AI-assisted technologies are changing the way shipping operates, helping to shed light on vessel performance, while cutting costs and reducing risk to human lives. Shipping companies such as CMA-CGM, Maersk, and Stena Line have deployed AI-assisted technologies to gain greater insight into their ships’ performance, while both established companies and start-ups are finding new ways to push the boundaries of AI.
Here are a few examples of how AI is being used to assist shipping operations.
Remote inspection
Artificial intelligence (AI) can be used to help remotely inspect or maintain difficult to reach places onboard ships, reducing risk to human life and cutting additional costs.
Drones are one example that are being used for remote inspections. Remotely operated and autonomous drones can reduce the need for human inspectors and lower the cost of inspections by eliminating the need for things like scaffolding that might be needed to inspect hard to reach places. Drones can reduce the time needed for inspections, thus improving efficiency and deliver faster reporting and decision making.
A project by classification society DNV GL called ADRASSO (Autonomous Drone-based Surveys of Ships in Operation) is using autonomous drone-based surveys, offering a safer and more efficient alternative to human ship tank inspection. The drones fly themselves into ships’ cargo or ballast tanks, and use AI to detect detect rust, cracks or poor coating conditions.
AI can also be used by robots to maintain vessels as well as inspect. Norwegian companies Kongsberg and Jotun have developed a robotic hull cleaning solution known as HullSkater that remains onboard a vessel at all times, and using big data and an intelligent analysis system, determines when its underwater surfaces might need cleaning. It then cleans using no divers or brushes to reduce risk to human life and ensure non-damaging cleaning.
For subsea inspection, maintenance and repair (IMR), a snake-like self-propelled robot like Eelume can be deployed to carry out IMR in confined spaces not accessible by conventional underwater vehicles.
Fuel efficiency
AI can also be used to improve the fuel efficiency of vessels by providing greater insight into ship performance.
Swedish ferry operator Stena Line has successfully used AI to predict the most fuel-efficient way to operate a vessel, reducing fuel consumption by 2-3 per cent per trip. The project began in 2018 onboard Stena’s Scandinavica vessel on the Gothenburg – Kiel route. The AI software, named Stena Fuel Pilot, will be installed on five more Stena Line vessels during the autumn. The ambition is to do a complete fleet wide roll-out on all 37 vessels in Europe during 2020.
Another Swedish firm, technology company Lean Marine, is also using AI to help manage fuel consumption. The company announced last month that its propulsion automation system FuelOpt had become AI-ready, allowing it to play a key role in enabling automated propulsion optimisation onboard AI-assisted vessels.
FuelOpt is also being used by Stena Line to help optimise propulsion in real-time based on AI commands on required values of power combined with the possibility for the operator to set limitations on fuel consumption and/or speed.
Japanese shipping company MOL is assessing the effectiveness of AI technology to help accelerate safe and environmentally efficient operations through the Ishin MOL Smart Ship Project.
Navigation and awareness
AI is also being deployed to help identify objects, reducing the risk of navigational hazards and improving safety at sea. According to Yarde Gross, CEO of start-up Orca AI, object identification capability is the backbone to autonomous operations, which we are seeing a rapid growth of in the maritime industry.
The Israeli company Orca AI is collaborating with other industry players including NYK Japan, Shell, Wärtsilä, and DNV GL to enhance navigation and enable the captain of a ship to improve his or her capabilities with the assistance of AI technology. The company created an algorithm that learns what to do as a result of continuous data that is collected and fed to it.
Shipping company Mitsui O.S.K. Lines (MOL) and Japan’s SenseTime announced a collaboration late last year on a new vessel image recognition and recording system using AI. The system’s graphic recognition engine was developed by applying artificial intelligence (AI) deep learning technology, allowing it to recognise surrounding vessels. The image recognition technology, which can be used even at night or during other periods of poor visibility, can detect small vessels that are not recognisable by vessel automatic identification systems (AIS).
Start-up Shone, which is headquartered in San Francisco and was formed in 2017 by three French engineers, has collaborated with shipping company CMA CGM to deliver AI-assisted voyages. Shone is accessing the shipping line’s vessels to develop its AI systems for container ships. According to the companies, CMA CGM’s AI-assisted voyages will help assist in decision support, safety and piloting assistance.
Predictive maintenance
AI also helps support predictive maintenance, enabling the early detection of equipment failures before they happen. Predictive maintenance is “all about the fleet management optimisation strategy,” says Rajan Vasudevan, CEO of OceanManager. “It’s about maintaining the weakest link.”
Predictive maintenance enables gains in operational efficiency to be achieved as the risk of equipment or vessel downtime is reduced by constant monitoring and alerts delivered from onshore experts. OceanManager’s vesInsight is an augmented intelligence solution that provides AI tools to enhance decision making for the management of vessel maintenance, supply, procurement, crewing and safety.
Another technology that uses AI for predictive maintenance is Wärtsilä’s Expert Insight. Leveraging AI and advanced diagnostics, equipment and systems are monitored in real-time. When anomalous behaviour is detected, it is flagged to specialists at Wärtsilä Expertise Centres, allowing them to support the customer proactively with an appropriate resolution to the issue.
Remote management
Last year, Maersk announced the launch of Captain Peter, an AI virtual assistant to help customers remotely manage their containers and gain greater insight into their cargo’s voyage.
The Remote Container Management (RCM) technology monitors the conditions inside the container from when the goods are locked inside, right up to when they are delivered to the final destination. Captain Peter keeps an eye on the container’s temperature, humidity, and CO2 levels, and notifies the customer if something needs attention. The data is now cloud-based for increased agility and can be easily shared as well as configured to the customer’s specific needs.
