Challenges when creating a cohesive digital twin ship: a data modelling perspective

Abstract

A digital twin is a digital asset that simulates the behaviours of a physical counterpart. Digital twin ship literature identifies that the concept is already being applied to specialised problems, but no clear guide exists for creating broader interdisciplinary digital twins. Relevant dimensions of product data modelling and previous attempts at standardizing ship data elucidate the requirements for effective data modelling in a digital twin context. Such requirements are placed in a broader perspective for digital twin implementation that encompasses challenges and directions for future development of services, networks, and software. Finally, an open standardization for digital twin data is proposed based on lessons extracted from this panorama, proposing its application to a research vessel.

Ship’s Digital Twin—A Review of Modelling Challenges and Applications

“…As described in [1], an SDT concept can be divided into fields of application, and collected data for each application field are processed for later modelling, simulation, and…”

“…As described in [1], an SDT concept can be divided into fields of application, and collected data for each application field are processed for later modelling, simulation, and behavioural analysis for the purpose of control, prediction, and decision-making support. Therefore, defining the purpose of the model might be considered as the initial step of the SDT formulation process.…”

“…SDT uses sensors installed onboard to collect data, such as heading, vessel speed, GPS coordinates, azimuth angles, propulsion power, engine speed of rotation, engine load, vibrations, and structural strains, depending on the purpose. However, sensor logs, or raw data, could not usually be immediately useful for a DT, i.e., it might be necessary to perform data processing to extract physical meaning from it, as noted in [1].…”

“…Generally, applications can vary; however, they could be classified into some basic groups such as condition-based maintenance and predictive maintenance, as in [8,23], decision-making support, as in [12,22,23], condition monitoring, as in [9,16,26], cost reduction, as in [17], testing and simulation, as in [24], testing the integration of various systems onboard, as in [1], training personnel, as in [50], and overall optimisation, as in [10,13,18].…”

“…Although the concept of a digital twin (DT) can be found described in the literature, e.g., as in [1], and although a digital twin as a term is being used increasingly in different fields of technical sciences, the concept remains unclear mostly in the way of differentiating a DT concept from an ordinary computer simulation model. Furthermore, the concept is sometimes related to a cyber-physical system (CPS) as described in [2][3][4][5], a part of the CPS as described in [6], or as a digital model and digital shadow, which can be combined to form a DT as described in [7].…”