Decision Tree Analysis is an analytic method used to quantitatively measure the possible outcomes as well as the probabilities to predict courses of action. The method uses a finite set of variables applied to nodes based on probability. While relatively easy to communicate results, using the method as a predictive tool could be problematic because each node forces a choice between a finite number of options.
- The process of creating a decision tree gives the user insight on the value of information.
- The technique allows for visually presenting the process and different outcomes of the problem.
- Can have various applications.
- Can be relatively simple to create.
- Can be internally or externally focused.
- Relatively easy to communicate findings.
- Can be descriptive or estimative.
- The technique is limited to using a finite set of variables.
- Can potentially lead to multiple outcomes.
- Multiple risks are hard to assess at the same time as it is a linear process.
- The use of a decision tree relies on the decisionmaker adhering to it.
- Event nodes require an evaluation of outcome estimates, which may force quantification of qualitative estimates.
- Tree may expand rapidly due to the possibilities from just a small number of variables.
- Establish the actors.
- Establish the possible events, and their sequence of importance (based on an evaluation of the actors’ goals and motivations).
- Order the events into the nodes and branches of the decision tree. Every node should have branches leading to it and possible choices following it.
- Assign probabilities to possible courses of actions.
- Evaluate the probability of each possible event occurring based on assigned values and position on the decision tree.
Personal Application of Technique:
The class was broken into three groups to create a decision tree based on the given situation:
“Imagine you only ever do four things on Saturday: go shopping, watch a movie, play tennis or just stay in. What you do depends on three things: how much money you have (rich or poor), the weather (windy, rainy or sunny), and whether or not your parents are visiting.” Our group created a decision tree with each criteria as independent variables allowing for probability of the choice, whereas the other groups built each decision based on answers to previous. The general outcome of our analysis was to stay in because there was an 80% chance of bad weather and a 70% chance that parents will not visit.
Rating: 3 out of 5