This document discusses modeling and analysis techniques used in decision support systems (DSS). It covers various categories of DSS models including optimization, simulation, and predictive models. It also describes static and dynamic analysis, decision making under certainty, risk, and uncertainty. Different modeling approaches like mathematical modeling, simulation, and heuristics are explained.
2. DSS modeling – Issues
• DSS – can be composed of multiple models
• Modeling Issues -
• Identification of problems and environment
analysis
• Variable identification
• Forecasting (predictive analysis)
3. DSS modeling – Categories
• Optimisation of problems with few
alternatives
• Optimisation via algorithm
• Optimisation via analytical formula
• Simulation
• Heuristics
• Predictive models
• Other Models
5. DSS modeling – Trends
• Model libraries and solution techniques
• Using web tools – perform modeling,
optimisation, simulation etc
• Multidimensional analysis
• Model for model analysis
6. Classification of DSS Models
Static Analysis:
• Static model takes a single snapshot of
situation
• Everything occurs in a single interval.
• E.g. Make or buy decision
• Stability of the relevant data is assumed.
7. Dynamic Analysis:
• Represents scenarios that change over time.
• E.g. 5-year profit and loss projection in which
the input data, such as costs, prices, and
quantities, change from year to year.
• Time dependent
• Important because they use, represent, or
generate trends and patterns over time.
• Shows average per period, moving averages
and comparative analysis.
8. Certainty, uncertainty, and risk
Decision situations are often classified on the
basis of what the decision maker believes about
the forecasted results. The categories are:
• Certainty
• Risk
• Uncertainty
9. Decision Making Under Certainty
• Complete knowledge is available
• Decision maker knows the outcome of each
course of action
• Situation involve is often with structured
problems with short time horizons
• Certain models are relatively easy to develop
and solve and they can yield optimal
solutions.
10. Decision making under uncertainty
• Several outcomes for each course of action.
• Decision maker does not know, or cannot
estimate the possible outcomes.
• More difficult because of insufficient
information.
• Involves assessment of the decision maker’s
attitude towards risk.
11. Decision making under risk
(Risk analysis)
• Decision maker must consider several possible
outcomes for each alternative.
• The decision maker can assess the degree of
risk associated with each alternative.
• Risk analysis can be performed by calculating
the expected value for each alternative and
selecting the one with best expected value.
12. Decision analysis with decision tables
and decision trees
Decision Table:
• Organize information and knowledge in
systematic tabular manner
13. Decision Trees:
• Alternative representation of the decision
table
• Shows the relationship of the problem
graphically and handle complex situations
• Can be cumbersome if there are many
alternatives or static nature.
• TreeAge Pro and Precision Tree: Powerful and
sophisticated decision tree analysis systems
14. Structure of mathematical models for
decision support
Components of decision
support mathematical
models:
• Result Variables
• Decision Variables
• Uncontrollable variables
• Intermediate result
variables
15. • Result Variables: reflect the level of effectiveness
of a system
• Decision Variables: describes alternative course
of action.
• Uncontrollable Variables: Some factors that
affect the result variables but not under the
control of decision maker.
• Intermediate result Variables: reflect
intermediate outcomes in mathematical models.
17. Sensitivity Analysis
• Attempts to assess the impact of a change in input data
on proposed solution.
• Important because it allows flexibility and adaptation
to changing conditions
• Provides a better understanding of the model and the
decision making situation
• Used for:
1.Revising models to eliminate too-large sensitivities.
2.Adding details about sensitive variables.
3.Obtainong better estimate of sensitive external
variables.
4.Altering a real-world system to reduce actual
sensitivities.
18. What-If-Analysis
• What will happen to the solution if an input
variables, an assumption, or a parameter
value is changed
• With the appropriate user interface, it is easy
for manager to ask a computer model
different questions and get the answers.
• Common in expert systems.
• User get an opportunity to change their
answers to some question’s.
19. Goal Analysis
• Calculates the values of the inputs necessary
to achieve a desired level of output.
• Represents a backward solution approach
20. Problem solving search methods
The choice phase of problem solving involves a
search for an appropriate course of action.
Search approaches are:
• Analytical Techniques
• Algorithms
• Blind Searching
• Heuristic Searching
21. Simulation
• Is a appearance of reality.
• A technique for conducting experiments with
computer on model of a management system
• Characteristics:
1.Simulation typically imitative.
2.Technique for conducting experiments.
3.Descriptive rather than a normative.
4.Used only when a problem is too complex to be
treated using numerical optimizing techniques.
22. Advantages of simulation
• Theory is fairly straightforward.
• Great time compression
• Descriptive rather than normative.
• Built from the manager’s perspective.
• Built for one particular problem and cannot solve
any other problem.
• A manager can experiment to determine which
decision variables and which part of environment
are really important, and with different
alternatives.
23. • Can handle an extremely wide variety of
problem types, such as inventory and staffing.
• Can include the real complexities of
problems.
• Automatically produce many important
performance measures.
• Relatively easy-to-use simulation packages.
• Often the only DSS modeling method that can
readily handle relatively unstructured
problem.
24. Disadvantages of simulation
• An optimal solution cannot be guaranteed.
• Model construction can be a slow and costly
process.
• Solutions are not transferable to other
problems
• Easy to explain to managers that analytic
methods are overlooked.
• Requires special skills because of the
complexity of the formal solution method.
25. The Methodology of Simulation
Test &
validate the
model
Real world
problem
Define the
problem
Construct
simulation
model
Implement
the result
Design the
simulation
experiments
Conduct the
experiments
Evaluates
the results
26. Simulation type
Probabilistic Simulation:
• One or more of the independent variables
• Follow certain probability distributions namely
1.Discete distribution
2.Continuous distribution
• Conducted with the aid of technique called
Monte Carlo simulation.
27. Time-Dependent Vs Time-Independent
Simulation:
• Time-independent-not important to know the
exact time of event
• Time-dependent-In waiting line problems, it is
important to know the precise time of arrival.
28. Object-Oriented Simulation:
• SIMPROCESS is an object-oriented process
modeling tool that allows user to create a
simulation model by using screen based
object.
• Unified Modeling Language(UML)- Designed
for object-oriented and object based systems
and applications.
• Java based simulations are essentially object
oriented.
29. Visual Simulation:
• Graphical display of computerized results
• Includes animations
• Is one of the most successful development in
computer-human interactions and problem
solving.
30. Quantitative Software Packages
• Are preprogrammed models and optimization systems.
• Serve as building blocks for other quantitative models
• A variety of these are available for inclusion in DSS as
major and minor modeling components.
• Revenue management systems focus on identifying
right product for right customer.
• Airlines have used such systems to determine right
price for each airline seat.
• System also available for retail operations,
entertainment venues, and many other industries.