Digitalisation is the biggest upheaval the maritime industry has seen since the intermodal freight container revolutionised cargo shipping in the second half of the twentieth century – that’s Rolls-Royce Marine’s view, in any case.
The company is keen to be at the forefront of this revolution with its Ship Intelligence arm and its work with intelligent asset management is a key part of the “digital landscape” that represents its response to the challenge posed, according to Rolls-Royce Marine business development and strategy manager for IAM Marte Ellingsen Tyldum.
Speaking at the company’s Intelligent Analytics Centre in Ålesund, Ms Ellingsen Tyldum explained that the data-enabled ship sits at the heart of a digital landscape, feeding data to four centres – the fleet operations centre, the remote operations centre, the simulation centre and the intelligent analytics centre itself. Data is also key to the other part of the Ship Intelligence unit, which is focused on remote and autonomous shipping, and to other projects the company is involved in, such as the Open Simulation Platform, she added.
A data-enabled ship is one that is transferring data – whether on energy usage, health, or conditions – to shore, allowing for remote monitoring. Ms Ellingsen Tyldum noted that it is not sufficient or practicable, given current reliance on satellite connections, to have a continuous stream of data being pumped, unfiltered, to shore. Therefore in Rolls-Royce’s system a lot of data processing will take place on board the vessel, and the compressed, timestamped end product will be displayed to the vessel’s crew and manager.
Key to this is the concept of edge computing. Currently, computing may be done in the cloud – for example, data is transferred in raw form from a vessel to a data centre for processing. With edge computing, the data is processed at or near its source, being transformed into something useful and valuable before it is distributed across the network. Ms Ellingsen Tyldum sees this as key to the digitalisation revolution, referencing the shipping container as “the box that changed the world” and adding “we have a new box that’s a lot smaller – it’s the Edge computer, that’s the new box that will change the world again.”
Noting that there are several asset-management solutions on the market, Ms Ellingsen Tyldum said what makes Rolls-Royce’s offering unique is it can offer an integrated picture, allowing for performance management and health management to take place at the same time. She gave the example of the integrated user interface not merely recommending a vessel should switch off an engine to improve performance but recommending the specific engine that should be switched off on the basis that that particular engine’s health data indicates it will require maintenance sooner.
Rolls-Royce’s offering is also strengthened by the company’s experience with interfacing various systems, understanding how, for example, the compressor affects the engine, how the engine affects the winches, and so on. Therefore, a lot of effort has been put into ensuring Rolls-Royce’s systems are able to interface well with different systems – “I can create a really good computer but if it doesn’t understand how to interface with all these different systems it’s not going to provide value,” said Ms Ellingsen Tyldum.
To meet the challenge of demonstrating an intangible product such as data to customers, Rolls-Royce has set up the Intelligent Analytics Centre Collaborative Canvas (IACCC). The IACCC provides a space for Rolls-Royce to present data visualisations and develop product features with customers to make efficient use of data to “provide a complete digital picture of the current and potential operational performance of their fleet”, said IACCC manager Jan Chirkowski. The company hope to thereby ensure its digital solutions provide comprehensive but relevant information in an intuitive format before they hit the market. “People need to trust the data,” said Ms Ellingsen Tyldum.
Alongside this is the Fleet Management Command Centre (FMCC), which operates as a proof-of-concept for a future product intended to provide a hub from which a fleet manager can access all and any information they require about their fleet. The suitably futuristic FMCC features a touch interface along with floor-to-ceiling 6-m wide curved display screens that show data gathered from portals such as the energy management and equipment health management portals.
Rolls-Royce has been offering some form of energy management analytics for some years now and it has undergone several revisions and improvements in this time precisely because of people not trusting the data, as Ms Ellingsen Tyldum put it. The issue was that early versions did not provide a detailed user interface on the vessel itself, just a simple dial, similar to that found on the dashboard of a car. Meanwhile, management on shore had access to a web portal that could provide information such as the number engines in use on a particular vessel. Rolls-Royce found that crews were wary of data gathered using this system as it did not provide enough information and context, so they invited crew members to participate in development in order to develop a suitable user-interface capable of adding value to the data being presented.
Equipment health management is another area that Rolls-Royce has a long history with. It has formed part of the parent company’s aircraft engine business for some time, with 12,000 engines sending data to Rolls-Royce’s control centres at a given time. On the maritime front, 300 thrusters and 600 steering gears are currently using this system, and it is also being developed for generators and engines. Ms Ellingsen Tyldum explained that early models using this system were very deterministic, attempting to predict a generalised outcome, but due to a lack of standardisation on ships, the volume of data produced and required for fault reanalysis was overwhelming. As a result, the company is experimenting with machine learning to sort the wheat from the chaff. One of the first areas this is being used in is with engines, initially land-based but now marinised, as the variety of data types make them good fits for machine learning algorithms to be trained on.
The machine learning model will process data from the equipment, and when it encounters something it perceives as being out of the ordinary it will flag it for labelling by a human operator in what is called ‘supervised learning’. The human operator then either gives the system as much information as they can about exactly what could cause the anomaly or informs it that the anomaly is not cause for concern. As the model develops, it learns from the labels applied to anomalies it encounters to judge what is actually cause for concern and what is within normal operating parameters and needs less and less human intervention. “Maybe the model is very stupid in the beginning, but you don’t need to have years and years of fault reanalysis,” said Ms Ellingsen Tyldum. Once operational, the goal is that the machine learning system will provide alerts when it encounters anomalies alongside advice on how to handle the anomaly and information such as time to failure.
Rolls-Royce is working with third parties such as original equipment manufacturers to enhance this technology by bringing them in to the anomaly classification and verification procedure.
What comes next
A key part of Rolls-Royce’s digitalisation strategy is the question of how to present data to users – or Ms Ellingsen Tyldum asked, “how do we make this intangible stream of value more tangible?” There are a number of technologies under development in this area.
The company is a keen participant in the Open Simulation Platform (OSP) initiative, operated by a consortium with the goal of producing ‘digital twins’ of vessels, which are simulated copies of real ships, including all systems and equipment.
These are produced by synthesising all information available about a vessel in a simulation and have applications in many areas such as design optimisation, maintenance, production and sustainability through a vessel’s entire lifecycle.
Alongside Rolls-Royce, other participants in the project so far include the Norwegian University of Technology Science, DNV GL, SINTEF Ocean, Hyundai Heavy Industries, Kongsberg Digital, Vard and the Offshore Simulator Centre.
A prototype is already operating with a simulated vessel and dynamic positioning system conducting a DP operation and Rolls-Royce is also utilising the tool in a project to analyse a vessel’s power and propulsion systems in a virtual test setup.
Another key area the company is devoting time and effort towards is using augmented and virtual reality technology. The distinction between the two is that augmented reality merely overlays information onto your surroundings, while virtual reality incorporates an entirely simulated environment.
One potential application would be to supply crew with suitably robust goggles with augmented reality technology built in. This would allow, for example, a lower-level service engineer working on a piece of equipment to bring the relevant manual up into their field of vision while they’re working on the equipment without the need for a physical copy. Having such technology networked would bring further benefits, for example the same engineer could call a colleague or service centre about the equipment and they would be able to highlight text in the manual, or annotate or circle relevant parts of diagrams, all while seeing exactly what the engineer is seeing live.
Similar technology could be utilised in the company’s work on remote and autonomous shipping, and how maintenance will be performed on it. One such area is live-controllable robots, which will ‘live’ on the autonomous vessel but will be controllable remotely for routine maintenance tasks such as changing filters on engines or performing checks. While this technology is not yet at a marketable stage, Rolls-Royce’s development team is already working on it, Ms Ellingsen Tyldum said.
Another application would be in producing virtual environments that accurately model a vessel’s engineroom for a chief engineer who may be based in a remote operation centre. This will incorporate data taken from the EHM system in the actual engineroom to highlight areas of concern that have been flagged as abnormalities by turning those components red in the virtual environment. In the same environment, that engineer would be able to pull up the manual for the affected equipment to establish what solutions can be found.
The concept of a virtual collaboration table is another area the company is looking into. This would be similar to a conference call, but rather than just seeing video footage of each other, the participants would share a virtual environment and see avatars of the other participants in that same environment. This would incorporate virtual models of equipment and systems, so that, for example, an engine could be viewed from all angles by all participants.