Application of Artificial Neural Network in the Modeling of Skidding Time Prediction

Correct skidding time prediction and having precise information about the efficiency of skidding system rescues in a high quality of harvesting in a variable conditions of woodlands. This paper represents one of the Artificial intelligence methods, that is called Artificial Neural Network (ANN), for creating the predicting time model of wheeled skidder Tiinberjack 450C. The study of components of project has been done by continuous time study method. Effective factors on time of skidding were: skidding distance, skidding slope, winching distance, number of logs, and volume of load. In the present study, the time of 105 skidding cycles were investigated and used as the training data for neural network. The determination coefficient and the root mean square error for the best trained network were, 71% and 0.1778, respectively. The results showed that ANN provides a good accuracy for estimating the time of one cycle skidding time.