Transcript Decision Support Systems

Decision Support Systems
Chapter 3: Decision Support Systems
Concepts, Methodologies and
Technologies: An Overview
Learning Objectives
• Understand possible decision support systems(DSS)
configurations.
• Understand the key differences and similarities between DSS
and BI systems.
• Describe DSS characteristics & capabilities.
• Understand the essential definitions of DSS.
• Understand DSS components and how they integrate.
• Describe the components and structure of each DSS
component: the data management subsystem, the model
management subsystem, the user interface (dialog)
subsystem, the knowledge-based management system and
the user.
Learning Objectives
• Explain internet impact on DSS and vice versa.
• Explain the unique role of the user in DSS versus management
information systems (MIS).
• Describe DSS hardware and software platforms.
• Understand important DSS classifications.
• Become familiar with some DSS application areas and
applications.
• Understand important current DSS issues.
DSS Configurations
• Depends on the management-decision situation and the
specific technologies used for support.
• Technologies are typically deployed over the web and are
assembled from:
–
–
–
–
Data
Models
User Interface
Knowledge (optional)
• Components are emphasized by the support provided (i.e.
Model-Oriented DSS -> Model (spreadsheets), Data-Oriented
DSS -> Database).
DSS Description
DSS
BI
Support the solution of a certain problem. Monitor Situations.
Identify problems and/or opportunities
Evaluate an opportunity.
using analytical methods.
User must identify wither a particular
situation warrants attention and then
analytical methods can be applied.
Utilizes models and data access, but they Utilized models and data access.
have their own databases that are used to Arguably considers DSS part of its
solve a specific problem or set of problems internal building blocks.
(DSS Applications)
Built to solve a specific problem and Focuses on reporting and identification
of problems by scanning data extracted
include their own databases
from a data warehouses.
Some DSS Definitions
• Systems designed to support managerial decision-making in unstructured
problems.
– Little (1970): Model based set of procedures for processing data and
judgements to assist manager in his decision making.. Must be simple, robust,
easy adaptive, complete
– Moore and Chang (1980): Structured problems are structured, because we
treat them in that way.. DSS is an expandable system capable of supporting ad
hoc data analysis and decision modeling for planning the future
– Bonczek (1980): A computerbased system with 3 interacting components, a
language system, a knowledge system, problem processing system
– Keen (1980): Final system can be developed by the adaptive process of
learning and evolution by the user, the DSS builder, and the DSS itself
Generic DSS Description
• DSS is an approach (or methodology) for supporting decision making.
• Uses Interactive, Flexible, Adaptive CBIS developed for supporting the
solution to a specific nonstructured management problem, it uses data,
provides an easy user interface, and can incorporate the decision maker
own insight.
• Includes models and is developed (possibly by end users) through an
interactive and iterative process.
• Supports all phases of decision making and may include knowledge
component.
• Can be used by a single user on a PC or can be Web based for use by many
people in several locations.
DSS characteristics and capabilities
• There is no consensus on exactly what a DSS is, and there is obviously no
agreement on the standard characteristics and capabilities of DSS.
Terms
• Business Analytics (BA): implies the use of models and data to improve
the organization’s performance or competitive posture. The focus is the
use of models, even if they are deeply buried inside the system.
• Data mining and OLAP systems have models embedded in them but are
still not well understood in practice.
• Web analytics: is an approach to using analytics tools on real-time Web
information to assist in decision making .
• Predictive analytics: describes the business analytics methods of
forecasting problems and opportunities rather that simply reporting them
as they occur. It utilized advanced forecasting and simulation models.
Component of DSS
• DSS application can be composed of
–
–
–
–
Data Management subsystem.
Model Management subsystem.
User Interface subsystem.
Knowledge Management subsystem.
Component of DSS
DATA MANAGEMENT SUBSYSTEM:
– Includes database that contains relevant data for the situation and is managed by DBMS.
– Can be interconnected with the corporate data warehouse [A repository for corporate
relevant decision-making data], usually, the data are stored or accessed via database
Web server.
MODEL MANAGEMENT SUBSYSTEM [MBMS]:
– Software package that includes financial, statistical, management science, or other
quantitative models that provide the system’s analytical capabilities and appropriate
software management.
– Modeling languages for building custom models are included.
– Often called Model Base Management System [MBMS].
– Can be connected to corporate or external storage of models.
– Model solution methods and management systems are implemented in Web
development systems (such as Java) to run on application servers.
Component of DSS
THE USER INTERFACE SUBSYSTEM
– User communicated with and commands the DSS through the user interface subsystem.
– User is considered part of the system.
– Researchers assert that some of the unique contributions of DSS are derived from the
intensive interaction between the computer and the decision maker.
– Web browser provides a familiar, consistent graphical user interface (GUI) for most DSS.
THE KNOWLEDGE-BASE MANAGEMENT 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 repository (part of a Knowledge
management system [KMS] (The Organizational Knowledge Base)
– Knowledge may be provided via Web servers.
By definition DSS must include the three major components DBMS, MBMS and
user interface, the KBMS is optional but it can provide many benefits by
providing intelligence in and to the three major components. The user may be
considered a component of a DSS.
How DSS Component integrate
-
Can be connected to a corporate intranet, extranet or the internet.
Component communicate through web technologies.
Web browsers are excellent choice for UI.
A Web Based DSS Architecture
Web Browser
Web
Server
Optimization/
Simulation, etc.
Server
Application
Server
Data
Server
Data
Warehouse
or DBMS
Read about DSS &
the Web mutual
impact
DATA MANAGEMENT SUBSYSTEM:
The Data management subsystem is composed of the following elements:
1- DSS database
2- DBMS
3- Data Directory
4- Query Facility
DATA MANAGEMENT SUBSYSTEM:
DATABASE
• Interrelated data extracted from various
sources, stored for use by the organization,
and queried.
– Internal data, usually from TPS.
– External data from government agencies, trade
associations, market research firms, forecasting
firms.
– Private data or guidelines used by decisionmakers.
DATA MANAGEMENT SUBSYSTEM:
Database Management System
• Data Organization
– Should DSS have their own Databases.
• Data Extraction ETL
– The process of capturing data from several
sources & the integration process.
Data Management Subsystem
Query Facility
• Access, manipulate and query data
– Accepts requests for data
– Consults the data directory
– Formulates the direct requests
– Reports the results (on a web structured page)
Ex: Search for all sales in the Southeast region
during June 2006 and summarize sales by
salesperson.
Data Management Subsystem
Data Directory
• Catalog of all data
– Contains data definitions
– Answers questions about the availability of data
items
– Source
– Meaning
– Allows for additions, removals, and alterations
Key Database Management System
Issues
•
•
•
•
Data Quality: [GIGO].
Data Integration: Single version of the truth.
Scalability.
Data Security.
Model Management Subsystem
• Components:
– Model base
– Model base management system
– Modeling language
– Model directory
– Model execution, integration, and command
processor
Model Management Subsystem
Models (Model Base)
Strategic, tactical operational•
Statistical, financial, marketing,•
Management science,
Accounting engineering
Model building blocks•
Model Directory
Model Base Management
Model execution,
integration and
command processor
Modeling commands: creation•
Maintenance: update•
Database interface•
Modeling language•
Data
Management
Interface
Management
Knowledge based
subsystem
Model Management Subsystem
Models in the Model Base
• Clasification with respect to time span
– Strategic models: Supports top management decisions
– Tactical models: Used primarily by middle management to allocate
resources
– Operational models: Supports daily activities
• Analytical models
– Used to perform analysis of data for strategic, tactical and operational
decisions
• Also there are model building blocks and routines, like
– Random number generation, curve fitting, present value computation
Model Management Subsystem
Model Management Activities
• Model execution
– Controls running of model
• Model integration
– Combines several models’ operations
• Model command processor
– Receives model instructions from user interface
– Routes instructions to MBMS or model
execution or integration functions
Model Management Subsystem
Model Base Management System
• Functions:
– Model creation
– Model updates
– Model data manipulation
– Generation of new routines
Model Management Subsystem
Model Directory
• Catalog of models and software
• Definitions
• Functions to answer questions about
availability and capability of the models
User Interface Management System
• Interacts with model, data and knowledge management
subsystems
• Includes a natural language processor or standard objects (pull
down menus, internet browsers)
• Includes GUI, frequently by web browsers
• Accomodates the user with a variety of input devices
• Provides output with a various formats and output devices.
• Provides help capabilities
User Interface Management System
• Stores data
• Process multiple functions concurrently
• Support cummunication b/w users and tech. Staff
• Provides training
• Provides flexibility and adaptiveness
• Captures, stores and analyzes the dialog usage
User Interface System
Data management
and DBMS
Knowledge-based
system
Model management
and MBMS
User Interface Management System (UIMS)
Natural Language Processor
Input
Action
Languages
Output
Display
Language
PC Display
Printers, Plotters
Users
Based on Figure 3.6, Schematic View of the User Interface
New User Interface Developments
• Voice/speech recognition (Ex: Clarissa developed at NASA Ames
Research Team)
• Handwriting recognition
• Translation of text into voice
• Automatic real time natural language speech translator (on
process)
• Displays are getting better by crisp images, holographic displays
(Ex: LCD panels developed at Philips Research)
New User Interface Developments
• Tactile interfaces (Ex: Immersion Corp.’s Cyberforce Sys.
includes a spandex glove that sense the doctors get
when performing surgery)
• Videoconferencing (Mİcrososft developed RingCam, an
omnidirectional videocamera to view the entire room
• Gesture interface that utilizes holographic displays
New Developments in DSS
• Access data from a data warehouse, use models from OLAP or
data mining tools.
• Web technologies
– Link components for accessing data and knowledge via web browsers
or web like user interfaces
– Enable virtual teams to collaborate
– Reduced technological barriers; made transactions easier and less
costly by mobile communication
• Hardware shrinks in size, increases in speed etc.
• Faster, intelligent search engines by AI techniques
• In the future some DSS may include emotions, mood, tacit
values and other soft factors
Knowledge-Based Management System
• Expert or intelligent agent system component to enhance the
operation of other DSS components
• Complex problem solving in unstructured/semistructured systems
• Gives aid in models selection and construction
• Enhances operations of other components
• A DSS that includes this optional component is called an intelligent
DSS, DSS/ES, expert support system or knowledge-based DSS
• Caution: A KMS is a text oriented DSS; not a Knowledge-based
Management System
DSS Hardware
• Hardware affects the functionality and usability of DSS,
De facto standard
• Major hardware options: mainframe, server,
workstation, PC, client/server system
• Distributed DSS runs on different networks including
internet, intranet, extranet
• Access by client pc’s or by mobile devices notebook pc’s,
PDA’s, cell phones
DSS Hardware
• Models run either on the server, mainframe, any
exernal system or client pc
• Web server with DBMS:
– Operates using browser
– Data stored in variety of databases
– Can be mainframe, server, workstation, or PC
– Any network type
– Access for mobile devices
DSS Classifications
• Association for Information Systems Special
Interest Group In DSS [AIS SIGDSS]
– Communications-driven and group DSS
– Data-driven DSS
– Document-driven DSS, data mining, and
management ES applications
– Model-driven DSS
• Holsapple and Whinston
– Text oriented, database oriented, spreadsheet
oriented, solver oriented, rule oriented, or
compound
DSS Classifications
• Alter
– Extent to which outputs can directly support or
determine the decision
– Data oriented or model oriented
DSS Classifications
• Donovan and Madnick
– Institutional
– Problems of recurring nature
• Ad hoc
– Problems that are not anticipated or are not
repetitive
• Hackathorn and Keen
– Personal support, group support, or organizational
support
DSS Classifications
• GSS v. Individual DSS
– Decisions made by entire group or by one decision
maker
• Custom made v. vendor ready made
– Generic DSS may be modified for use
• Database, models, interface, support are built in
• Addresses repeatable industry problems
• Reduces costs
Web and DSS
•
•
•
•
•
•
•
Data collection
Communications
Collaborations
Download capabilities
Run on Web servers
Simplifies integration problems
Increased usability features