What Is Decision Tree Analysis?

The decision tree analysis method refers to the analysis of each decision or event (that is, the natural state), which leads to two or more events and different results, and draws a branch of this decision or event into a graph. The branch of a tree is called decision tree analysis.

Decision tree analysis

Right!

Decision tree analysis refers to the analysis of each decision or event (that is, the natural state), which leads to two or more events and different results, and draws a branch of this decision or event into a graph, which is similar to The branch of a tree is called decision tree analysis.

Chinese name

Decision tree analysis

Foreign name

decision tree

Types of

Analytical method

Features

Top-down

Generally generated from the top down. Each

advantage:

1) can generate understandable rules;

2) The amount of calculation is relatively small;

3) Can handle continuous and category fields;

4) The decision tree can clearly show which fields are important.

Disadvantages:

1) It is difficult to predict continuous fields;

2) For chronological data, a lot of pre-

Decision trees are a quick and easy way to determine production capacity options. Not only can decision trees help people understand problems, they can also help people solve problems. A decision tree is a method that lists the steps involved in solving a problem and the conditions and results of each step. Many specialized software packages that have appeared in recent years can be used to build and analyze decision trees. With these specialized software packages, it becomes easier to solve problems.

The decision tree consists of decision nodes, branch nodes between opportunity nodes and nodes. Generally, people use boxes to represent decision nodes, circles to represent opportunity nodes, branch lines from decision nodes indicate choices that decision makers can make, and branch lines from opportunity nodes to represent opportunity nodes. Probability of the event shown.

When using a decision tree to solve a problem, you should start from the end of the decision tree, step by step, and advance to the beginning of the decision tree. As you move forward, the expected value of an event should be calculated at each stage. It is important to note that if the planning period of the problem handled by the decision tree is long, the time value of the funds should be considered in the calculation.

After the calculation is completed, the decision tree is pruned, and all branches other than the highest expected value are deleted at each decision node, and the last step is advanced to the first decision node. At this time, the most problematic node is found. Best plan.

Let's take the Southern Hospital Supply Company as an example to see how to use the decision tree to make an appropriate production capacity plan.

Southern Hospital Supply Company is a company that manufactures overalls for medical staff. The company is considering expanding production capacity. It can have the following options: 1. Do nothing; 2. Build a small plant; 3. Build a medium-sized plant; 4. Build a large plant. The newly added equipment will produce a new type of gown, and the potential or market of this product is still unknown. If you build a large factory and the market is good, you can achieve a profit of $ 100,000. A bad market will result in a loss of $ 90,000. However, if the market is better, building a medium-sized plant will get $ 60,000, small-scale plants will get $ 40,000, and if the market is not good, building a medium-scale plant will lose $ 10,000, and small plants will lose $ 5,000. Of course, there is another option to do nothing. Recent market research shows that the probability of a good market is 0.4, which means that the probability of a bad market is 0.6. See the figure below:

Based on these data, the options that will produce the largest expected monetary value (EMV) can be found.

EMV (Jianda Factory) = (0.4) * ($ 100,000) + (0.6) * (-$ 90,000) =-$ 14,000 EMV (Medium-sized Factory) = (0.4) * ($ 600,000)) + (0.6) * (-$ 10,000) = + $ 18,000 EMV (build a small factory) = (0.4) * ($ 40,000) + (0.6) * (-$ 5,000) = + $ 13,000 EMV (do not build a factory) = $ 0 According to the EMV standard, Southern Corporation should build a medium-sized factory.