MAROFF KPN: Digital Twins for Vessel Life Cycle Service (TwinShip) (2018-2021)

The project is supported by Knowledge-building Project for Industry (KPN)

Project partners

  • NTNU in Ålesund
  • Rolls-Royce Marine AS
  • DNV GL
  • SINTEF Ålesund

Project Description

There is a strong demand for innovation and efficiency within marine system design, operations, and life cycle service. Norwegian maritime industrial cluster is a world leader in developing complex, customized ships and offshore vessels to the global market, particularly for demanding operations, where safety and environment are focused areas.

Nowadays, modern marine vessels operate increasingly autonomous through strongly interacting subsystems [2]. These systems are dedicated to a specific, primary objective of the vessel or may be part of the general essential ship operations. Between sub-systems, they exchange data and make coordinated operational decisions, ideally without any user interaction. Designing, operating and life cycle service supporting such vessels is a complex and intricated engineering task requiring an efficient development approach to consider the mutual interaction between subsystems and the inherent multi-disciplinarily. Scalable simulation technologies should take the lead in this process. Furthermore, the work flow in maritime industry does not stop after vessel delivery. Through system updating or due to life cycle maintenance, subsystems can change. The overall behaviour of the entire vessel still needs to be efficient and value robust. To make sure of that, product design and product use need to be coupled already during early stages of design, which requires traceability through a performance data management system that spans the entire vessel lifecycle.

The recent years have seen an increasing interest in developing and employing digital twins, big data and cloud computing for maritime industrial system design, ship intelligence, and operational service. Digitalization has become a key aspect of making the maritime industries more innovative, efficient and fit for future operations. Increased use of advanced tools for designing and evaluating system performance, safety and structural integrity are generating a range of digital models of a vessel and its equipment. In the operational phase, cheaper sensors and increased connectivity together with increasing data storage and computational power enable for new ways of managing a vessel’s safety and performance.

The goal of this research is to develop digital twins of maritime systems and operations, which is an open virtual simulator as the next generation of marine industrial infrastructure not only for overall system design, allowing configuration of systems and verification of operational performance, but also more focusing to provide early warning, life cycle service support, and system behaviour prediction (Figure 1).

centered image
Figure 1. The development of digital twins system for marine industry

Today’s maritime engineering systems are operating in highly dynamic environments. The challenge is to develop a concept leveraging on the different levels of system specific services already provided by manufacturers, where safety and efficient performance of complex integrated systems can be managed from the early stages of a new build vessel project and throughout the vessel’s life-cycle. In order to be able to respond fast to unexpected events and anomalies, future systems need to become more autonomous. Advanced marine systems should be able to decide between different actions, adapt to dynamic environments and execute high-level task specifications without explicitly being programmed. To meet requirements, they will need access to very realistic models of the current state of the process, and in addition, their own behavior in interaction with the environment in the real world. From a simulation point of view, digital twins are the next wave in modeling, simulation and optimization technology. It will be a significant scientific and operational achievement for the maritime industry, keeping Norway on the technological lead.

The marine digital twin tracks information on all parameters to define how each individual module and sub modules behaves over its entire useful life, including the initial design and further refinement, manufacturing related deviations, modifications, uncertainties, updates as well as sensor data from on-board systems, maintenance history and all available historical and fleet data obtained using data mining. Using digital twins for marine engineering could bring the following advantages.

Working packages

  • WP1 – Develop an open digital twins platform for marine design, operation, and maintenance.
  • WP2 – Development tools for early warning, prediction, and optimization based on digital twins for maritime industry.
  • WP3 – Demonstrators – Subsystem and operational verification process.


  • KPN Twinship Kick-Off Meeting, 20 June 2018, NMK II.
  • KPN Twinship WP2 workshop, 13 December 2018, NTNU i Ålesund.


  1. Xu Cheng, Guoyuan Li, Robert Skulstad, Shenyong Chen, Hans Petter Hildre and Houxiang Zhang: Modeling and analysis of motion data from dynamically positioned vessels for sea state estimation, IEEE International Conference on Robotics and Automation (ICRA), Montreal, QC, Canada, pp. 6644-6650, 20–24 May, 2019.
  2. Xu Cheng, Guoyuan Li, Robert Skulstad, Shengyong Chen, Hans Petter Hildre and Houxiang Zhang: Data-driven uncertainty and sensitivity analysis for ship motion modeling in offshore operations, Ocean Engineering, vol. 179, pp. 261-272, 2019, DOI: 10.1016/j.oceaneng.2019.03.014.
  3. Xu Cheng, Guoyuan Li, Robert Skulstad, Shengyong Chen, Hans Petter Hildre and Houxiang Zhang: A neural network based sensitivity analysis approach for data-driven modeling of ship motion, IEEE Journal of Oceanic Engineering, 2018, DOI: 10.1109/JOE.2018.2882276.
  4. Robert Skulstad, Guoyuan Li, Houxiang Zhang, and Thor I. Fossen. “A Neural Network Approach to Control Allocation of Ships for Dynamic Positioning”. IFAC-PapersOnLine 51, no. 29, pp. 128-133, 2018.
  5. Robert Skulstad, Guoyuan Li, Thor I. Fossen, Bjørnar Vik and Houxiang Zhang: Dead reckoning of dynamically positioned ships: using an efficient recurrent neural network, IEEE Robotics & Automation Magazine, vol. 26, no. 3, pp. 39-51, 2019, DOI: 10.1109/MRA.2019.2918125.
  6. Guoyuan Li, Hans Petter Hildre and Houxiang Zhang: Toward time-optimal trajectory planning for autonomous ship maneuvering in close-range encounters, IEEE Journal of Oceanic Engineering, 2019, DOI: 10.1109/JOE.2019.2926822.
  7. André Listou Ellefsen, Emil Bjørlykhaug, Vilmar Æsøy, Sergey Ushakov, and Houxiang Zhang. “Remaining Useful Life Predictions for Turbofan Engine Degradation Using Semi-Supervised Deep Architecture”. Reliability Engineering & System Safety, accepted, 2018.