Automate the Swedish port call
| Reference number | |
| Coordinator | Sjöfartsverket - Offentliga medel |
| Funding from Vinnova | SEK 499 033 |
| Project duration | May 2020 - May 2021 |
| Status | Completed |
| Venture | AI - Competence, ability and application |
| Call | Start your AI-journey! For public organizations |
Important results from the project
The project has aimed to create the basis for automation of reporting and pilot orders concerning the Swedish Maritime Administration´s parts of the Swedish port arrivals. The project has delivered capacity for prediction of arrival at the end point of the call. The project has delivered predictions of the pilot´s start time, end time and thus also the pilotage time consumption. All predictions for places places and times can have a more general use regardless of whether the ship is subject to pilotage or not.
Expected long term effects
We have developed a ready to use prototype that involves a proof of concept for creating predictions for which ships will need assistance from pilots, from and to which boarding place the pilotage will begin and end, when the ships need piloting and when it is expected to call at the port. This provides the conditions for automation of both ship registration and pilot order and lays the foundation for further digitalisation of other government reporting. The potential for efficiency and simplified reporting is great as the effects can be applied to up to 80,000 calls.
Approach and implementation
LiU and the SMA have developed an AI-based prototype for predictions of the call process based on calls to Bråviken. The system monitors individual ship objects in real time and draws conclusions about their status based on, among other things, AIS information. The prototype development used already existing technology, SOMA-AI, developed by LiU. SOMA-AI is an AI platform for interpreting large amounts of streaming data and tracking objects in real time.