Decision Tree defined:
There are many technical definitions of a decision tree is, you may also use decision trees for many different types of processes or in different industries. The simplest definition of a decision tree is that it is an analysis diagram, which can help aid decision makers, when deciding between different options, by projecting possible outcomes. The decision tree, gives the decision maker an overview of the multiple stages by that will follow each possible decision. Each branch shows the probability of the outcome.
A technical definition taken from our Foundations of Operations Management textbook is “A schematic model of alternatives available to the decision maker, along with their possible consequences.” Decision trees can be used when making a wide variety of choices, from the simplest decisions to very complex ones regarding:
product planning and process management.
How to use a Decision Tree:
A decision tree is used from right to left, and calculations are made to show the benefits or potential problem areas. An event node is represented with a square; a decision node is represented with a circle. Visually a typical decision tree looks like a sideways tree (it may also look like the decision tree at the top of the page); branches are represented by the choices that you have to make, accompanied with the probability factor.
The probability of each option is shown above each branch. The two or more branches that sprout from the same node must all add up to one (100%); for example, if one node has two options, and one of the branches probabilities is 40 percent, then the other branch probability must be 60 percent. The probability percent is based on historical facts; unfortunately there is not always historical data to follow, so a decision tree is only as good as the research and educated guesses that supports it.
Each branch is also represented with the payoff; this is usually expressed as the present value of net costs or profits. To get a dollar figure you start with the percentage of the branch and multiply it by the dollar figure. Your choice(s) will based on the getting the highest payoff from each decision. This will become clear once you view the example based on the Fit Pit.
When creating a decision tree you need to identify the major decisions denoted by a square and the major uncertainties, denoted by a circle. The decision tree analysis can lead to paths the decision maker may not have considered in the beginning.
.GIF)
Decision Tree Defined:
The process of calculating the value of each node is known as “folding back” the tree. Event nodes are represented by things that the decision maker has no control over. To make your calculations you multiply the values of the event node by the probabilities. The value of the node depends on the nodes on the right.
Advantages of using a decision tree:
There are a couple of advantages of using a decision tree a few of them are:
(click on link to see more advantages of using a decision tree)
1. It is very easy to understand and interpret for any reader.
2. Small details that may have been missed are taken into consideration.
3. Saves time in the long run; once it is laid out your path to success is easy to follow.
4. They work well with multi stage/phase decisions.
There are lots of different types software available to help you design a decision trees; there are links provided on the right hand side of the blog; you may also use an excel add in for decision trees, or research further applications that can help you design your decision tree.
Example of a Decision Tree:
For this article BCIT’s Fit Pit the example that is used. The Fit Pit is the gym/recreational center on campus. Through research it has been found that they have a declining rate of use for their facilities; dropping approximately 20 percent each semester.
The Fit Pit can use a decision tree to help support or help make the decisions needed that may help them with the declining rate of usage. A process that that has been identified as something that needs to changes is that the Fit Pit lacks any process to track the coming and going of students and teachers. This means that they do not know who is using the gym, which programs’ are successes and which are not.
By purchasing or creating a system that could track who is using the gym they would be able to use this information, identify their market and as well they would be able to see what the peak time of use are, decide which classes are not needed and which are full and could replace the empty ones. With the ability to identify their market they may be able to increase the use of the gym and recreational center; make better use of the classes and wasted space. There are many possible options when deciding on getting a tracking system or if it is needed at all. A decision tree will eliminate doubt and allow BCIT to make the right decision.
To view the decision tree model please click on the link on the top right hand side of the page.