Affective factors weight estimation in tree felling time by artificial neural networks

One of the most important functions of forests is timber production. The variation in working conditions and places, the complexity and variety of factors affecting the productivity of workers, and the nonlinear relationships among these factors in timber production make it difficult to solve with traditional methods. Today, artificial neural networks (ANN) designed to explain the biological structure of human brain is one of the most commonly used methods in the examination of system behaviors including complex events. In this paper, the affects of factors interacting the felling process on work time by using artificial neural networks (ANN) were identified. The measurement and observation were done in the mountainous forests of The Eastern Black Sea region, Turkey. The felling time, amounts, and the affecting factor measurements were mostly used in The Eastern spruce and beech forests. The data collected were evaluated statistically and the regression equation counting the time according to affective factors were formed. At the end of the study, it was found out that ANN models are more realistic and reliable for different land and working conditions when compared to regression equation based on direct relations in calculating the work time. The ANN models can be used for similar conditions on the planning of felling time, the control of applications and the determination of unit of price for felling workers.