The subject of this thesis is management of forest plantations considered as a decision-making process. The approach adopted has made use of some of the ideas and terminology of operations research and of control theory. Special attention has been given to the conceptual basis of the development of decision models in forest management. In addition, available knowledge is briefly analyzed, as well as current problems and possible approaches to their solution.
Precise formulation of the problem, clearly defining the objectives and alternatives of possible action, is of prime importance. Effectiveness measures for management plans can take the form of the present value of the total net income, or of an internal rate of return, depending upon the objectives and characteristics of the firm.
The construction of the model can be approached through separate consideration of various submodels: a model of evaluation, a model of prices and costs, and a model of the forest. In many cases it is true that optimum management of the forest is obtained by optimizing independently the management of the stands of which it is composed. The models for stands may represent various levels of detail. In addition, these models may be deterministic or stochastic, dealing in this case with the development of stands as a Markovian stochastic process,
The results of the application of a management plan can be predicted by means of analog or digital simulation. Besides, for complicated models, simulation may be the only method of determining a near optimum plan. If a mathematical model which is not very complicated is used, dynamic programming is the optimizing technique most adequate.
The estimation of the parameter values of the model can be based on data from permanent plots, and/or temporary plots in which increments have been measured.
The random characteristics of forest development and of the market conditions make that the real values tend to differ more each time from the values predicted by the model, as time elapses. This brings out the problem of determining the most adequate periodicity and characteristics of the updating and recalculating of the management plan. This problem could be approached through stochastic models, including the characteristics of this control as a part of the management plan.
As an appendix, an example of a simple deterministic stand model is included, as well as its simulation in an analog computer.