Wave forecasts in the proximity of a harbour area based on artificial neural networks

Nowadays maritime transportation has experienced an extremely rapid growth. Indeed, it drives approximately 90% of the global trade and, in the last decade, it registered a continuous growth, about 3% per year. This involves the necessity for improving different aspects related to navigation and, in particular, the navigation safety which is greatly influenced by whether conditions. In this regard, the development of a forecast system of wave characteristics could help the vessel movement in proximity to the harbour entrance or within harbour basin. Wave characteristics are usually estimated using numerical models which generally require high computational costs. The present study reports on the implementation of a forecast methodology based on an Artificial Neural Network (ANN) which aims to provide a reliable response as well as the numerical model but with a significant reduction of the computational time.