What is big data and how will it affect the maritime industry?

The maritime industry is a strange dichotomy, on the one hand, it is a highly innovative industry that is on the cutting edge of efficiency and is the living core of some of the most complex logistics operations on the planet. On the other hand, it’s painfully slow to adapt to new technology, disparate and disconnected from the wider world, and well behind its younger sibling aviation when it comes to safety.

Technology has always evolved, and ships have always slowly adapted but the pace of global technological change is now so fast that the shipping industry is in danger of being disrupted by big data in much the same way that the media, communications, financial services, retail, and education industries have been in the last decade. Shipping isn’t “sexy” which has left it relatively sheltered from the venture capitalists and skinny jeans of silicon valley. But the industry that moves 90% of world trade is full of opportunities for the technologically capable and, to use an increasingly cliché phrase, is ripe for disruption.

We’re not breaking new ground here, if you’ve not heard the term big data banded around like it’s going out of fashion in the last five years then you’re speaking to the wrong people. It seems that every time I check my email I’m being invited to another seminar on big data analytics. There’s a reason it’s the current buzz word, not just in shipping but in almost every industry: it’s a fundamental game changer.

What makes big data so big?

We’ve all been doing “data” for decades – business analysts have been pouring over spreadsheets for years examining, for example, the fuel efficiency of a fleet of ships and recommending actions. But spreadsheets are limited to about 1 million rows – which is a lot of data but it’s not “big data”. Big data platforms can capture, store, and analyse petabytes of data at a time. If you’re not sure how big a petabyte is we can put it into context; 1 petabyte of storage space is about the same as 13 years of HD video, 50 petabytes of data is about the size of all of the written works of mankind, from the beginning of recorded history, in every language.

We’ve now moved far beyond the realm of spreadsheets. Instead of tracking the fuel data of a few ships, it’s now technically possible to track the fuel data of all 50,000 ships in the global commercial fleet – the barrier to mass data capture, storage, and analysis is no longer a technical one but a commercial and political one.

Size isn’t everything…

It’s not just about how much data can be processed, the other major big data breakthrough is the ability to process and make use of unstructured data.

To go back to our spreadsheet analogy – previously any data had to be neatly placed in columns and rows before it could be analysed by computers or it was about as useful as a cardboard lifeboat. Nowadays artificial intelligence has moved on to the point where you can throw “unstructured data” into the system and it will work out what it all means, neatly arrange it in a way that’s useful and come back with some helpful analysis.

This could be anything from swathes of text to photos and video, or even audio. Instead of just adding fuel numbers to a spreadsheet it’s now possible to take those fuel numbers, a video of the ship’s last pilotage and berthing, the VDR recording from the ship’s bridge, and all of the email correspondence between the ship and the office and discover answers to questions we were previously unable to consider.

Big data has existed for decades, a ship will produce billions of data points over the course of even a short voyage. We’ve just not been able to process it effectively and intelligently until now. Taking vast numbers of unstructured data points and being able to make sense of them helps us to do everything from better predicting the weather to better predicting port congestion. Imagine being able to predict with pinpoint accuracy a port delay weeks in advance and change a ship’s schedule accordingly, or accurately predict a stress fracture in the ship’s hull five years before it becomes a fatal event.

Realistically, how will this impact the industry?

The true impact of this technology remains unknown, but we can look at how it’s being successfully used in other industries to understand what the impact could be to maritime. Each of the approx. 3,700 Rolls-Royce aviation engines in service today has hundreds of sensors installed which are monitored by a team who use the data to predict when and where breakdowns will occur. Engineers and parts are constantly moved around the world based on the data they receive and Rolls Royce have completely changed their business model based on this capability. 70% of their aircraft division’s revenue now comes from their engine monitoring and repair service.

The USA’s Cheesecake Factory uses IBM’s big data platform to keep their food consistent across their 175 restaurants. It takes structured data like supplier name or location and combines it with unstructured data like the colour of ingredients to better diagnose issues with the food and predict them before they occur. All of the data is processed in real time meaning a strange tasting sauce or some dodgy beef can be caught before it makes it to your plate It  can also isolate the actual root cause of a problem and ensure that only food that is  affected is recalled rather than    pulling everything for testing.

Ship’s are becoming better connected with the outside world through the ever increasing bandwidth of satellite networks. Swap out aircraft engines with ship’s engines, cargo equipment, or navigation systems and swap Cheesecake factory ingredients with oil samples, ballast water, or exhaust emissions and you start to build a picture of the capability of this new technology. Before long it will be possible for an intelligent mooring line to warn the chief mate, the office procurement team, and its manufacturer that it needs to be replaced before it’s even been inspected.

The closer you look at the possibilities of big data the deeper the rabbit hole gets. A competitive advantage is increasingly hard to come by in modern shipping. Just blindly cutting corners and costs is not a sustainable practice and should be avoided if you even half believe in crew safety and welfare. Investing in your data is not just an intelligent way to increase efficiency, reduce costs, and improve safety, it’s one of the few ways left to gain a genuine competitive advantage. If you don’t do it, someone else will…