Track: Operations Research
Abstract
This paper proposes an approach to predict the operational performance of a port terminal using the combination of Exponential Smoothing, ARIMA and Artificial Neural Networks forecasting models. The empirical research was performed using data from the unloading operation of iron ore from the Carajás mine (Brazil), which is the largest open pit iron mine in the world. The results showed that the proposed combined forecasting approach generates better predictions than the univariates models when compared to other possible methods. The proposed approach can be applied to other complex decision problems that need to predict operational performance with high accuracy in the industrial and logistic context.