Improve Safety in Critical Maritime Operations

Project with the objective of implementing a prototype software for the extraction of the most advisable routes to be followed by vessels to improve the safety of maritime operations.


The aim of the project was to define a Machine Learning based methodology to perform a clustering of trajectories and innovative approaches for the extraction of trajectory waypoints, with tolerances in the space and time domain.

The definition and extraction of “preferred routes” was based on the analysis of existing historical AIS data recordings, regularly collected and stored during the last years by institutions like national Coast Guards, according to EMSA.

The most advisable routes to be followed by the vessels are extracted on the basis of the vessel characteristics such as length, width and draught.


  • Increasing port efficiency and safety: port traffic planning, aimed at increasing safety conditions together with port efficiency.
  • Traffic monitoring: the Fault Isolation Module was deployed, integrating the set of diagnostic models developed for each of the powertrain subsystems, to allow identifying the anomaly root cause. Aid to port navigation for pilots and master of ships and management of port infrastructure


  • Preferred routes: Preferred routes described as series of waypoints with space and time tollerances.
  • Statistical insights: Statistical insights on route characteristics (e.g., ship types distribution, flow direction, average speed,…).