposed solution
one change at a time
2. Trial and error
some input data are changed
Rafi A Khan
5/12/2010 15 151
• What-if-analysis
What will happen to the solution if an input variable or
a parameter is changed?
e.g. what will happen to the total inventory cost if the cost of
carrying inventories increases by 10%?
• Goal seeking analysis
Computes the amount of inputs necessary to achieve a desired level
of an input (goal).
e.g. How many nurses are needed to reduce the average waiting time
of a patient in the emergency room to less than 10 minutes.
Rafi A Khan
5/12/2010 15 152
Rafi A Khan
5/12/2010 15 153
Literature:
1. (a) Decision Support Systems and Intelligent Systems, Fifth Edition
E.Turban, Jay Aronson,
Prentice Hall, 1998.
(b) Decision Support Systems and Expert Systems,
Management Support Systems, E.Turban, Fourth Edition,
Prentice Hall, 1995.
2. Knowledge-based Decision Support Systems, With Applications

in Business, 2nd Edition, M. Klein, L. Methlie,
Wiley, 1995.
Rafi A Khan
5/12/2010 15 154
SUMMARY
 Systems are composed of inputs, outputs, processes, and

decision makers.
 A model is simplified representation or abstraction of reality.

They can be iconic, analog, or mathematical.
 Decision making involves four major phases: intelligence, design,

choice, and implementation.
Rafi A Khan
5/12/2010 15 155
Models
A model is a simplified representation or abstraction of reality.
1. Iconic model is a physical replica of a system.
2. Analog model gives a symbolic representation of reality,

behaves like the real system but does not look like it.
3. Mathematical (quantitative) models use mathematical relationships

Benefits:
– compression of time
– easy model manipulation
– low cost of the analysis
– cost of making mistakes is less than mistakes on real system
– can model risk and uncertainty
– a very large number of solutions can be analysed
– enhance learning and training
Rafi A Khan
5/12/2010 15 156
3. Optimisation
model generates an optimal solution
Limitations:
– works if the problem is structured and deterministic
4. Heuristics
Informal knowledge of how to solve problems efficiently and
effectively, how to plan steps in solving a complex problem,
how to improve performance, and so forth.
Rafi A Khan
5/12/2010 15 157
Modelling Process
Example: How much to order for the grocery?

The Question: How much bread to stock each day?
1. Trial-and-error
experimentation on the real system
Not appropriate if:
– too many alternatives to explore
– the cost of making errors is very high
– the environment keeps changing
2. Simulation
assume the appearance of the characteristics of reality
Problems:
– no guarantee that the solution is optimal one
– professional development
Rafi A Khan
5/12/2010 15 158
Definition of DSS
 DSS is an interactive computer-based systems, which help decision makers

utilize data and models to solve unstructured problems.
 DSS is an interactive computer-based systems, which help decision makers

utilize data and models to solve unstructured problems.
Rafi A Khan
5/12/2010 15 159
Definition of DSS
 Decision Support Systems (DSS) are a class of computerized information

systems that support decision-making activities. DSS are interactive
computer-based systems and subsystems intended to help decision makers
use communications technologies, data, documents, knowledge and/or
models to successfully complete decision process tasks.
Rafi A Khan
5/12/2010 16 160
Components of DSS
Other computer Internet, intranet,

based systems extranet.
Data management Model management External models
Knowledge-based
subsystems
User interface
Manager (user)
Organizational KB
Rafi A Khan
5/12/2010 16 161
Components of DSS
Data management subsystem
 The data management subsystem includes a data base, which contains
relevant data for the situation and is managed by software call the database
management system (DBMS) .the data management subsystem can be
interconnected with the corporate data warehouse, a repository for
corporate relevant decision making data.
Model management subsystem
 This is software package that includes financial, statistical, management
science, or other quantitative models that provide the system analytical
capabilities and appropriate software management. Modeling languages in
building custom models are also included, this software is often called a
model base management system (MBMS). This component can be
connected to corporate or external storage of models.
Rafi A Khan
5/12/2010 16 162
Components of DSS
Knowledge based management subsystem
 This subsystem can support any of the other subsystems or act as an

independent component. It provides intelligence to augment the decision
maker‘s own. It can be interconnected with the organization‘s knowledge
depository, which is called the organizational knowledge base.
User interface subsystem
 The user communicates with and commands the DSS through this
subsystem. The user is considered part of the system. Researchers assert that
some of unique contributions of DSS are derived from the intensive
interaction between the computer and the decision maker.
Rafi A Khan
5/12/2010 16 163
Internal data
THE DATA MANAGEMENT SUBSYSTEM s ources
External
data Finance Marketing Production Personal Other
Source
Extraction
Private personal
Organizational data
knowledge base
Decision support
database
Corporate data
warehouse
Query
Facility
Database management
System Interface
oRetrieval management
oInquiry
oUpdate
oReport generation
oDelete
Data directory Model management
Knowledge-based
Rafi A Khan s ubsystem
5/12/2010 16 164
THE DATA MANAGEMENT SUBSYSTEM
 The data management subsystem is composed of the following elements:
 DSS database
 Database management system
 Data directory.
 Query facility.
Rafi A Khan
5/12/2010 16 165
THE DATABASE
 A database is a collection of interrelated data organized to meet the needs
and struc-ture of an organization and can be used by more than one person
for more than one ap-plication
 The data in the DSS database are extracted from internal and external data
sources, as well as from personal data belonging to one or more Users.
Rafi A Khan
5/12/2010 16 166
DATA ORGANISATION
 In small ad hoc DSS, data can be entered directly into models some times
extracted directly from larger databases.
 In large organizations that use extensive data ,such as Wal-Mart, AT&T,

and United Air Lines data are organized in a data warehouse and used
when needed .
Rafi A Khan
5/12/2010 16 167
EXTRACTION
 To create a DSS database or a data warehouse it is often necessary to capture
data from several sources. This operation is called extraction.
 It basically consists of importing of files, summarization, standardization,

filtration, and condensation of data.
 The data for the warehouse are extracted from internal and external sources.
The extraction process is frequently managed by a DBMS.
Rafi A Khan
5/12/2010 16 168
DATABASE MANAGEMENT SYSTEM
 A database is created, accessed, and updated by a DBMS.
 Most DSS are built with a standard commercial relational DBMS that
provides capabilities such as it captures or extracts data for inclusion in a
DSS database ,it updates (adds, deletes, edits, changes) data records and
files, retrieves data ,provides data security etc.
Rafi A Khan
5/12/2010 16 169
THE QUERY FACILITY
 Query facility is necessary to access, manipulate, and query data.
 The query facility includes a special query language.
 Important functions of DSS query system are selection and manipulation

operation (e.g., the ability to follow a computer instruction such as "Search
for a sales in zone B during June 2000 and summarize sales by salesperson").
Rafi A Khan
5/12/2010 17 170
THE DIRECTORY
 The data directory is a catalog of all the data in the database.
 It contains data definitions and its main function is to answer questions

about the availability of data items, their source, and their exact meaning.
 The directory is especially appropriate for supporting the intelligence phase

of the decision-making process by helping to scan data and identify problem
areas or opportunities.
 It supports the addition of new entries, deletion of entries, and retrieval of

information on specific objects.
Rafi A Khan
5/12/2010 17 171
General Functions of the DBMS
Data Definition
 Provides a data definition language (DDL) that allows users to describe

the data entities and their associated attributes and relationships
 Allows for the interrelation of data from multiple sources
Data Manipulation
 Provides the user with a query language to interact with the database
 Allows for capture and extraction of data
 Provides rapid retrieval of data for ad hoc queries and reports
 Allows for the construction of complex queries for retrieval and data
manipulation Rafi A Khan
5/12/2010 17 172
Data Integrity
 Allows the user to describe rules (integrity constraints) to maintain the

integrity of the database
 Assists in the control of erroneous data entry based on the defined integrity
constraints
Access Control
 Allows identification of authorized users
 Controls access to data various elements and data manipulation activities

within the database
 Tracks usage and access to data by authorized users
Concurrency Control
 Provides procedures for controlling simultaneous access to the same data

by more than one user
Rafi A Khan
5/12/2010 17 173
5/12/2010 174
Improved data sharing.
The DBMS helps create an environment in which end users have

better access to more and better-managed data. Such access takes it
possible for end users to respond quickly to changes in their
environment.
Transaction Recovery
Provides a mechanism for restart and reconciliation of the database in
the event of hardware failure
Records information on all transactions at certain points to enable
satisfactory database restart
Minimized data inconsistency.
Data inconsistency exists when different versions of the same data

appear in different places.
Rafi A Khan
17
Improved decision making.
 Better-managed data and improved data access make it possible to

generate better quality information, on which better decisions are based.
Increased end-user productivity.
 The availability of data, combined with the tools that transform data
into usable information, empowers end users to make quick, informed
decisions that can make the difference between success and failure in the
global economy.
Rafi A Khan
5/12/2010 17 175
Structure of Model Management System
Models (model base)
•Strategic, tactical, operational
•Statistical, financial, marketing,
Model directory
mgt. science, accounting etc
•Model building blocks
Model execution,
integration, and command
Model base management processor
•Modeling commands : creation
•Maintenance: update
•Database interface
•Modeling language
Data management Interface management Knowledge –

based
subsystem
Rafi A Khan
5/12/2010 17 176
Functions of the MBMS
 Creates models easily and quickly, either from scratch or from the building
blocks
 Allows users to manipulate models so that they can conduct experiments

and sensitivity analyses ranging from what-if to goal seeking
 Stores, retrieves and manages a wide variety of different types of models

in a logical and integrated manner
 Accesses and integrates the model building blocks
 Catalogs and displays the directory of models for use by several

individuals in the organization
Rafi A Khan
5/12/2010 17 177
Functions of the MBMS
 Tracks model data and application use
 Interrelates model with appropriate linkages with the database and

integrates them within the DSS
 Manages and maintains the model base with management functions

analogous to database management: store, access, run, update, link, catalog,
and query
 Use multiple models to support problem solving
Rafi A Khan
5/12/2010 17 178
USER INTERFACE MANAGEMENT SYSTEM
Data management Knowledge- based Model management
and DBMS subsystem and MBMS
User Interface
Management System
(UIMS)
Language Processor
Input Output
Action Display
Languages Languages
Printers, plotters
Users
Rafi A Khan
5/12/2010 17 179
General Functions of the DSS Interface
 Allows for interaction with the DSS in a variety of dialog styles
 Accommodates the user with a variety of input devices
 Presents data with a variety of formats and output devices
 Gives user help capabilities, prompting, diagnostic and suggestion routines,

or any other flexible support.
 Stores input and output data.
 Provides support for communication among and between multiple DSS

users
Rafi A Khan
5/12/2010 18 180
General Functions of the DSS Interface
 variety of formats included menu driven, question/answer, procedural
command language, or natural command language
 Provides for the presentation of data in a variety of formats
 Allows for detailed report definition and generation by the DSS user
 Allows for the creation of forms, tables, and graphics for data output
 Can provide multiple ―windows‖ or views of the data to be available

simultaneously
Rafi A Khan
5/12/2010 18 181
CHARACTERISTICS OF DSS
 DSS provides support for decision makers mainly in semi-structured and
unstructured situations by bringing together human judgment and
computerized information.
 Support is provided for various managerial levels, ranging from top

executives to line managers.
 Support is provided to individuals as well as to groups.
 DSS provides support to several interdependent or sequential decisions.

The decisions may be made once, several times or repeatedly.
 DSS supports all phases of decision making process; intelligence, design,

choice and implementation.
Rafi A Khan
5/12/2010 18 182
 DSS are adaptive over time. DSS are flexible and so users can add, delete,
combine, change or rearrange basic elements.
 User Interface – Interactive and friendly.
 DSS attempt to prove the effectiveness of decision making rather than its
efficiency.
 The decision maker has complete control over all steps of the decision
making process in solving a problem. A DSS specifically aims to support
and not to replace the decision maker.
Rafi A Khan
5/12/2010 18 183
 End users should be able to construct and modify simple systems by
themselves. Larger systems can be built with assistance from information
system (IS) specialists.
 A DSS usually utilizes models for analyzing decision making situations. The
modeling capability enables experimenting with different strategies under
different configurations.
Rafi A Khan
5/12/2010 18 184
Benefits of DSS Use
 Extend the decision-maker‘s ability to process information and knowledge
 Extend the decision-maker‘s ability to tackle large-scale, time-consuming,

complex problems
 Improve the time associated with making a particular decision
 Improve the reliability of a particular decision process or outcome

 Encourage exploration and discovery on the part of the decision-maker
 Reveal new approaches to thinking about a particular problem space or

decision context
 Generate new evidence in support of a particular decision or confirmation
of existing assumptions
 Create a strategic or competitive advantage over competing organizations
Rafi A Khan
5/12/2010 18 185
Limitations of DSS Use
 DSSs cannot yet be designed to contain distinctly human decision-

making talents such as creativity, imaginativeness, or intuition
 The power of a DSS is limited by the computer system upon which it is

running, its design, and the knowledge it possesses at the time of its
use
 Language and command interfaces are not yet sophisticated enough to

allow for natural language processing of user directives and inquiries
 DSSs are normally designed to be narrow in scope of application thus

limiting their generalizability to multiple decision-making contexts
Rafi A Khan
5/12/2010 18 186
DSS Classification
1. Alter’s Output Classification (1980)
2. 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
Rafi A Khan
18 187
Alters' Classification of DSS
Rafi A Khan
5/12/2010 18 188
Alter’s Classification of DSS
Data-Driven DSS
 Data-Driven DSS take the massive amounts of data available through the
company's TPS and MIS systems and cull from it useful information which
executives can use to make more informed decisions.
 Data- Driven DSS emphasize access to and manipulation of large databases

of structured data
Rafi A Khan
5/12/2010 18 189
Alter’s Classification of DSS
Model-Driven DSS
 A second category, Model-Driven DSS (accounting and financial models,

representational models, and optimization models).
 Model-Driven DSS emphasize access to and manipulation of a model.
 Model-Driven DSS use data and parameters provided by decision-makers to

aid them in analyzing a situation, but they are not usually data intensive.
 Very large databases are usually not needed for Model-Driven DSS.
Primarily used for the typical "what-if" analysis. That is, "What if we
increase production of our products and decrease the shipment time?"
Rafi A Khan
5/12/2010 19 190
DSS Classifications
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
Rafi A Khan
5/12/2010 19 191
Holsapple and Winston Classification
TEXT ORIENTED DSS
 Textually represented information that could have a bearing on decision.
 Documents to be electronically created, revised and viewed as needed.
 Information Technologies such as documents imaging, hypertext etc can be

incorporated into this type.
 DMS, KMS, Content Mgt System, Business rule system
DATABASE ORIENTED DSS
 In this type of DSS the database plays a major role in the DSS structure.
 Strong report generation and query capabilities.
 Data are organized in a highly structured format.
Rafi A Khan
5/12/2010 19 192
SPREADSHEET ORIENTED DSS
 Spreadsheet is a modeling language that allows the user to write models to
execute DSS analysis.
 Tools- Statistical packages, linear programming package (Solver), financial
and management science models.
 The most popular tools used are Excel and Lotus 1-2-3.
SOLVER ORIENTED DSS

 A solver is an algorithmic or procedure written as a computer program for
performing certain computations for solving a particular problem type.
 EOQ for calculating optimal ordering quantity or a linear regression routine
for calculating trend.
 Excel, Lotus 1-2-3 and quatro pro can be used to develop such a system.
 C++, Lingo etc

Rafi A Khan
5/12/2010 19 193
RULE ORIENTED DSS
 The knowledge component of DSS includes both procedural and inferential

(Reasoning) rules, often in an expert system, format.
 Assignment Algorithm for Flight Scheduling
COMPOUND DSS
 It is a hybrid system that includes two or more of the fine basic structures
explained above. It can be built by using a set of independent DSS, each
specializing in one area.
Rafi A Khan
5/12/2010 19 194
Other DSS Classification
 Personal
 Group
 Organizational
 Custom VS Readymade
Rafi A Khan
5/12/2010 19 195
DSS Classification
OTHER CLASSIFICATIONS OF DSS
INSTITUTIONAL DSS
 Deal with decisions of a recurring nature. An institutionalized DSS can be

developed and refined as it evolves over a number of years because the DSS
is used repeatedly to solve identical or similar problems.
 Portfolio Management
ADHOC DSS
 Deals with specific problems that are usually neither anticipated nor
recurring. Adhoc decisions often involve strategic planning issues
sometimes management control problems.
Rafi A Khan
5/12/2010 19 196
Knowledge-Driven DSS
Knowledge-Driven DSS
 It suggest or recommend actions to managers.
 These DSS are computer systems with specialized problem-solving

expertise.
 The "expertise" consists of knowledge about a particular domain,

understanding of problems within that domain, and "skill" at solving some
of these problems.
 A related concept is Data Mining. It refers to a class of analytical

applications that search for hidden patterns in a database.
 Data mining is the process of searching through large amounts of data to

produce data content relationships.
Rafi A Khan
5/12/2010 19 197
Document-Driven DSS
 A new type of DSS, a Document-Driven DSS is evolving to help managers

retrieve and manage unstructured documents and Web pages.
 The Web provides access to large document databases including databases

of hypertext documents, images, sounds and video.
 Examples of documents that would be accessed by a Document-Based DSS

are policies and procedures, product specifications, catalogs, and corporate
historical documents, including minutes of meetings, corporate records, and
important correspondence.
 A search engine is a powerful decision aiding tool associated with a

Document-Driven DSS.
Rafi A Khan
5/12/2010 19 198
Communications-Driven and Group DSS
 Group Decision Support Systems (GDSS) came first, but now a broader
category of Communications-Driven DSS or groupware can be identified.
 It includes communication, collaboration and decision support technologies

that do not fit within those DSS types identified.
 A Group DSS is a hybrid Decision Support System that emphasizes both the
use of communications and decision models.
 A Group Decision Support System is an interactive computer-based system

intended to facilitate the solution of problems by decision-makers working
together as a group.
 Groupware supports electronic communication, scheduling, document

sharing, two-way interactive video, White Boards, Bulletin Boards, and
Email.
Rafi A Khan
5/12/2010 19 199
1. Impact on Structured Tasks, where standard procedures, decision
rules and information flows can be reliably Predefined.
MIS
2. Payoff – Improvement in efficient by reducing costs, turnaround

time , replacing clerical personnel or increasing their productivity.
1. Impact mostly on Structured problems (rather than tasks), in which
Mgt. Sci / OR
the objective, data and constraints can be prespecified.
2. Payoff – generation of better solutions for general categories of

problems (e.g. inventory).
1. Impact is on decisions in which there is sufficient structure for
computer and analytic aids to be of value but where the managers
DSS
judgment is essential.
2. Payoff – extending the range and capability of managers decision

process to help them improve their effectiveness.
Rafi A Khan
20
3. Relevance for managers decision making – indirect (e.g. by
providing reports and access to data.
MIS
4. MIS application is routine and done periodically.
3. Relevance for managers – provision of detailed recommendation

Mgt. Sci / OR
and new methods handling complex problems.
4. Application are nonroutine, as needed.
3. Relevance for managers – creation of supportive tool, under their

own control..
DSS
4. Application are nonroutine, as needed.
Rafi A Khan
20
5/12/2010
The Business School

Rafi A Khan
20
 Ancient
 Collaboration at the organizational level
 Could revolutionize collaboration and computing
5/12/2010 10:21:04 PM Rafi

RafiA Khan
A Khan
20
Helps organizations
 Identify
 Select
 Organize
 Disseminate
 Transfer
Important information and expertise within the organizational

memory in an unstructured manner
5/12/2010 10:21:04 PM Rafi

RafiA Khan
A Khan
20
Knowledge
 Understanding gained through experience or study
 Know-how or familiarity with how to do something
 Information that is contextual, relevant, and actionable
 Accumulation of facts, procedural rules or heuristics
 Knowledge is INFORMATION IN ACTION
 Actionable (relevant) information available in the right format, at

the right time, and at the right place for decision making
(TIWANA2000)
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
20
Knowledge
 Fact – statement of some element of truth about a subject matter or a domain.
Example: milk is white, sun rises in east.
 Heuristics – rule of thumb based on years of experience.
Example: strike on independence day in our state
 Intelligence – capacity to acquire, improve and apply knowledge.
 Experience – what we have done and what has happened in past in a specific
area of work
 Common sense – natural ability to sense, judge or perceive situations ; grows

stronger over time.
 Memory – ability to store and retrieve relevant experience at will, is part of

intelligence.
 Learning
5/12/2010 – is knowledge
10:21:05 PM or skill that
Rafi A is
RafiA acquired by instruction or20
Khan
Khan study
Knowledge Types
 Explicit knowledge
– Objective, rational, technical
– Policies, goals, strategies, papers, reports
– Codified
– Leaky knowledge
 Tacit knowledge
– Subjective, experiential learning
– Highly personalized
Rafi
– Difficult to formalize
5/12/2010 10:21:05 PM RafiA Khan
A Khan
20
Knowledge Types
 Shallow (surface) knowledge
– Indicates minimal understanding of the problem area
Example – If u don‘t have petrol in your car, the car

wont start
 Deep knowledge
– Indicates maximal understanding of the problem area
Example – why don‘t a car starts when it has no petrol

(need to know various
5/12/2010 10:21:05 PM components
Rafi
RafiA Khan
A Khan of car) 20
Knowledge Types
– Descriptive – data, information
– Procedural – how to do something
– Reasoning – policies or rules
– Linguistic – vocabulary or grammar
– Presentation – graphing, messaging
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
20
Data, Information and Knowledge
INFORMATION
Processed Relevant and
actionable
DATA KNOWLEDGE
Relevant and actionable data
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
21
Knowledge Management (KM)
 A process of capturing, transformation, and diffusion of

knowledge throughout an enterprise so that it can be
shared and thus REUSED
 Helps organizations find, select, organize, disseminate,

and transfer important information and expertise
 Transforms data / information into actionable knowledge

to be used effectively anywhere
5/12/2010 10:21:05 PM Rafi
RafiA Khan in the organization
A Khan
21 by
KM Objectives
 Create knowledge repositories
 Improve knowledge access
 Enhance the knowledge environment
 Manage knowledge as an asset
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
21
KMS Manage
 Knowledge creation through learning
 Knowledge capture
 Knowledge sharing and communication through

collaboration
 Knowledge access
 Knowledge use and reuse
 Knowledge archiving
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
21
Cyclic Model of KM
 Create knowledge
 Capture knowledge
 Refine knowledge
 Store knowledge
 Manage knowledge
 Disseminate knowledge
5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
21
Cyclic Model of KM
Create Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
21
Cyclic Model of KM
Create Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
21
Cyclic Model of KM
Capture
Knowledge
Create Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
21
Cyclic Model of KM
Capture
Knowledge
Create Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
21
Cyclic Model of KM
Capture
Knowledge
Create Knowledge Refine

Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
21
Cyclic Model of KM
Capture
Knowledge

Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
22
Cyclic Model of KM
Capture
Knowledge

Knowledge
Store
Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
22
Cyclic Model of KM
Capture
Knowledge

Knowledge
Store
Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
22
Cyclic Model of KM
Capture
Knowledge

Knowledge
Store
Knowledge
Manage
Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
22
Cyclic Model of KM
Capture
Knowledge

Knowledge
Store
Knowledge
Manage
Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
22
Cyclic Model of KM
Capture
Knowledge

Knowledge
Disseminate
Store
Knowledge
Knowledge
Manage
Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
22
Cyclic Model of KM
Capture
Knowledge

Knowledge
Disseminate
Store
Knowledge
Knowledge
Manage
Knowledge
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
22
Why Adopt KM
 Cost savings
 Better performance
 Demonstrated success
 Share Best Practices
 Competitive Advantage
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
22
KM Methods, Technologies, and Tools
 Email or messaging
 Document management
 Search engines
 Enterprise information portal
 Data warehouse
 Groupware
 Workflow management
 Web-based training
 Others
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
22
5/12/2010
Knowledge Acquisition
Techniques
The Business School
Rafi A Khan
22
Knowledge Acquisition
The following are main methods of knowledge acquisition :
• Production Rule
• Frames
• Semantic Network
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
23
Production Rules
IF-THEN
 Independent part, combined with other pieces, to produce

better result
 Model of human behavior
 Examples
– IF condition, THEN conclusion
– Conclusion, IF condition
– If condition, THEN conclusion1 (OR) ELSE conclusion2

5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
23
Frames
 Organized structure of knowledge
 Put related knowledge in one area called frame
 A frame consists of slots representing a part of

knowledge
 Each slot has a value in the form of data,
information, process and rules
 Frame can be related to other frames
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
23
Frames(Engine Overheating)
Slot : Symptoms Value
Temp More than 80 deg

Water Boiling Slot : Inspection
Speed Retardation Value
Check Water Level
Oil in Engine Slot : Treatment Value
Carburetor
Stop Engine & Drain
Water
Start Engine & pour cold
Water
Increase oil level
Adjust Carburetor
5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
23
Semantic Networks
 Graphical
depictions
 Nodes and links

connecting nodes
 Node represents an
Entity & link
represents
Association
 Hierarchical
relationships
5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
23
Inferencing
 Inferencing means deriving a conclusion based on statements

that only imply that conclusion.
 Every rule in knowledge base can be checked to see whether

its premise (principle) or conclusion can be satisfied by
previously made assertions.
 This process can be done in two directions :
–Forward
–Backward
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
23
Inference Techniques
Forward Chaining
Forward chaining is a data-driven approach . We start from available
information as it becomes available or from a basic idea, and then we try to
draw conclusions.
Backward chaining
Backward chaining is a goal-driven approach in which you start from an
expectation of what is going to happen (hypothesis) and then seek evidence
that supports (or contradicts) your expectation.
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
23
Example
Investment Decision : Whether to invest in IBM Stocks

The following variables are used:
– A= Have Rs.10,000
– B= Younger than 30
– C= Education at college level
– D= Annual income of atleast Rs.40,000
– E= Invest in securities
– F= Invest in growth stocks
– G= Invest in IBM stock (the potential goal)
The facts: we assume that an investor has Rs.10,000(that A is true) and

that she is 25 years old (B is true). She would like advice on
investing in IBM stock(yes or no for the goal).
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
23
Example
The Rules:Our knowledge base includes the following five rules:

 R1: IF a person has Rs10,000 to invest and has a college degree THEN she should
invest in securities
 R2: IF a persons annual income is atleast Rs40,000 to invest and has a college
degree THEN she should invest in growth stocks
 R3: IF a person is < 30 and is investing in securities THEN she should invest in
growth stocks
 R4: IF a person is < 30 and >22 THEN she has a college degree
 R5:IF a person wants to invest in growth stocks then the stock should be IBM
– R1: IF A and C, THEN E
– R2: IF D and C,THEN F
– R3: IF B and E, THEN F
– R4: IF B, THEN C
– R5: IF F, THEN G
5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
23
Forward Chaining
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
23
Backward Chaining
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
24
The End
5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
24
Experts
 Experts
– Have special knowledge, judgment, and experience
– Can apply these to solve problems
• Higher performance level than average person
• Faster Solutions
• Recognize Patterns
 Expertise
– Task specific knowledge of experts

5/12/2010 10:21:05 PM Rafi
RafiA
• Acquired from reading, Khan
training,
A Khan
practice 24
Expert System
Expert Systems: a computer application that employs a set of rules based

on human knowledge to solve problems that require human expertise
 Information systems that solve problems by capturing knowledge for a
very specific and limited domain of human expertise are called expert
systems
– For example, diagnosing a cars ignition system, classifying biological

specimen
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
24
Common Expert System Architecture
User
User
Interface
Inference
Engine
Knowledge Base
User Environment
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
24
KBES
Knowledge based expert system (KBES) has three basic

components:
• Knowledge base
• User control mechanism
• Inference Mechanism
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
24
User Interface
 Design of the UI focuses on human concerns such as ease of use,

reliability and reduction of fatigue
 Design should allow for a variety of methods of interaction
(input, control and query)
 Mechanisms include touch screen, keypad, light pens, voice
command, hot keys
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
24
Knowledge Base
 Contains the domain-specific knowledge acquired from the
domain experts
 Can consist of all the theoretical foundations, facts, judgments,
rules, formulas, intuitions and experience
 The success of an ES relies on the completeness and accuracy of
its knowledge base
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
24
Inference Engine
 Here, the knowledge is put to use to produce solutions
 Interprets the knowledge available and performs logical
deductions in a given situation.
 It is a strategy used to search through rule base
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
24
Applications of Expert Systems
 DENDRAL project
– Applied knowledge or rule-based reasoning commands
– Deduced likely molecular structure of compounds
 MYCIN
– Rule-based system for diagnosing bacterial infections
 XCON
– Rule-based system to determine optimal systems
configuration
 Credit analysis
– Ruled-based systems for commercial lenders
 Pension fund adviser
5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
24
Applications
 Finance
– Insurance evaluation, credit analysis, tax planning, financial planning
and reporting, performance evaluation
 Data processing
– Systems planning, equipment maintenance, vendor evaluation, network
management
 Marketing
– Customer-relationship management, market analysis, product planning
 Human resources
– HR planning, performance evaluation, scheduling, pension management,
legal advising
 Manufacturing
– Production planning, quality management, product design, plant site
selection, equipment maintenance and repair
5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
25
Benefits of Expert Systems
 Increased outputs
 Increased productivity
 Decreased decision-making time
 Increased process and product quality
 Reduced downtime
 Capture of scarce expertise
 Flexibility
 Ease of complex equipment operation
 Elimination of expensive monitoring equipment
 Operation in hazardous environments
 Access to knowledge and help desks
5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
25
Benefits of Expert Systems
 Ability to work with incomplete, imprecise, uncertain data
 Provides training
 Enhanced problem solving and decision-making
 Rapid feedback
 Facilitate communications
 Reliable decision quality
 Ability to solve complex problems
 Ease of knowledge transfer to remote locations

5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
25
Limitations
 Knowledge not always readily available
 Difficult to extract expertise from humans
 Lack of end-user trust
 Knowledge subject to biases
 Systems may not be able to arrive at conclusions
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
25
The End
5/12/2010 10:21:05 PM Rafi
RafiA Khan
A Khan
25
5/12/2010
The Data Warehouse
Department of Management Studies
University of
Rafi A Khan Kashmir
25
Objective
 How operational data and decision support data differ
 What a data warehouse is, how data for it are prepared, and how it is
implemented
 What data mining is and what role it plays in decision support
 What online analytical processing (OLAP) is
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
25
The Need for Data Analysis
 Managers must be able to track daily transactions to evaluate how the

business is performing
 By tapping into operational database, management can develop strategies

to meet organizational goals
 Data analysis can provide information about short-term tactical

evaluations and strategies
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
25
Operational Data vs. Decision Support Data
 Operational Data
– Mostly stored in relational database
– Optimized to support transactions representing daily operations
 DSS Data
– Give tactical and strategic business meaning to operational data
– Differs from operational data in following three main areas:
• Time span
• Granularity
• Dimensionality
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
25
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
25
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
26
The Data Warehouse
 Integrated, subject-oriented, time-variant, nonvolatile collection of data

that provides support for decision making
 Usually a read-only database optimized for data analysis and query

processing
 Requires time, money, and considerable managerial effort to create
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
26
Characteristics of a Data Warehouse
• Subject orientation: data is organized based on how the users refer to it.
• Integrated: all inconsistencies regarding naming convention and value
representations are removed.
• Non-volatile: data is stored in read-only format and do not change over
time.
• Time Variant: data are not current but normally time-series.
• Summarized: operational data are mapped into a decision-usable
format.
• Large Volume: time-series datasets are normally quite large.
• Not Normalized: DW data can, and often is, redundant.
• Metadata: data about data is stored.
• Data Sources: internal and external unintegrated operational systems.
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
26
The Data Warehouse (continued)
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
26
The Data Warehouse (continued)
5/12/2010 10:21:05 PM Rafi

RafiA Khan
A Khan
26
Dependent
Data Mart
Extract/Summarize Data
External
Data
OLAP tools
ETL Routine
Operational (Extract/Transform/Load) Data
Database(s) Warehouse Data Mining Tools
Ad-hoc Queries
Independent
Data Mart
Reporting Tools
Monitoring/
Fig. Data warehouse process model
5/12/2010 10:21:06 PM Rafi
RafiA Khan
A Khan
26
Data Warehousing Benefits
 Increase in knowledge worker productivity
 Supports all decision makers‘ data requirements
 Provide ready access to critical data
 Insulates operation databases from adhoc 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
266
5/12/2010 10:21:06 PM Rafi
RafiA Khan
A Khan
26
The Data Mart
 Data mart
– Small, single-subject data warehouse subset
– Each is more manageable data set than data warehouse
– Provides decision support to small group of people
– Typically lower cost and lower implementation time than data

warehouse
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
26
Twelve Rules that Define
a Data Warehouse
 Data warehouse and operational environments are separated
 Data warehouse data are integrated
 Data warehouse contains historical data over long time horizon
 Data warehouse data are snapshot data captured at given point in time
 Data warehouse data are subject oriented
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
26
a Data Warehouse (continued)
 Data warehouse data are mainly read-only with periodic batch updates
from operational data
– No online updates allowed
 Data warehouse development life cycle differs from classical systems

development
 Data warehouse contains data with several levels of detail: current detail
data, old detail data, lightly summarized data, and highly summarized
data
 Data warehouse environment is characterized by read-only transactions to

very large data sets
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
26
a Data Warehouse (continued)
 Data warehouse environment has system that traces data sources,

transformations, and storage
 Data warehouse‘s metadata are critical component of this environment
 Data warehouse contains a mechanism for resource usage that enforces

optimal use of data by end users
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
27
OLAP Activities
 Generating queries
 Requesting ad hoc reports
 Conducting statistical and other analyses
 Developing multimedia applications
271
5/12/2010 10:21:06 PM Rafi
RafiA Khan
A Khan
27
Online Analytical Processing
 Advanced data analysis environment that supports decision making,

business modeling, and operations research
 OLAP systems share four main characteristics:
– Use multidimensional data analysis techniques
– Provide advanced database support
– Provide easy-to-use end-user interfaces
– Support client/server architecture
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
27
Multidimensional Data Analysis Techniques
 Data are processed and viewed as part of a multidimensional structure
 Particularly attractive to business decision makers
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
27
OLTP vs OLAP
On-Line Transaction Processing On-Line Analytical Processing

• Time-critical • Small delays tolerable
• In-place data update • Append only
• Current data • Historical and current data
• Functional transaction focus • Reporting (information
delivery) focus
• Store details only • Store summary + details
(e.g. counts and aggregates)
• Only keeps company internal data • Warehouse also keeps external data
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
27
Multidimensional Data Analysis Techniques
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
27
Multidimensional Data Analysis Techniques (continued)
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
27
OLAP Architecture
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
27
OLAP Architecture (continued)
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
27
OLAP Architecture (continued)
 Designed to use both operational and data warehouse data

 Defined as an ―advanced data analysis environment that supports decision
making, business modeling, and an operation‘s research activities‖
 In most implementations, data warehouse and OLAP are interrelated and
complementary environments
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
27
Multi Dimensional Data
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
28
Multi Dimensional Data
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
28
Drill Down & Roll Up
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
28
Drill Down & Roll Up
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
28
Fresh Produce 60 82 63 59 July
Tinned Food 30 84 79 64 August
Toiletries 50 15 46 73 Sept
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
28
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
28
30 July
Fruit
Fresh Produce Vegetables 15 August
15 Sept
Dairy
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
28
30
Fruit July
15 Sept
Dairy
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
28
30
Fruit July
15 Sept
Dairy
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
28
30
Fruit July
15 Sept
Dairy
2 4 1 1 21-30 sept
3 3 4 2 10-20 sept
1st-10 sept
2 1 3 4
5/12/2010 10:21:06 PM Rafi
RafiA Khan
A Khan
28
30
Fruit July
15 Sept
Dairy
Apples 2 4 1 1 21-30 sept

Fruits Mangoes 3 3 4 2 10-20 sept
1st-10 sept
Oranges 2 1 3 4
5/12/2010 10:21:06 PM Rafi
RafiA Khan
A Khan
29
Some Tools in the Marketplace
• Data Warehousing
• Microsoft SQL Server 2000 Data Transformation Service
• Oracle 9i Warehouse Builder
• IBM Red Brick Warehouse, and DB2
• NCR/Teradata
• SAS Data Warehousing (Warehouse Administrator)
• OLAP
• Cognos PowerPlay
• Business Objects Analytics
• Microstrategy 7i
• Microsoft SQL Server 2000 Analysis Service
+ MDX query language for decision support
+ Microsoft Data Analyzer
• Oracle 9i OLAP
Data Warehouse, OLAP, and Data Mining solutions are sometimes listed
under the title ‗Business Intelligence‘ (BI) software.
5/12/2010 10:21:06 PM Rafi
RafiA Khan
A Khan
29
Knowledge Discovery in Databases (KDD)
• Knowledge Discovery in Databases (KDD) is the automated discovery

of patterns and relationships in large databases.
• Knowledge Discovery in Databases (KDD) as it is also known, is the

nontrivial extraction of implicit, previously unknown, and potentially
useful information from data.
• KDD is the search for relationships and global patterns that exist in
large databases but are `hidden' among the vast amount of data, such
as a relationship between patient data and their medical diagnosis.
These relationships represent valuable knowledge about the database
and the objects in the database
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
29
Knowledge Discovery in Databases (KDD)
 Selection:
Extraction of the data from a database that is relevant to the
data mining analysis.
 Preprocessing:
Ensuring that values have a uniform meaning, eliminating
missing values in the data, and inaccurate (inconsistent) data.
 Data Transformation:
Converting the data into a two-dimensional table and
eliminating unwanted fields so the results are valid.
 Data mining:
The extraction of patterns from the data using by an
appropriate set of algorithms.
 Interpretation:
The transformation of the identified patterns into knowledge
5/12/2010 10:21:06 PM Rafi
RafiA Khan
A Khan
29
KDD PROCESS
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
29
Trends leading to Data Flood
 More data is generated:
– Bank, telecom, other business

transactions ...
– Scientific data: astronomy,

biology, etc
– Web, text, and e-commerce
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
29
Big Data Examples
 Europe's Very Long Baseline Interferometry (VLBI) has 16 telescopes, each

of which produces 1 Gigabit/second of astronomical data over a 25-day
observation session
– storage and analysis a big problem
 AT&T handles billions of calls per day
– so much data, it cannot be all stored -- analysis has to be done ―on the
fly‖, on streaming data
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
29
Largest databases in 2003
 Commercial databases:
– Winter Corp. 2003 Survey: France Telecom has largest decision-support

DB, ~30TB; AT&T ~ 26 TB
 Web
– Alexa internet archive: 7 years of data, 500 TB
– Google searches 4+ Billion pages, many hundreds TB
– IBM WebFountain, 160 TB (2003)
– Internet Archive (www.archive.org),~ 300 TB
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
29
From terabytes to exabytes to …
 UC Berkeley 2003 estimate: 5 exabytes (5 million terabytes) of new data

was created in 2002.
www.sims.berkeley.edu/research/projects/how-much-info-2003/
 US produces ~40% of new stored data worldwide
 2006 estimate: 161 exabytes (IDC study)
– www.usatoday.com/tech/news/2007-03-05-data_N.htm
 2010 projection: 988 exabytes
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
29
Largest Databases in 2005
Winter Corp. 2005 Commercial Database

Survey:
1. Max Planck Inst. for Meteorology ,

222 TB
2. Yahoo ~ 100 TB (Largest Data

Warehouse)
3. AT&T ~ 94 TB
www.wintercorp.com/VLDB/2005_TopTen_Survey/TopTenWinners_2005.asp
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
29
Data Growth
In 2 years, the size of the largest database TRIPLED!
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
30
Data Growth Rate
 Twice as much information was created in 2002 as in 1999 (~30% growth

rate)
 Other growth rate estimates even higher
 Very little data will ever be looked at by a human
Knowledge Discovery is NEEDED to make sense and use of data.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
30
Machine Learning / Data Mining
Application areas
 Science
– astronomy, bioinformatics, drug discovery, …
 Business
– CRM (Customer Relationship management), fraud detection, e-commerce,

manufacturing, sports/entertainment, telecom, targeted marketing, health care,
…
 Web:
– search engines, advertising, web and text mining, …
 Government
– surveillance, crime detection, profiling tax cheaters, …
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
30
Application Areas
What do you think are some of the most important and

widespread business applications of Data Mining?
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
30
Data Mining for Customer Modeling
 Customer Tasks:
– attrition prediction
– targeted marketing:
• cross-sell, customer acquisition
– credit-risk
– fraud detection
 Industries
– banking, telecom, retail sales, …
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
30
Customer Attrition: Case Study
 Situation: Attrition rate at for mobile phone customers is around

25-30% a year!
 With this in mind, what is our task?

– Assume we have customer information for the past N
months.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
30
Customer Attrition: Case Study
Task:
 Predict who is likely to attrite next month.
 Estimate customer value and what is the cost-effective offer to

be made to this customer.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
30
Customer Attrition Results
 Verizon Wireless built a customer data warehouse

 Identified potential attires
 Developed multiple, regional models
 Targeted customers with high propensity to accept the offer
 Reduced attrition rate from over 2%/month to under 1.5%/month (huge
impact, with >30 M subscribers)
(Reported in 2003)
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
30
Assessing Credit Risk: Case Study
 Situation: Person applies for a loan
 Task: Should a bank approve the loan?
 Note: People who have the best credit don‘t need the loans, and people
with worst credit are not likely to repay. Bank‘s best customers are in the
middle
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
30
Credit Risk - Results
 Banks develop credit models using variety of machine learning methods.
 Mortgage and credit card proliferation are the results of being able to
successfully predict if a person is likely to default on a loan
 Widely deployed in many countries
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
30
e-commerce
 A person buys a book (product) at Amazon.com
What is the task?
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
31
Successful e-commerce – Case Study
 Task: Recommend other books (products) this person is likely to buy
 Amazon does clustering based on books bought:
– customers who bought ―Advances in Knowledge Discovery and Data

Mining‖, also bought ―Data Mining: Practical Machine Learning
Tools and Techniques with Java Implementations‖
 Recommendation program is quite successful
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
31
Security and Fraud Detection - Case Study
 Credit Card Fraud Detection

 Detection of Money laundering
– FAIS (US Treasury)
 Securities Fraud
– NASDAQ KDD system
 Phone fraud
– AT&T, Bell Atlantic, British Telecom/MCI
 Bio-terrorism detection at Salt Lake Olympics 2002
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
31
Data Mining and Privacy
 in 2006, NSA (National Security Agency) was reported to be mining years

of call info, to identify terrorism networks
 Social network analysis has a potential to find networks
 Invasion of privacy – do you mind if your call information is in a gov

database?
 What if NSA program finds one real suspect for 1,000 false leads ?
1,000,000 false leads?
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
31
Related Fields
Machine Visualization
Learning
Data Mining and
Knowledge Discovery
Statistics Databases
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
31
Statistics, Machine Learning and
Data Mining
 Statistics:
– more theory-based
– more focused on testing hypotheses
 Machine learning
– more heuristic(experience-based techniques that help in
problem solving, learning and discovery)
– focused on improving performance of a learning agent
– also looks at real-time learning and robotics – areas not part of
data mining
 Data Mining and Knowledge Discovery
– integrates theory and heuristics
– focus on the entire process of knowledge discovery, including
data cleaning, learning, and integration and visualization of
results
 Distinctions are fuzzy
5/12/2010 10:21:06 PM
witten&eibe Rafi
RafiA Khan
A Khan
31
Data mining
Many Definitions…
– A short one…
Search for Valuable Information in Large Volumes of Data.
– A long one…
Exploration & Analysis, by Automatic or Semi-Automatic Means,
of Large Quantities of Data in order to Discover Meaningful
Patterns & Rules.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
31
Data Mining
 Data Mining is the step in the process of knowledge discovery in

databases, that inputs predominantly cleaned, transformed data, searches
the data using algorithms, and outputs patterns and relationships to the
interpretation/evaluation step of the KDD
 Data mining is a process that uses a variety of data analysis tools to

discover patterns and relationships in data that may be used to make valid
predictions.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
31
Data Mining
 Data Mining constitutes one step in the KDD process.
 The transformed data is used in the data mining step. It is in this step that
the actual search for patterns of interest is performed.
 The appropriate data mining algorithm (linear/logistic regression, neural

networks, association rules, etc.) for the data mining task (classification,
database segmentation, rule generation, etc) are applied.
 It is necessary to remove redundant and irrelevant patterns from the set of

useful patterns. Once a set of good patterns have been discovered, they
then have to be reported to the end user. This can be done can be done
textually, by way of reports or using visualizations such as graphs,
spreadsheets, diagrams, etc.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
31
Data Mining
 Data mining tools do the following:
– Analyze data
– Uncover problems or opportunities hidden in data relationships
– Form computer models based on their findings
– Use models to predict business behavior
 Require minimal end-user intervention
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
32
Data Mining (continued)
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
32
Major Data Mining Tasks
 Classification: predicting an item class

 Clustering: finding clusters in data
 Associations: e.g. A & B & C occur frequently
 Visualization: to facilitate human discovery
 Summarization: describing a group
 Deviation Detection: finding changes
 Estimation: predicting a continuous value
 Link Analysis: finding relationships
 …
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
32
Data Mining Tasks: Classification
Learn a method for predicting the instance class from

pre-labeled (classified) instances
Many approaches:
Statistics,
Decision Trees,
Neural Networks,
...
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
32
Data Mining Tasks: Clustering
Find “natural” grouping of

instances given un-labeled data
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
32
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
32
We want to know ...
 Given a database of 100,000 names, which persons are the least likely to
default on their credit cards?
 Which types of transactions are likely to be fraudulent given the

demographics and transactional history of a particular customer?
 If I raise the price of my product by Rs. 2, what is the effect on my ROI?
 If I offer only 2,500 airline miles as an incentive to purchase rather than

5,000, how many lost responses will result?
 If I emphasize ease-of-use of the product as opposed to its technical

capabilities, what will be the net effect on my revenues?
 Which of my customers are likely to be the most loyal?

Data Mining helps extract such information
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
32
Major Data Mining Characteristics and Objectives
 Data are often buried deep
 Client/server architecture
 Sophisticated new tools--including advanced visualization tools
 End-user miner 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
327
5/12/2010 10:21:06 PM Rafi
RafiA Khan
A Khan
32
Difference between OLAP & Data Minig
OLAP is part of the spectrum of decision support tools. Traditional query and
report tools describe what is in a database. OLAP goes further; it‘s used to
answer why certain things are true. The user forms a hypothesis about a
relationship and verifies it with a series of queries against the data. For
example, an analyst might want to determine the factors that lead to loan
defaults. He or she might initially hypothesize that people with low
incomes are bad credit risks and analyze the database with OLAP to verify
(or disprove) this assumption. If that hypothesis were not borne out by the
data, the analyst might then look at high debt as the determinant of risk. If
the data did not support this guess either, he or she might then try debt
and income together as the best predictor of bad credit risks.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
32
Difference between OLAP & Dataminig
OLAP analysis is essentially a deductive process.
The OLAP analyst generates a series of hypothetical patterns and

relationships and uses queries against the database to verify them or disprove
them.
It becomes much more difficult and time-consuming to find a good

hypothesis, when the number of variables being analyzed is in the dozens or
even hundreds? and analyze the database with OLAP to verify or disprove it.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
32
Difference between OLAP & Dataminig
Data mining , rather than verify hypothetical patterns, it uses the data itself to
uncover such patterns. It is essentially an inductive process.
For example, suppose the analyst who wanted to identify the risk factors for
loan default were to use a data mining tool. The data mining tool might
discover that people with high debt and low incomes were bad credit risks (as
above), but it might go further and also discover a pattern the analyst did not
think to try, such as that age is also a determinant of risk.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
33
Data mining Applications
 Many organizations are using data mining to help manage all phases of the
customer life cycle, including acquiring new customers, increasing revenue
from existing customers, and retaining good customers.
 Telecommunications and credit card companies are two of the leaders in

applying data mining to detect fraudulent use of their services.
 Insurance companies and stock exchanges are also interested in applying

this technology to reduce fraud.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
33
Retail/Marketing
 Identify buying patterns from customers
 Find associations among customer demographic characteristics

 Predict response to mailing campaigns
 Market basket analysis

 Retailers are making more use of data mining to decide which products to
stock in particular stores (and even how to place them within a store), as
well as to assess the effectiveness of promotions and coupons.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
33
Medicine
 Characterise patient behaviour to predict office visits
 Identify successful medical therapies for different illnesses
 Medical applications are another fruitful area: data mining can be used to
predict the effectiveness of surgical procedures, medical tests or
medications.
 Pharmaceutical firms are mining large databases of chemical compounds
and of genetic material to discover substances that might be candidates for
development as agents for the treatments of disease.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
33
Banking
 Detect patterns of fraudulent credit card use
 Identify `loyal' customers
 Predict customers likely to change their credit card affiliation
 Determine credit card spending by customer groups
 Find hidden correlations between different financial indicators
 Identify stock trading rules from historical market data
 Companies active in the financial markets use data mining to determine

market and industry characteristics as well as to predict individual
company and stock performance.
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
33
Insurance and Health Care
 Claims analysis - i.e which medical procedures are claimed together
 Predict which customers will buy new policies
 Identify behaviour patterns of risky customers
 Identify fraudulent behaviour
Transportation
 Determine the distribution schedules among outlets
 Analyse loading patterns
 Analyse loading patterns
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
33
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
33
5/12/2010 10:21:06 PM Rafi

RafiA Khan
A Khan
33
Friday, July 18, 2008 RafiRafi
A. Khan
A Khan
33
Systems Development Life Cycle
 Four phases
– Planning
– Analysis
– Design
– Implementation
 Cyclical
 Can return to other phases
 Waterfall model

A. Khan
A Khan
33
Tools
 Computer-aided software design tools

– CASE tools- Oracle 9i developer suite, Rational rose,
Paradigm Plus
 RAD design tools- Sybase power Designer. Oracle Internet

Development Suite, Rational RequisitePro
 Code debugging methods -
 Testing and quality assurance tools - Red Views WebLoad,

Load Runner, Rational RequisitePro, SilkPerformer

A. Khan
A Khan
34
Successful Project Management
 Define requirements
 Manage change
 Get support from upper management
 Establish timelines, milestones, and budgets based on realistic

goals
 Involve users
 Document everything

A. Khan
A Khan
34
Implementation Failures
 Lack of stakeholder involvement
 Incomplete requirements
 Unrealistic expectations
 Project champion leaves
 Lack of skill or expertise
 Inadequate human resources
 New technologies

A. Khan
A Khan
34
Project Management Tools
 Project management software can allow:
– Collaboration among disparate teams
– Resource and program management
– Portfolio management
– Web enabled
– Analyses of project data
S/W Examples : Microsoft project, PlanView, ActiveProject

A. Khan
A Khan
34
Alternative Development Methodologies
 Parallel Development
– Multiple development on separate systems (Design & Implementation
Phases)
– Database, Model base, UI and Knowledge can be developed in parallel
 RAD
– Quick development allowing fast, but limited functionality
– Methods of RAD
• Phased development
– Sequential serial development (Break system into Pieces)
• Prototyping ( Analysis, Design & Implementation repeatedly)
– Rapid development of portions of projects for user input and
modification
– Small working model or may become functional part of final
system
• Throwaway prototyping
– Pilot test or simple development platforms
A. Khan
A Khan
34
A. Khan
A Khan
34
A. Khan
A Khan
34
Tools
 Computer-aided software design tools
– CASE tools - Oracle 9i developer suite, Rational rose, Paradigm Plus

 RAD design tools- Sybase power Designer. Oracle Internet Development
Suite, Rational RequisitePro
 Code debugging methods -
 Testing and quality assurance tools - Red Views WebLoad, Load Runner,
Rational RequisitePro, SilkPerformer

A. Khan
A Khan
34
DSS Prototyping
 Short steps
– Planning
– Analysis
– Design
– Prototype
 Immediate user feedback
 Iterative
– In development of prototype

A. Khan
A Khan
34
Successful Project Management
 Define requirements
 Manage change
 Get support from upper management
 Establish timelines, milestones, and budgets based on realistic

goals
 Involve users
 Document everything

A. Khan
A Khan
34
Implementation Failures
 Lack of stakeholder involvement
 Incomplete requirements
 Unrealistic expectations
 Project champion leaves
 Lack of skill or expertise
 Inadequate human resources
 New technologies

A. Khan
A Khan
35
Project Management Tools
 Project management software can allow:
– Collaboration among disparate teams
– Resource and program management
– Portfolio management
– Web enabled
– Analyses of project data
S/W Examples : Microsoft project, PlanView, ActiveProject

A. Khan
A Khan
35
Agile Development
 Rapid prototyping used for rapidly changing requirements
 Used for:
– Unclear or rapidly changing requirements
– Speedy development
 Heavy user input
 Incremental delivery with short time frames
 Tend to have integration problems
Example : Extreme Programming (XP), Scrum, Crystal.

A. Khan
A Khan
35
DSS Prototyping
 Advantages  Disadvantages
– User and management – Changing requirements
involvement – May not have thorough
– Short user-reaction understanding of benefits
time(Feedback from user) and costs
– Short intervals between – Poorly tested
iterations – Dependencies, security,
– Low cost & Short and safety may be ignored
development time – High uncertainty
– Improved user – Problem may get lost
understanding of system
– Reduction in quality
– Higher costs due to
multiple productions
A. Khan
A Khan
35
Change Management
 Crucial to DSS
 People resistant to change
 Examine cause of change
 May require organizational culture shift
 Lewin-Schein change theory steps
– Unfreeze
• Create awareness of need for change
• People support what they help create
– Move
• Develop new methods and behaviors
• Create and maintain momentum
– Refreeze
• Reinforce desired changes
• Establish stable environment

A. Khan
A Khan
35
DSS Technology Levels
DSS primary tools
–Fundamental elements
•Programming languages, graphics, editors, query systems
DSS generator (engine)
–Integrated software package for building specific DSS
•Modeling, report generation, graphics, risk analysis
•These range from spreadsheets such as Excel—perhaps with some add-
ins or a more sophisticated generator such as MicroStrategy‘s DSS
Architect.
Specific DSS
–For some problem types there may be a commercially available package
that can be acquired and customized
DSS primary tools are used to construct integrated tools that are used to
construct specific tools

A. Khan
A Khan
35
DSS
 Hardware
– PCs to multiprocessor mainframes
 Software
– Involves multiple criteria
– Develop in house, outsource, or buy off the shelf
– Off the shelf software rapidly updated; many on market
– Prices fluctuate
– Different tools available

A. Khan
A Khan
35
DSS
 Team developed DSS requires substantial effort to build and
manage
 End user developed DSS
– Decision-makers and knowledge workers develop to
solve problems or enhance productivity
• Advantages
– Short delivery time
– User requirements specifications are eliminated
– Reduced implementation problems
– Low costs
• Risks
– Quality may be low
– May have lack of documentation
– Security risks may increase

A. Khan
A Khan
35
DSS
 Microstrategy 8
 Hyperion System 9 BI+
 Business Object XI
 Microsoft Biztalk server2004
 IBM Websphere Commerce Suite
 Oracle Daily Business Intelligence(DBI)

A. Khan
A Khan
35

Decision Support System

Caricato da

Informazioni sul documento

Descrizione originale:

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Decision Support System

Caricato da

Copyright:

Formati disponibili

5/12/2010

Data & Information

The Business School

 Fact – statement of some element of truth about a subject matter or a domain.

Example: milk is white, sun rises in east

 Intelligence – capacity to acquire, store, improve and apply knowledge

 Common sense – natural ability to sense, judge or perceive situations ; grows

 Memory – ability to store and retrieve relevant experience at will, is part of

 Learning – is knowledge or skill that is acquired by instruction or study

5/12/2010 10:21:02 PM Rafi A Khan 2

 Knowledge. Information once analyzed, understood, and explained is

 knowledge or foreknowledge (predictions or forecasts).

 Data vs. Information

• A ―given,‖ or fact; a number, a statement, or a picture

• Represents something in the real world

• The raw materials in the production of information

• Data that have meaning within a context

• Data after manipulation

– Example: customer survey

• Reading through data collected from a customer survey with

• When manipulated, the surveys may provide useful

– Computer-based ISs take data as raw material, process it,

Figure 1.1 Input-process-output

Activities in an Information System

INPUT PROCESS OUTPUT

Daily sales account

List of low stock items to be re-ordered

List of overstock items

Long overdue payments

Profit and loss account

Used to streamline day to day operations called Operational

 Slow or fast moving items

 Reliable supplier of items

Used to improve profitability of shop called Tactical information

Whether to stock different varieties of items

Whether to start a new branch in a different locality

Whether to start an e-shop

Information to expand business and explore new opportunities

Strategic : Needed for long range planning and directions.

• This is less/un- structured.

Tactical : Needed to take short range decisions to improve

• Profitability and Performance.

Operational : Needed for day to day operations of the

• Eg: Daily Sales, Billing.

System as a group of interrelated components working together toward a

Such a system has three basic interacting components or functions:

 Processing: Involves transformation process that converts input into output.

 Output: Involves transferring elements that have been produced by

Input Process Output

Feedback and Control

Fig. Showing Elements of a System

A system with feedback and control components is sometimes called a

 Feedback: It is data about the performance of a system. It is actually

 Control: It involves monitoring and evaluating feedback to determine

 A system does not exist in a Vacuum; rather, it exists and function in an

 If a system is one of the components of a larger system, it is then

 A large system can be split or decomposed into smaller subsystems

 The decomposition of a system into subsystems can be in a serial

 In a serial system processing, the entire output of a subsystem is the

 In the matrix arrangement the different outputs go to different sub-

 If the process of input transformation is not visible and

 Most of the systems can be viewed in a hierarchical structure.

The systems can be classified in different categories based on the