Second S302 class session

Download Report

Transcript Second S302 class session

Chapter 6:
Applying Information
Technology Managerial Support Systems
Review: Organizational Systems
Transaction Processing Systems
 Enterprise Resource Planning Systems
 Data Warehousing
 Office Automation
 Groupware
 Intranets
 Factory Automation

Managerial Support Systems
Decision Support Systems
 Data Mining
 Group Support Systems
 Geographic Information Systems
 Executive Information Systems
 Artificial Intelligence, including Expert
Systems and Neural Networks
 Virtual Reality

Decision Support Systems
Computer-based system, usually
interactive, designed to assist managers in
making decisions
 Incorporates both data and models, and
usually intended to assist in the solution of
semi- or unstructured problems

Components of a DSS



Data Management
Model Management
Dialog Management,
or the User Interface
DSS Examples
Scheduling ambulances and ambulance
technicians in Montreal
 Evaluating motor-vehicle legislation in the
state of Idaho
 Media planning for print media
(advertising) in India
 Police-beat allocation in a California city
(also Geographic I.S.)

DSS Examples (continued)
United Airlines’ System Operation Advisor
-- helps aircraft controllers deal with
aircraft shortage problems arising because
of delayed flights or mechanical problems
 Planning municipal solid waste
management
 Scheduling and routing home health care
nurses

Data Mining

Employs a variety of
techniques (such as
neural networks) to
search or “mine” for
small “nuggets” of
information from the
vast quantities of data
stored in an
organization’s data
warehouse
Uses of Data Mining
Market segmentation -- identify the
common characteristics of customers who
buy the same products from your company
 Customer churn -- predict which customers
are likely to leave your company and go to
a competitor
 Fraud detection -- identify which
transactions are most likely to be fraudulent

Uses of Data Mining (continued)
Direct marketing -- which prospects should
be included in a mailing list
 Interactive marketing -- predict what each
individual accessing a Web site in most
likely interested in seeing
 Market basket analysis -- understand what
products or services are commonly
purchased together

Uses of Data Mining (continued)
Trend analysis -- reveal the difference
between a typical customer this month
versus last month
 Identification of patterns/trends -- scrutiny
of the data to identify patterns

Group Support Systems
System designed to make group sessions
more productive by supporting such group
activities as brainstorming, issue
structuring, voting, and conflict resolution
 A variant of DSS in which the system is
designed to support a group
 A specialized type of groupware

Motivations for GSS


Increased number of
meetings and teams
Many group-based
activities are
inefficient
GSS Characteristics
Parallel human processing
 Equal opportunity for participation
 Anonymity
 Complete record of meeting
 Output of one phase leads to next
 Can more easily apply structure

Geographic Information Systems
A computer-based system designed to
collect, store, retrieve, manipulate, and
display spatial data
 A spatially based DSS
 Typically a digitized map with other data
linked to the map coordinates

How GIS Works

Two basic ways to represent spatial data:
– By rasters
» Grids of equal-sized cells grouped
or linked to make lines and shapes
» Values of cells vary
» Example: Satellite images, pixels on screen
– By vector
» Points, Lines, and Polygons
» Approximates curves, can link into networks
» Example: Property boundaries, sales territories
A Closer Look: Vectors

Geographic data are:
– type of feature
» several lines connect to form a road
» polygons start and end at same point
– where it is in reference system
» (x,y) for start and end of each line segment

Attribute data are:
– descriptive values associated to feature
» name of street segment
» number of people in house
» average household income for county
Spatial Analysis

Organizing features in layers like clear map
overlays allows comparisons:
– between features at same place
– between same attributes at different places

Can answer questions like:
– What is adjacent to this feature?
– Where is the closest something to this feature?
– What points are contained within this feature?

Can display map features based on attribute
values (called Thematic Mapping)
Sources of GIS Data

Geo-referenced data:
– #1 Government
» Census Bureau, USGS, NASA, Dept. of Defense
– #2 Create yourself
» Digitize from map, Global Positioning System,
Geocode from own databases
– #3 Buy it or download it
» Businesses that collect, repackage, aggregate public
and private sources of information
Executive Information Systems
A computer application designed to be used
directly by top managers, without the
assistance of intermediaries, to provide the
executive easy on-line access to current
information about the status of the
organization and its environment
 Now usually made available to most levels
of management

EIS continued


Traditionally, an EIS was to support strategic
planners at the “top” of the organization
Today, an EIS may be used to provide status
information to all workers who “need to know”
EIS at
strategic
apex
versus
at all
levels
Characteristics of an EIS
Primarily used for tracking and control
 Customized to the individual executive
(at least top-level executives)
 Graphical
 Easy to use
 Incorporates both hard and soft data

Artificial Intelligence

The study of how to
make computers do
things that are
presently done better
by people
AI Research Areas
Natural languages
 Robotics
 Perceptive systems
 Expert systems
 Neural networks

Expert Systems


One branch of
Artificial Intelligence
(AI)
Concerned with
building systems that
incorporate the
decision-making logic
of a human expert
Major Pieces of an Expert
System
Knowledge base (developed by a
knowledge engineer working with the
expert or experts)
 Inference engine
 User interface

Obtaining an Expert System
Buy a fully developed system (e.g.,
Lending Advisor)
 Use an artificial intelligence (AI) shell, also
called an expert system shell
 Have an expert system custom-built by
internal or external knowledge engineers

Examples of Expert Systems
MYCIN - diagnose blood diseases
(Stanford)
 CATS-1 - diagnose mechanical problems in
diesel locomotives (GE)
 Dipmeter - provide advice when drill bit
gets stuck while drilling an oil well
 Magic - determine human services benefits
(Merced County, CA)

More Examples of Expert
Systems
Credit Clearing House - provide
information to subscribers about firms in
apparel industry (D&B)
 MOCA - schedule routine maintenance on
American Airlines’ entire fleet
 MSE - Market Surveillance Expert - assist
in investigating insider trading (American
Stock Exchange)

Neural Networks
Named after study of how human nervous
system works
 Use statistical analysis to recognize patterns
from vast amounts of data by a process of
adaptive learning
 Consist of software that attempts to emulate
the processing patterns of the biological
brain

Examples of Neural Networks
BankAmerica - neural network evaluates
commercial loan applications
 American Express - system reads
handwriting on credit card slips
 State of Wyoming - system reads handprinted numbers on tax forms
 Arco and Texaco - neural network helps
pinpoint oil and gas deposits

More Examples of Neural
Networks
Speigel uses a neural network to prune its
catalog mailing list to eliminate those who
are unlikely to order again
 The Fidelity Disciplined Equity Fund
(Bradford Lewis) uses a neural network to
select stocks that are undervalued

Virtual Reality
The use of computer-based systems to
create an environment that seems real to
one or more sense (usually including sight)
 Used to design dashboard and controls of
car, simulate a tank battle, and enable
pinpoint control of radiation therapy
