Академический Документы
Профессиональный Документы
Культура Документы
Chapter 13
Decision Support Systems
13-1
Copyright 2001 Prentice-Hall, Inc.
Programmed decisions
repetitive and routine
have a definite procedure
Nonprogrammed decisions
Novel and unstructured
No cut-and-dried method for handling
problem
Intelligence
Design
Choice
Review
13-5
Degree of
problem
structure
Semistructured
Unstructured
Management
control
Strategic
planning
Accounts
receivable
Tanker fleet
mix
Order entry
Short-term
forecasting
Warehouse and
factory location
Inventory
control
Production
scheduling
Mergers and
acquisitions
Cash
management
Budget
preparation
New product
planning
PERT/COST
systems
Sales and
production
R&D planning
13-6
13-7
13-8
13-9
Retrieve
information
elements
Little
Analyze
entire
files
Prepare
reports
from
multiple
files
Estimate
decision
consequences
Degree of
complexity of the
problem-solving
system
Propose
decisions
Make
decisions
Degree
of
problem
solving
support
Much
13-10
A DSS Model
Environment
Individual
problem
solvers
Report
writing
software
Other
group
members
GDSS
GDSS
software
software
Mathematical
Models
Database
Decision
support
system
Environment
Legend:
Data
Communication
Information
13-12
Database Contents
Report writers
Special reports
Periodic reports
COBOL or PL/I
DBMS
Mathematical models
Simulations
Special modeling languages
Groupware or GDSS
13-13
13-14
Improved communications
Improved discussion focus
Less wasted time
13-15
GDSS Environmental
Settings
Synchronous exchange
Members meet at same time
Committee meeting is an example
Asynchronous exchange
Members meet at different times
E-mail is an example
GDSS Types
Decision rooms
Legislative session
Computer-mediated conference
MEMBER
PROXIMITY
Dispersed
Smaller
Larger
Decision
Room
Legislative
Session
Local Area
Decision
Network
ComputerMediated
Conference
13-18
Groupware
Functions
E-mail
FAX
Voice messaging
Internet access
Lotus Notes
Popular groupware product
Handles data important to managers
13-19
O = optional feature
Novell
GroupWise
X
X
X
X
O
X
X
3
X
X
O
O
13-20
13-21
History of AI
Early history
Past 2 decades
Expert
systems
AI
hardware
Robotics
Natural
language
Learning
Neural
networks
Perceptive
systems
(vision,
hearing)
Artificial Intelligence
13-23
13-24
User
Instructions &
information
Solutions &
explanations
Knowledge
User
interface
Inference
engine
Expert
system
Development
engine
Expert and
knowledge engineer
Knowledge
base
Problem
Domain
An Expert
System Model
13-25
User interface
Knowledge base
Inference engine
Provides reasoning
Interprets knowledge base
Development engine
13-26
User Interface
User enters:
Instructions
Information
Questions
Problem solutions
13-27
Knowledge Base
13-28
Conclusion
Evidence
Evidence
Conclusion
Evidence
Evidence
Evidence
Evidence
Evidence
Evidence
13-29
Rule Selection
13-30
Inference Engine
13-31
Forward Reasoning
(Forward Chaining)
Rule 1
IF A
THEN B
Rule 2
IF C
THEN D
Rule 3
IF M
THEN E
Rule 4
IF K
THEN F
Rule 5
IF G
THEN H
Rule 6
IF I
THEN J
The Forward
Reasoning
Process
T
Rule 7
F
IF B OR D
THEN K
T
Rule 8
IF E
THEN L
Rule 10
IF K AND
L THEN N
Rule 12
IF N OR O
THEN P
Legend:
Rule 9
IF
IF(F
(FAND
ANDH)
H)
OR
ORJJ
THEN
THENMM
T
Rule 11
IF M
THEN O
First pass
T
Second pass
Third pass
F
13-33
Step 4
Rule 1
IF A THEN
B
T
Rule 2
Step 3
Rule 7
IF B OR D
THEN K
IF C
THEN D
Step 5
Rule 3
IF N OR O
THEN P
Rule 8
IF M
THEN E
IF E
THEN L
Rule 11
Rule 9
13-35
Step 1
Rule 12
IF (F AND H)
OR J
THEN M
IF M
M
IF
THEN O
THEN
O
Legend:
Problems to
be solved
If N Or O
Then P T
Rule 4
If K
Then F
Rule 5
Step 9
If G
Then H
Rule 6
If I
Then J
Rule 12
Step 7
Step 6
IF (F And H)
Or J
Then M T
If M
Then O
Rule 9
Rule 11
Legend:
Problems to
be solved
13-36
13-37
Development Engine
Programming languages
Lisp
Prolog
For managers
Consider more alternatives
Apply high level of logic
Have more time to evaluate decision rules
Consistent logic
For the firm
Better performance from management
team
Retain firms knowledge resource
13-39
13-40
Keys to Successful ES
Development
Neural Networks
13-43
Axonal Paths
(output)
Synapse
Axon
Dendrites
(input)
13-44
Evolution of Artificial
Neural Systems (ANS)
McCulloch-Pitts mathematical
neuron function (late 1930s) was the
starting point
Hebbs learning law (early 1940s)
Neurocomputers
Marvin Minskys Snark (early 1950s)
Rosenblatts Perceptron (mid 1950s)
13-45
Current Methodology
13-46
w1
y2
w2
y3
w3
wn-1
yn-1
13-47
OUT1
OUTn
The Multi-Layer
Perceptron
Input
Layer
Y1
Yn2
OutputL
ayer
IN1
INn
13-48
Knowledge-based Systems
in Perspective