Forecasting road transport energy consumption using artificial neural networks

The focus of this paper is to understand the relationships between road transport sector fuel consumptions and input data. Different kind of input variables have been taken into account: socio-economic data input (as population, industrial production index, fuel price, car registration, income, gross domestic product), and weather input (average temperature, average rain). A comparison among statistical methods (regression models) and artificial neural networks has been carried out, in order to evaluate the accuracy of various models, using statistics indexes (the root mean squared error, the mean absolute error and the correlation coefficient) as forecasting performance measures.