We’ve detected that you are using an outdated browser. This will prevent you from accessing certain features. Update browser

GreenSteam

Improving efficiency through machine learning.

LR’s recent acquisition will help maritime optimise operational efficiency and vessel performance

Some may view digitalisation as disruptive, but the technology and more importantly, the data it brings with it, provides opportunity. Yet, there are caveats. Maritime is fragmented with many companies developing solutions and then trying to find the problem, instead of solving the problem directly. We now have a jigsaw puzzle of niche solutions that don’t quite hit the mark. This has left the industry losing trust in tech promises and more specifically, digital solutions.

A collective effort is needed to equip maritime with the data it needs to tackle future challenges and make better decisions. Moving away from the silos we have today, LR aims to aggregate siloed data sources and use analytics to provide insights and outcomes to clients that will reduce their overall operational expenditures and maintain their competitive advantage.

To support this journey, LR acquired GreenSteam, a marine data company specialising in improving vessel efficiency through machine learning. The new business will be integrated within LR’s i4 Insight (‘Integrated Information . . . Intelligent Insight’) digital platform, part of the group’s Maritime Performance Services Division. This is all part of LR’s growth strategy aiming to give shipowners, operators and charterers the best possible view of vessel performance and fuel consumption across their fleets.

GreenSteam was founded almost 15 years ago by a group of three Faroe Islanders from the Technical University of Denmark (DTU), two of whom are still with the company – Daniel Jacobsen, Chief Scientific Officer and Jóan Petur Petersen, Analytics and R&D Head of machine learning.

“I think the cultural fit looks really good,” CEO Simon Whitford (pictured right) told Horizons. “We’ve received a really warm welcome from everyone at i4 Insight and LR. And when it comes to current and potential customers,” he continued, “we’ve been really surprised at how good a reception there’s been now that we’ve got the full i4 suite of services to offer compared to what in hindsight seems like quite a niche ‘point solution’ when we were on our own.”

GreenSteam’s founding vision was of bringing Machine Learning to the difficult task of decoding all of the impacts that external factors – such as weather, temperature/density of the seas, current etc. – can have on a conventional vessel, Whitford explained, as well as operational factors – like type of fuel, engine, choice of operating speed – and how they collectively combine to determine the amount of fuel needed to get the vessel from A to B.

“That adds up to approximately 15-plus variables you’re trying to understand in order to optimise something that before GreenSteam and Machine Learning came along was almost impossible to solve,” he opined. “But if you can understand how those factors conspire, then you can also control what you understand and obviously minimise your fuel consumption. And if you can do that, you have the opportunity to operate your vessel more efficiently and ‘steam green’.”

After all, “every tonne of marine fuel generates more than three tonnes of carbon dioxide emissions,” the Greensteam CEO pointed out.

That vision of having “a machine learning model for vessel performance at the heart of everything” hasn’t really changed over years, he continued, since in many ways GreenSteam was ‘ahead of the curve’ – except that in recent times “we’ve developed models that are super-applicable to short-term chartered-in vessels that can use the machine learning technology.

“What has changed is the tools that hang off that capability, and they go in two directions. Both looking backwards as analytical tools so we can now itemise all the components that make up a vessel’s emissions – such as the effect of wind, waves, fouling on the hull. And secondly, looking forward, we can use the models as vessel-specific prediction engines to then combine with weather forecasts and plan the future routes and operations of vessels in a way that keeps the cargo owners – the people whose freight our customers’ vessels carry – happy while minimising the environmental impact.”

“In terms of the current customer base, all i4 customers can be enriched by GreenSteam’s Machine Learning expertise across their fleets,” Whitford said. “All the information that i4 has really helps in a new objective that we have got to create Day One models. These are the ability to create a machine learning model that is better than anything else available and for a vessel, you haven’t collected data from yet.

“For example, if I’m thinking of chartering a vessel, I want a view from i4 Insight as to how that vessel is going to perform, and that’s where we’re heading now. What powers that is the vast amount of data that i4 Insight and LR have – we can learn from all the information that we didn’t have access to before. It’s like getting the keys to the library!”

The company has also developed a product called GreenSteam's Capture service that Whitford believes can be of real importance not only to i4 Insight but potentially to LR as well. This is a handheld smartphone or tablet Optical Character Recognition (OCR) based data acquisition app that “you can take onboard, or download from Google PlayStore, and then immediately start sending auditable, verified time/date-stamped data from any meter on the vessel.

“We’ve shown that doubles the accuracy of modeling compared to using data from traditional manual noon reports that ships send in,” Whitford added.

Sea

Read more articles like this in Horizons

Thought leadership, maritime insights and expert advice from Lloyd’s Register.

Explore the new issue

INSIGHTS

What we think

LR's experts regularly share their research and insights.

Can't find what you are looking for?

Hit enter or the arrow to search Hit enter to search

Search icon

Are you looking for?