Those stories come not just from tech giants like Facebook, Google and Alibaba, but also from smaller fangled-tech firms, such as SetelHellas, a maritime technology company in Piraeus!
SetelHellas' SeeMBox-V© team helped accelerate the pace of research into a form of machine learning, leading to huge advances in patterns recognition to reduce the potential for human miscalculations or poor decision making in the vessel operational chain.
As SeeMBox-V© machines, trained with large number of vessel data to develop clever algorithms, have become capable of carrying out more and more tasks, so our interest has grown.
Vassilis Balabanos, SetelHellas’ Project Manager and one of the project’s data scientists in Piraeus, says he is trying to apply machine-learning algorithms to the data feeds from the onboard systems in the onshore database. Fleet engineers employ behavioural experts to watch such reports, so they can work out how vessel operate and where to place corrective actions to the best advantage. With the right algorithms, SeeMBox-V© could automate the process and run it in real time.
Andrej Kozlov, SeeMBox-V developer director in Cyprus, says instead of algorithms presenting deterministic “yes” or “no” and "red" or "green" results to queries, our new system is able to offer up more probabilistic inferences about the sea world.
The system being developed are just beginning to be a broadly useful technology in shipping, and new algorithms presented at the press conference are likely to be adopted rapidly. Powerful computer systems and large volumes of data lie waiting for more exploitation. Moreover, SetelHellas have grasped the power of machine learning, and they are unlikely to let go.