Decision Trees Essay

The process of decision making is extremely important, especially in the highly competitive business environment.

In such a situation, the effectiveness of the decision making may be the determinant factor that defines the success of the performance of a company. At the same time, the effective decision making process is important not only as the means of the improvement of the competitive position of a company in the market but also as an effective tool to maintain the stability within the organization. It proves beyond a doubt that the stable relationships between people working within a company may exist only in the situation when the decision making process does not cause any harm to some members of the personnel and it does not provoke the resistance to the decision taken from the part of the personnel.

However, nowadays, it is hardly possible to speak about the ideal approach to the decision making, which could guarantee the effectiveness of the decision making process and its successful implementation. Nevertheless, among the variety of strategies and techniques that companies use in the contemporary business environment it is possible to single out decision trees, which provide ample opportunities to increase the effectiveness of the decision making process. In fact, decision trees help consistently to model a decision, to forecast its possible consequences, to choose the most effective strategy of its implementation and calculate decision probabilities to increase its reliability and effectiveness of its implementation.

The role of decision trees in decision modeling and forecasting its consequences

On analyzing the role of decision trees in the decision making process, it should be pointed out that decision trees are widely applied in operations research. Basically decision trees are traditionally defined as a decision support tool that uses model of decisions and their possible consequences, including chance event outcomes, resource costs, and utility (Yuan and Shaw, 1995, p.129). Obviously, decision trees fulfill very important functions because modeling of the decision and forecasting its possible outcomes are the most important components of the process of decision making.

As a rule, it is impossible to take a really effective decision if its practical implementation is not modeled and its outcomes are not taken into consideration. A typical decision tree is a closely related model form which is used as a visual and analytical decision support tool. On the basis of the model developed in the process of decision making, policymakers actually define the possible ways of the development and forecast possible outcomes of the decision. In order to understand the practical implementation of decision trees it is possible to model a decision. For instance, it is possible to presuppose that a company has a choice between the development of a new product and consolidation.

To take an effective decision, the company can use a decision tree, modeling decision and analyzing variants that the company has at the moment (Figures 1-2).

First of all, it is necessary to create a diagram that reflects the proposed decision and represents the options that the company has at the moment. On the one hand, the company can develop a new product, while, on the other hand, it can consolidate. In fact, it is necessary to carefully model each of the options suggested, analyze the implementation of each option and forecast the outcome of the practical realization of either option, through the analysis of the possible market reaction. It is highly recommended to carefully plan each step that each option involves and forecast its possible outcomes. For instance, the introduction of a new product may need thorough or rapid development.

Each of these options will have different outcomes and effects, which may be rated from good to poor. On creating a diagram, the entire model of existing options of taking a decision is created and, what is more, the developer of the model can forecast consequences of either option selected in the process of the decision making.

In such a way, it is possible to speak about the highly effective model, which provides ample opportunities to analyze in details different stages of the implementation of a decision. Also it suggests different variants and, as a rule, a decision tree offers several decisions and the policymaker should just chose the most appropriate one. This may be done on the basis of the analysis of the possible outcomes and possible challenges a company may face in the process of the decision making. Consequently, a decision tree offers an effective model of the decision that helps assess objectively all of the options a company has.

The choice of strategy using decision trees

However, along with the practical aspect of the decision making process, decision trees may also play an important strategic role. To put it more precisely, the modeling of the decision making and analysis of possible options that an organization has while taking a decision, provides ample opportunities to forecast consequences of each option. In such a way, policymakers, responsible for decision making, can have ample opportunities to assess each of the options they have before choosing the most appropriate and effective decision. The choice is basically made on the basis of possible consequences and costs of the decision.

In such a situation, it is obvious that the process of decision making naturally provides a lot of information which clarifies the process of the practical implementation of the decision. What is meant here, using decision trees, it is possible to develop an effective strategy to implement the decision and maximize possible benefits a company can receive in case of the successful implementation of the decision. Returning to the example of the decision making concerning the choice between the introduction of the new product and consolidation, it is possible to estimate that the strategy of the implementation of the decision should be developed on the basis of the diagram and possible effects of decisions. For instance, if the analysis of the effects of the introduction of new products proves that this decision is more likely to bring positive results than the company should focus on the introduction of new products.

Naturally, this means that the company should focus on the introduction of innovations, motivate employees to develop new products, etc. In such a way, the company should focus its strategic development on the implementation of innovations. At the same time, to maximize profits, the company should select whether to develop the new product thoroughly or rapidly. Probably, the latter may be more effective and guarantee the company leadership in the industry that makes the strategy of the introduction of innovations the dominant direction of the development of the company because it is the most effective one. In the similar way, the company may analyze each of the options modeled within a decision tree and, depending on the option selected, the company should choose the most effective strategy that meet either option. Obviously, the strategy will vary consistently depending on the decision taken.

Calculating decision probabilities using decision trees

At the same time, decision trees have not only strategic but also practical value. To put it more precisely, using decision trees it is possible to calculate decision probabilities. On analyzing the example of the company that have to choose between the development of the new product and consolidation and using the data from Figure 2, it is possible to calculate decision probabilities (Figure 3). In this respect, it should be pointed out that, in order to calculate decision probabilities, it is necessary to multiply the value of outcomes by their probability. The total for that node of the tree is the total of these values. Using such calculations it is possible to forecast that the value for new product will be as follows:

0,4 (probability of good outcome) * 1,000,000 (value) = 400,000

0,4 (probability of moderate outcome) * 25,000 (value) = 20,000

0,2 (probability of poor outcome) * 1,000 (value) = 400

The calculating of decision probabilities is very important because it makes the forecasts more reliable and facilitates the process of decision making.

Advantages of decision trees

Obviously, decision trees have a lot of benefits and advantages compared to other tools used in the process of decision making. In this respect, it is necessary to underline that decision trees are very simple to understand and interpret. In fact, it is quite simple to build the model and forecast possible outcomes, while the information about the resources the company possesses and the marketing research provide essential data that could be used for the calculation using decision trees. The simplicity of understanding and interpretation of the model is very important because it facilitates consistently the implementation of the decision selected.

It proves beyond a doubt that a decision can be applied effectively when it is understandable for the personnel and when it is positively accepted by the personnel. The use of decision trees may easily convince the personnel as well as policymakers in the necessity to take one of the possible decisions and the simplicity of decision trees makes them very convincing since advantages and drawbacks of either option are obvious.

Furthermore, it should be said that decision trees have value even with little hard data. In fact, in such a situation it is possible to involve experts and specialists, who can describe the situation in details, assess its probabilities, alternatives and costs, and give their professional preferences for outcomes of the option selected. In addition, it is possible to use decision trees for optimizing an investment portfolio. To put it more precisely, decision trees can demonstrate the effectiveness of investments and facilitate the process of taking decision concerning investments and, what is more, it can contribute to the more objective selection of the most effective decision concerning investments.

Finally, it should be said that decision trees use a white box model (Yuan and Shaw, 1995, p.131). What is meant here is the fact that if a given result is provided by a model, the explanation for the results is easily replicated by simple math.

Weaknesses of decision trees

Nevertheless, in spite of the numerous strengths and advantages of decision trees compared to other tools used in the process of decision making process, this tool has certain drawbacks and weaknesses. In this respect, it is necessary to underline that one of the major drawbacks is the high probability of the subjectivity of the decisions and options developed in the process of the construction of a model. In fact, often it is difficult to take into consideration all possible outcomes and existing options in the process of decision making. As a result, some options may be simply omitted, while potentially they may be stronger compared to those included in the model. However, this drawback may be minimized through the involvement of experts in the development of a decision tree.

Conclusion

Thus, taking into account all above mentioned, it is possible to conclude that the use of decision trees in the process of decision making can be very effective. At any rate, the advantageous of decision trees are obvious and they may be easily implemented in different environment that makes this tool widely spread in the process of decision making.



Leave a Reply