This paper proposes a machine learning based ship speed over ground prediction model, driven by the eXtreme Gradient Boosting (XGBoost) algorithm. The dataset is acquired from a world-sailing chemical tanker with five years of full-scale measurements. The model is trained using encountered metocean environments and ship operation profiles in two scenarios, i.e., through RPM or propulsion power. This model is further combined with the particle swarm optimization (PSO) algorithm to integrate a sailing time control method. It optimizes constant RPM or power operation strategy to meet the requirements of fixed ETA.
The uncertainty over estimated time of arrival (ETA), i.e., the expected arrival date and time of a shipment to a planned destination, is one of the main problems in the decision-making stage within the entire maritime transport chain (Valcic et al., 2011, Wang et al., 2021). This uncertainty reduces the reliability of the schedule, resulting in increased delays and decreased productivity for inland transportation (Rob et al., 2015). The ship arrival delays add to the cost of vessel operating and supply chain management, and efficient logistic plan cannot be formulated (Vernimmen et al., 2007). Thus, it is important to accurately predict the required sailing time, and the ship speed over ground (SOG) is the most essential factor in determining the ETA (Wang et al., 2020). The inevitable ship speed decrease due to various weather conditions always leads ship operators to frequently revise ETA 24 hours before arrival (Fancello et al., 2011). A reliable ship speed prediction is becoming more essential in improving marine traffic control, fleet management, cargo handling operations (Prpic-Orsic et al., 2016).
A ship's speed reduction can be divided into voluntary or involuntary (Faltinsen et al., 1980). The voluntary speed reduction avoids slamming, propeller racing, and extreme ship motions by intentionally adjusting ship engine power or RPM. The added resistance causes the involuntary speed loss due to wind, waves, and propulsion efficiency loss in harsh weather conditions (Chuang and Steen, 2013). The prediction of a ship's speed requires comprehensive knowledge about resistance, propulsion, machinery, and automatic control. Since modern ships’ complexity and exposure to external factors such as wind, waves, and currents, it is not easy to accurately estimate ship actual sailing speed. And there is always a discrepancy between the theoretical calculation and real measurements. Sometimes this difference can be very significant.