Exis Technologies in conversation with HCB
Peter Mackay, Editor, Hazardous Cargo Bulletin spoke to James Douglas, CEO of Exis Technologies about the Hazcheck Detect cargo screening tool, dealing with fires aboard containerships and plans for the future in a recent Q&A Session.
With Maersk Line’s encouragement, Exis developed a new system, Hazcheck Detect, to look at shipping documents before containers arrive at the port. It searches for the tell-tale signs of misdeclaration and the alternative descriptions the more unscrupulous shippers are known to use to avoid the cost of declaring their goods under the IMDG Code.
James Douglas, CEO of Exis Technologies, says that the system is now screening around 13 million cargo items per month, getting some 500 ‘hits’ every day. Not all of those are critical and the issues may be resolved before the goods get onto the ship. Nevertheless, he says, between 40 and 50 boxes are stopped each week, any one of which could have led to a fire aboard the ship.
It is fundamental, Douglas says, for screening systems such as this to be aligned right across the industry. And, while Exis is still fine-tuning the Hazcheck Detect platform, it is already showing its worth. One great benefit is that it works in real time; scanning takes seconds, allowing the carrier to be alerted immediately if something looks amiss and to prevent the box being loaded onto the ship. Maersk Line says this is a really major safety benefit.
The keywords used to find undeclared dangerous goods – and there some 10,000 of them, working through 4,500 ‘rules’ – can be shared with third parties and this would, Exis believes, help achieve that level of alignment necessary to be sure that undeclared dangerous goods do not find their way onto a ship.
NCB and Exis have naturally been thinking about where they can take all this expertise and information in the future. Douglas is excited about the potential to apply machine learning and artificial intelligence (AI) to make some sense out of the massive data sets that are now available. “We can do a lot of things with all this data,” he says.
For instance, with their knowledge of those containers that have failed inspections, it should be possible to create a model to predict those containers that might fail at the next inspection, allowing NCB to target its resources more accurately.
The full conversation is available on YouTube here: https://youtu.be/6PiwG8VixFI