Improve poultry farm efficiency in Iran: using combination neural networks, decision trees, and data envelopment analysis (DEA)

Since, poultry meat farming sub-sector has high potential for enhancing the agriculture industry in comparison to other sub-sectors. Therefore, evaluation of decision making units (DMUs) of poultry in provinces and hence improving them is important task to the whole agriculture. Besides, there exist several proposed approaches to resolve this problem. However, a different methodology is proposed due to its powerful discriminatory performance, in this research. For this purpose, combination of data envelop analysis (DEA) and requisite data mining techniques same as artificial neural network (ANN) and decision tree (DT) are employed in order to enhance the power of predicting the DMUs evaluation performance because of their well-known efficiency, and thereby to present precise decision rules for improving their efficiency. To illustrate the proposed model, all poultry companies in Iran were taken into account. However, in this case there is a small dataset and because the large dataset is necessary to collect data as well as to apply data mining methodology, so, we employed k-fold cross validation method to validate our model. Consequently, applied model is supposed to predict efficiency of DMUs and thereby to present decision rules in order to improve the efficiency precisely and accurately according to used optimizing techniques.