Data Ingestion and Harmonisation for the Maritime Domain

The Maritime Industry is a massive business, connecting the entire world, as the main means of trading of essential goods. Nevertheless, there are challenges with the ever-increasing maritime traffic complexity, safety, performance, energy efficiency and automation. These challenges are driving the industry to embrace a digital transformation of the sector, with the application of state-of-the-art Artificial Intelligence, Big Data and High-Performance Computing technologies. With the extremely large amount of data generated by shipping, it is possible to apply these technologies to model the ships and their behaviours, create digital twins of the ships, as well as to model the traffic patterns in the sea, make optimal route predictions, etc. However, due to the vast number of actors in the Maritime Industry, the large amounts of data generated by the different actors is wildly varied, heterogeneous and complex. To use this data to train Machine Learning models and Artificial Intelligence technologies, there is a need for all the data coming from the different actors in the industry to be homogenised into a single unified format. To accomplish this, the authors propose the creation of the VesselAI Data Ingestion and Harmonisation Services, a tool that enables ingestion and harmonisation of generic maritime datasets. This tool provides the ability to map a raw dataset of choice to a harmonised schema with the application of Natural Language Processing algorithms, with no need to use scripts or develop code.