A Semi-Supervised Machine Learning Model to Forecast Movements of Moored Vessels

The good performance of the port activities in terminals is mainly conditioned by dynamic response moored ship system at a berth. An adequate definition highly multivariate processes involved berth crucial for an appropriate characterization operability. availability efficient forecast movements ships essential planning, performance, and safety development operations. In this paper, inference model to predict motions, based on semi-supervised Machine Learning methodology, presented. A comparison with different supervised unsupervised techniques, as well existing Deep Learning-based models predicting has been performed. highest obtained. Additionally, influence infragravity wave parameters introduced predictor variables analyzed compared typical ocean waves, wind, sea level variables. prediction developed validated available dataset measured data from field campaigns Outer Port Punta Langosteira (A Coruña, Spain).