Voyage performance evaluation based on a digital twin model

Abstract

Using digital twins in voyage performance evaluation is becoming critical for ocean vessels to reduce GHG emissions. A novel GBM approach is proposed in this paper to establish a digital twin model for voyage performance prediction. The weather hindcast data are introduced to enrich noon reports (NR) and automatic identification system (AIS) datasets, which are split into training and validation sets to develop GBM. The NR and AIS datasets collected from a 57000DWT bulk carrier are used to demonstrate the fidelity and capability of the proposed GBM. The voyage performance prediction from the GBM shows better accuracy than those from pure WBM or pure BBMs. An arrival time forecast and a weather routing showcase are also presented to demonstrate the application effects of GBM. The proposed GBM provides a satisfying prediction of ship speed and fuel consumption without mandatory sensor-collected data, thus applicable for a varity of vessels. In those cases where more sensors are available onboard, the proposed approach can incorporate sensor data to improve the model accuracy further.