Sachpazis Costas: Geotechnical Engineering: A student's Perspective Introduction
Decision Tree Analysis
1. Decision Tree Analysis
Prof. Priyanka P. Dange
Civil Engineering Department
Sanjivani College of Engineering,
Kopargaon.
2. Introduction
A decision tree is essentially a graphic diagram.
One of the best ways to analyze a decision is to
use so-called decision trees.
It is a useful tool for management decision
makers.
It is useful in cases where a manager is required
to make a number of sequential decisions.
Decision trees depict in the form of a ‘tree’, the
decision points, change events and probabilities
involved in various courses that might be
undertaken.
3. Why Decision Tree?
A common problem occurs in business
when a new product is introduced. Manager
must decide whether to install expensive
permanent equipment to ensure production
at the lowest possible cost or to undertake
cheaper temporary tooling that will
involve a higher manufacturing cost but
lower capital investments and will result in
lower losses if the product does not sale as
estimated.
4. . Invest(0.6)
Undertake Development Success (0.7)
Research
Not to Invest(0.4)
Development Fails (0.3)
Discard
Discard
Decision Outcome Termination
3
Decision Tree Example
5. Decision Tree Analysis
This diagram is read it from left to right.
The leftmost node in a decision tree is called as a
root node or a decision node and can be
represented by small square( )
The branches emanating to the right from a
decision Node represent the set of decision
alternatives that are available. One and only one
of these alternative can be selected.
The small circles( ) in the tree are called chance
nodes.
6. The number shown in parentheses on each
branch of a chance node is the probability that
the outcome shown on that branch will occur at
the chance node.
The right end of each pass through the tree is
called an endpoint and each endpoint
represents the outcome of following a path
from the root node of the decision tree to that
endpoint.
7. Steps in Decision Tree Analysis
Identifying the problem and
alternatives.
Delineating the decision tree.
Specifying probabilities and monetary
outcomes.
Evaluating various decision
alternatives.