Decision Making &

Decision Support

MIS 320
Kraig Pencil
Winter 2011
PPT Slides by Dr. Craig Tyran and Kraig Pencil
Game Plan
Decisions and IS support
Decision making process
Decision support systems (DSS)
Expert Systems

PPT Slides by Dr. Craig Tyran and Kraig Pencil
A. Decisions and IS Support
1. Business people encounter many types of decisions
Follow a decision making process
Variety of information systems to support decision making
From Haag, et al, MIS for the Information Age, 3
rd
Edition, 2002.
PPT Slides by Dr. Craig Tyran and Kraig Pencil
B. Decision Making Process
1. Simons decision
making model
Simple, yet enduring
Decision process
modeled as a flow of
events
Can proceed in linear or
iterative fashion
2. Information systems can
support each phase of
process

This gentleman won a Nobel Prize!
PPT Slides by Dr. Craig Tyran and Kraig Pencil
Business Scenario
Scenario: Orcas Kayaks Inc. is a company that makes
sea kayaks. A new Operations Manager has been hired
to promote smooth and efficient manufacturing
operations.
Note: The previous Ops Mgr got fired
Frequent problems with factory equipment
Much factory downtime
Poor decision making

Desired:
Better decision making process
Support for mgmt decision making
PPT Slides by Dr. Craig Tyran and Kraig Pencil
B. Decision Making Process (cont.)
3a. Intelligence phase
Gather data that may be used for
intelligence purposes
Does there seem to be a
problem(s) or opportunity(s)?
Define the problem or opportunity

Operations Mgr scenario:
Gather data
Review of production log reveals
significant equipment downtime
Problem: Poor maintenance? (Or
could it be something else?)

Can IS help? Which type(s) of
IS could be useful for this phase?
PPT Slides by Dr. Craig Tyran and Kraig Pencil
B. Decision Making Process (cont.)
3b. Design phase
Identify key variables
Create model to aid decision making
Validate model
Establish criteria to be used to make a
choice
Identify alternative solutions

PPT Slides by Dr. Craig Tyran and Kraig Pencil
B. Decision Making Process (cont.)
3b. Design phase




Operations Mgt scenario:
Variable: e.g.,
current maintenance schedule,
age of equipment,
cost of maintenance,
cost of lost productivity ...
Model(s): e.g.,
statistical regression,
cost-benefit forecast model
Establish criteria: e.g., $ total 5-year cost
Alternatives: e.g.,
repair as needed
purchase new equipment
change maintenance schedule

Can IS help? How?
PPT Slides by Dr. Craig Tyran and Kraig Pencil
Example: Statistical Model Factory
Equipment Maintenance Cost vs. Age
Should we be
concerned
about model?
The model
sounds good
Age
C
o
s
t
s

PPT Slides by Dr. Craig Tyran and Kraig Pencil
B. Decision Making Process (cont.)
3c. Choice phase
Evaluate potential solutions using
model(s) developed earlier
e.g., What if analysis,
Sensitivity analysis
Use criteria to choose the
preferred solution

Operations Mgr scenario:
Tasks: See above
Make it easy to explore
different scenarios of interest
Convey useful information

Can IS help? How?
PPT Slides by Dr. Craig Tyran and Kraig Pencil
B. Decision Making Process (cont.)
3d. Implementation phase
Implement the decision
Monitor
Make adjustments

Operations Mgr scenario:
Tasks: See above

How could an IS provide
support for the above?
PPT Slides by Dr. Craig Tyran and Kraig Pencil
C. Decision Support Systems
1. Used for decisions that are
Complex, Messy
Non routine, Non structured
2. Include models (financial, prediction, etc.)
e.g., Allow decision maker to manipulate inputs,
Support what if analysis
3. Particularly helpful for ______ and/or _______ phases of
decision process
Example: Cow culling DSS to support cattle management:
http://ag.arizona.edu/AREC/cull/culling.html
choice design
PPT Slides by Dr. Craig Tyran and Kraig Pencil
DSS for Cattle Management
DSS results in this
case are displayed
using a graphical
output
Flexible: can evaluate
multiple cases or
scenarios.
Does the DSS make
the decision???
No, the user may need
some expertise to
interpret the results or
to consider other
factors.
PPT Slides by Dr. Craig Tyran and Kraig Pencil
C. Decision Support Systems (cont.)
4. Three primary components of DSS include (see figure)
User interface
Model management
Data management


PPT Slides by Dr. Craig Tyran and Kraig Pencil
D. DSS Components

PPT Slides by Dr. Craig Tyran and Kraig Pencil
C. Decision Support Systems (cont.)
5. Excel-based DSS Example: (next slide)
Break-even analysis for a manufacturing business
Decision: What sales price will lead to breakeven after 1,000
units sold?
DSS components and features
Interface
Graphical output
Advanced Excel feature: User input
form
Model management
Break-even model
Data management
Assumption data table (very simple)
Easy access to the workbook through
the WWW

PPT Slides by Dr. Craig Tyran and Kraig Pencil
Break-Even Analysis Model
PPT Slides by Dr. Craig Tyran and Kraig Pencil
D. Introduction to Expert Systems (ES)
1. Expert systems
Are Knowledge-based systems
Replicate the thought processes of human
experts
Follow a structured set of rules
My computer wont turn on
Did you press the power button?
If no, press it and start again.
If yes, is it plugged in?
If no, plug it in and try again.
If yes, is the monitor on? and so on.
Image: www.tubecad.com/2006/10/05/Genius.png
PPT Slides by Dr. Craig Tyran and Kraig Pencil
D. Introduction to Expert Systems (ES)
Figure: http://www.generation5.org/content/2004/bdt-implementation.asp
Expert Systems
tend to use
branching
logic, like this
one.
bare
PPT Slides by Dr. Craig Tyran and Kraig Pencil
D. Introduction to Expert Systems (ES)
1. Expert systems (continued)
An expert system can replace the expert
Identify the whale species: http://www.aiinc.ca/demos/whale.shtml




An expert system can support the expert
Paper machine diagnostic system: http://www.exsys.com/demomain.html

The Cs: Create and Convey
PPT Slides by Dr. Craig Tyran and Kraig Pencil

Notice that the recommendations
still leave the human expert some
options.
PPT Slides by Dr. Craig Tyran and Kraig Pencil
D. Introduction to Expert Systems (cont.)

2. How is an ES different from a DSS?
DSS
Allows decision maker a platform for exploring and evaluating the options
ES
Delivers advice/recommendations based on answers to a set of questions
Where does the knowledge reside for DSS? For ES?

3. Business examples of ES?
Help desk / Customer support center
Other ???

4. Would it be easier to develop a set of expert rules for
a) a narrow, well-defined field of expertise?
b) a wide-ranging, poorly-defined field of expertise?
What are the implications for ES applications?
PPT Slides by Dr. Craig Tyran and Kraig Pencil
E. Knowledge Base Systems
Knowledge Base*
Use indexes and search technology (key word searches) to catalog
and retrieve relevant information recorded by experts.
Is also known as a Human-readable knowledge base. The
Microsoft Knowledge Base, for example, is a vast collection of
brief articles, papers, manuals, etc. on technical topics for
Microsoft products. Example: http://support.microsoft.com/kb/931836
Some organizations have tried to capture solutions to problems that
their employees have encountered, so that employees in the future
wont need to waste time reinventing the solution

* Knowledge Base has multiple meanings. In an Expert System, a knowledge base is a set
of facts (situations) and rules stored in specialized databases. The knowledge base for an
expert system is complex and is designed to be read by the ES, not by people.