The forest sector is one of the most important pillars of the Brazilian economy due to its high wood productivity. Several studies are being carried out seeking to develop a computational method capable of estimating production efficiently in order to reduce production costs. The objective of the study was to develop machine learning models capable of estimating present and future eucalyptus production with high precision, evaluating relevant supervised learning models, like neural networks and support vector machine (SVM), in relation to the Clutter model, widely adopted by the forestry industries. A case study conducted on real data obtained from a continuous forest inventory showed that SVM is efficient to estimate growth and production of eucalyptus.