Application of Artificial Neural Network (ANN) for Modelling the Economic Efficiency of Broiler Production Units

The present study addressed the economic analysis of broiler production units. Therefore, required data was collected from 50 broiler production units using personal questionnaires in the winter, 2013. Cronbach alpha coefficient was estimated for these questionnaires to be 0.81, which indicates the reliability of the questionnaire is acceptable. The results showed that the feedforward neural network with two hidden layers (4 and 17 neurons for the economic model) had the best results and it can be used to estimate the energy ratio with high precision. The optimal model performance was performed using measures such as the coefficient of determination (R2), MSE, MAPE and MAE. Value of R2 was reported for the economic model as 96%.