LAUSR

This study presents the application of Artificial Neural Network (ANN) techniques to estimate the total energy use of broiler farms. Chicken meat is shown as one of the important parameters in the modeling of energy use efficiency of broiler farms. However, the measurement of this extremely important parameter is difficult and takes a long time to obtain the desired results. In order to overcome such difficulties, scientists have tried to develop alternative methods. The farm-scale data used in the study was obtained from 30 broiler farms in Mersin (Turkey) province in 2018. In the application of ANN model, consumed feeds, electricity, fuel, water, broiler farms, chicks, human labor and machinery parameters used in the farm are used as input; broiler poultry meat and fertilizer parameter are used as output. In addition, the total energy equivalent estimates of chicken meat were made using various input combinations to investigate the best results model. The highest coefficient of determination (R) (0.936) and the lowest root mean square error (RMSE) and the mean absolute error (MAE) values were found to be 0.232 and 0.019, respectively. The results showed that the ANN model is a very promising approach for the estimation of total energy equivalent of chicken meat in broiler farms.