Skip to main navigation Skip to main content
The University of Southampton
Engineering

Expanding impact of machine learning solution reduces emissions in maritime shipping

Published: 27 October 2021
Amy Parkes
Amy Parkes

An innovative machine learning approach that predicts ship power usage in changing sea conditions is being rolled out across the globe to enhance sustainability in shipping.

The digital solution is helping captains optimise the amount of fuel and power needed in any given situation, reducing fuel consumption and lowering emissions.

Researchers from the University of Southampton and Shell Shipping and Maritime collaborated on the development of the Just Add Water (JAWS) app through the partners’ Centre for Maritime Futures .

The JAWS software is now available to hundreds of liquefied natural gas (LNG) carriers around the world thanks to an agreement between Shell and shipping technology provider Kongsberg Maritime.

Advances in the new modelling technique were driven by Southampton postgraduate research student Amy Parkes, who completed her PhD this summer.

“It is really rewarding to apply this fundamental machine learning research to challenges in the marine space,” she says.  “I developed a new error measure which makes machine learning techniques more reliable and interpretable, and therefore more trustworthy. I hope that that use of machine learning will continue to improve the sustainability of shipping and that this work can also encourage a shift of perspective in the machine learning field to concentrate more on the useability and interpretability of methods.”

The JAWS app interprets depths and angles of a ship known as the draught and trim. The software uses historic, high-frequency data from the vessel to determine the optimal conditions on previous voyages, which enables the system to advise on how best to enhance draft and trim. It also monitors and reports live fuel and emissions savings back to managers, to give real-time insight into the benefits of deploying this technology across a fleet.

During Amy’s PhD, engineers trialled the system on a fleet of over a dozen 300m-long LNG carriers for 12 months, cumulatively recording the saving of 250,000 tonnes of carbon dioxide emissions, equivalent to a fuel saving of $90 million.

Last September, Kongsberg Maritime announced that JAWS would be included as an application in the K-IMS suite ; a portfolio of specialised applications to support complex operations.

Amy’s postgraduate research was funded by the Engineering and Physical Sciences Research Council (EPSRC) and Shell Shipping and Maritime, with her time has been divided between Southampton and Shell. Her project was supervised by the School of Engineering’s Dr Adam Sobey and Professor Dominic Hudson .

“I really enjoyed the PhD,” Amy says. “I particularly enjoyed the fundamental research side that allowed me to really use my mathematics background to its fullest. But I also felt that collaborating with the real-life industrial partner really improved my ability to produce research.”

During her PhD study, Amy also applied machine learning techniques to a fuel-saving air lubrication system at sea. The partnership with Silverstream Technologies optimises the performance of the system that reduces frictional resistance between a vessel’s hull and the water.

Amy is now working with maritime consultancy Arcsilea who engage technically with global organisations such as the International Maritime Organisation (IMO) and advise on regulation for the environmental impact of the maritime industry.

“I am directly using my machine learning skills in the marine domain to help the implementation of international maritime GHG regulation,” she explains, “ensuring that vessels are as efficient and sustainable as possible.”

Privacy Settings
Powered by Fruition