Covering Disruptive Technology Powering Business in The Digital Age

Home > DTA news > News > ‘Predicting the future’ with big data
‘Predicting the future’ with big data
September 6, 2016 News


Shipowners’ ability to harvest and analyse operational data from their vessels offers huge potential for ship design improvements, enabling shipbuilders to “predict the future” and build more efficient ships, said a panel of experts at the SMM Maritime Future Summit.

“A modern vessel is really using a tremendous amount of data,” said Paolo Tonon, VP of Maersk Maritime Technology. “Our Triple-E vessels have more than 3,000 sensors on board, each collecting two terabytes of data and 30 terabytes from the fleet, but the question is how we use that data to make decisions.”

“It’s a tremendous advantage to be a design company within an operator. You design a vessel, build it and then operate it. We use big data to optimise the vessels – this is a process that is happening on a daily basis at Maersk. We are continuously upgrading and modifying our assets.”

On-board sensors are becoming a key component in day-to-day ship operation as a result of advances in maritime satellite connections and their increasing accuracy and decreasing cost. Denis Morais, cto of SSI, a Canadian ship design firm, explained that in many cases these sensors “don’t even need power”.

“Every component can have a sensor – they can power themselves,” he said.

“There is no better way to predict the future than big data,” continued Morais. “Because it takes two or three years to build a ship, what you’re really trying to do is predict the future. Designing and engineering a ship is a lot of different processes.”

The exponentially increasing capability to do these calculations, or what Morais called ‘infinite computing’ gives designers flexibility to try almost infinite ship design scenarios.

Benjamin Vernooij, Internet of Things (IOT) and Edge Computing Technologist at Dell, agreed: “In shipbuilding, everything will be sensored-up. With that data I can predict when each component will malfunction. I can get statistical data, and with that I can predict the future,” he said.

The accuracy of data will be particularly vital, in order to avoid broken algorithms, and there are already concerns over sensor reliability. Vernooij stressed that shipping needed an “extra line of defence” against poor data.

This article was originally published on can be viewed in full