HPC-based vessel predictive maintenance optimization through Natural Language Assistance

Presentation of the problem and objective of the experiment

FIGAL Innova and SREC Solutions developed a system to monitor and assist vessel’s crew and operators for the predictive maintenance of the ship machinery all along their working period and conditions: when to make a maintenance operation in which machinery and what to do. The partners update the system to optimize crew effort using Natural Language Understanding, an Artificial Intelligence technology that enables the interaction between people and machines using predefined commands in human language.

Short description of the experiment

MARINE WORDS targets to experiment with advanced HPC services to train the Natural Language Understanding algorithm for assisting marine transportation predictive and preventive maintenance processes in multiple languages, with specific jargon and in noisy environments. The final aim of using this application is to cut vessel’s operating expenses of predictive maintenance in an efficient manner, all along their working period and conditions: in port, navigating in the sea, with connectivity, and periods without connectivity.

Organisations involved:

ISV: SREC Solutions S.L.
HPC Provider: CESGA