An era where artificial intelligence will help ports manage berth allocation and tackle congestion may not be far off – but to make the most of algorithms in such safety-critical settings as ports and shipping operations, we must ensure that they are trained judiciously, writes Grant Ingram, CEO of Innovez One for the UK and EMEA.
The launch of ChatGPT has given millions around the world a first-hand experience of dialoguing with AI, with reactions varying from awe at the chatbot’s remarkable potential, to concern about its potential implications, to derision about some of its blatant mistakes.
Beyond the social media buzz, the arrival of conversation algorithms has also turbocharged interest for AI in companies’ boardrooms, with many in our industry wondering how best to approach a technology destined to transform the way we work, trade, and create.
Artificial intelligence already has a track record of powering breakthroughs in sectors as varied as finance and drug discovery – but what exactly can it achieve for the maritime sector? With ports and shipping operations, there is no room for mistakes. Unlike conversation algorithms, we can’t experiment at will with port efficiency and safety. Yet this new era nevertheless presents an opportunity to seize.
New challenges call for new solutions, and AI can provide an invaluable helping hand at a time when ports are reinventing themselves to support decarbonisation across supply chains in an era of ever-increasing supply chain disruption. We know that AI can transform port operations to boost efficiency and sustainability, because it is already doing so in several ports, including important hubs like Tanjung Priok and Tanjung Pelepas.
The right question to ask, therefore, isn’t whether or not we should use AI in maritime, but how we can make the right choices to ensure that AI-powered solutions are adapted to our sector, safe and reliable. Much of the answer lies in how algorithms are developed and trained.
Reality checks
One area where AI is already at play today is by automating and optimizing port, tug and pilotage operations. Using machine learning, a subset of AI, algorithms can “learn” from a port’s data to predict the duration of each upcoming job depending on the ship type and the different jobs required. Using these predictions, they can deploy resources such as tugs and pilot boats in the most efficient way, solving complex puzzles far beyond what humans and spreadsheets can achieve.
This may seem like a small step in the grand decarbonisation scheme, but it is essential to ensure that all the moving pieces fall into place seamlessly to welcome ships exactly when they arrive in ports, which has a tangible impact on idling times for visiting ships, congestion, and the port’s overall emissions. For example, in Tanjung Priok, our AI-powered MarineM system has reduced the overall distance travelled during tug and pilot operations by 20% and slashed average waiting times for visiting ships from 2.4 hours to 30.6 minutes.
To achieve these reliable results, having the right training is essential. With port operations, we can be objective, and teach the algorithm what a good answer is, with a carefully curated dataset and clear parameters. In practice, a key part of the training is a “reality check” where the algorithm sees its predictions and schedules tested against real durations – which helps it refine its answers and tailor them to the port’s specific operations and constraints.
A smart future
Going forward, we need to keep the same focus on solid and reliable results as AI is expanded to more areas of port operations. For example, machine learning can optimise berth management, which will become even more important going forward, as vessels will be powered by different types of fuels and clean technologies, making their needs for port services more specific.
Moreover, we can imagine a future where AI-powered algorithms would be trained to assess and predict levels of port congestion from aerial images, for instance, which could help ports identify critical situations, and take early action to ease congestion. The potential of artificial intelligence to support smarter and greener supply chains is undeniable, but not all AI is the same. Therefore, greater use of AI in maritime should be combined with greater understanding of the technology being developed – to ensure we train it, and use it, judiciously.
