ABSTRACT
The main objective of this work was to evaluate the consequences of the use of yield and growth estimates from different prediction yield and growth models in a Linear Programming model applied to forest regulation. Thus, using data of continuous forest inventory, the yield estimation in future volume was obtained using a yield model based on age, another one based on age and site index, and a third model that used age and site index besides density, which was represented by the basal area per hectare. Also a model based on data of continuous forest inventory which uses data of volume on a period of time to make linear projections of yield for the next period was tested. Next, a simplified forest regulation problem was proposed and solved by the model I by means of Linear Programming, using data from the four volume prediction models. At the end, the conclusion was: a) that models of forest regulation, supplied with estimates from different yield models, when solved by Linear Programming, result in different ways of treating the forests; b) that the matrix of technological coefficients, supplied with different yield data and for a same objective function affects in a significant way the decision making process.
Key words:
yield and growth models; management model; optimization