Decision trees

Decision trees are part of the evaluation and decision-making instruments and serve in particular for the graphic clarification of multi-level decision problems. Decision trees can be divided into two groups:

⦁ deterministic decision trees, which are characterized by the fact that a multi-level decision problem is broken down into partial activities, to which reliable (partial) results can be assigned;

⦁ stochastic decision trees in which there is the possibility that the results of the individual sub-activities can still be influenced by different environmental developments that cannot be reliably foreseen at the present time.
The end result of a decision problem arises only from the superposition of the partial decisions to be made in the chronological sequence and the development of the environmental constellations. Decision trees are constructed dynamically and determine sequential decision points, some of which cannot be influenced, by resolving a complex decision problem into successive, sometimes alternative, partial decisions. With the prerequisite of successive and alternative action and resource planning, resource planning can, however, only determine the target achievement value of the successive alternative branches when the planning has taken place up to the last period relevant for the entire complex of measures.

The successive derivation enables decision trees to generate planning results that are available simultaneously. The analysis of multi-period alternative branches on the basis of statistical probabilities or expected values also implies that decisions are made at the latest possible point in time, which does not delay the implementation of the overall planning and to the extent that flexibility in terms of content and timing is characteristic of flexible planning. It should be noted, however, that adjustment decisions can only be taken into account in the alternative follow-up planning, since with the decision that has already been made, a general alternative branch was selected according to the highest degree of target achievement, and new information can only be related to follow-up planning.

Decision trees are similar in their formal structure to the relevance trees, but differ in their content. While relevance trees serve a multi-level analysis of criteria and alternative effects, decision trees are used to make sequential decisions about alternative effects in the respective decision context. With the use of decision trees, the planning and decision problem is structured at the same time (open system simulation).

Was the explanation to "Decision trees"Helpful? Rate now:

More explanations about controlling