Prediction of two-wheel drive tractor repair and maintenance costs using artificial neural network in comparing with regression

The highest profit is the ultimate goal of any farming firm. The management of machine replacement (specifically tractor) is one of the most critical factors for achieving this goal. Therefore, accurate prediction of tractor repair and maintenance costs is a must. The objective of this research was to predict tractor repair and maintenance costs using neural network and regression methods and comparing their performance for these predictions. The study was conducted using empirical data on 60 two-wheel drive tractors from Astan Ghodse Razavi agro-industry. The results indicated that both neural network and regression models have high rather accurate prediction of repair cost but the performance of regression model in comparison with neural network model for prediction of oil and fuel costs was very low with much less accuracy.