HPC VESSEL MAINTENANCE OPTIMIZATION BY NATURAL LANGUAGE ASSISTANCE

  1. /
  2. POC
  3. /
  4. HPC VESSEL MAINTENANCE OPTIMIZATION...

FIGAL Innova and SREC Solutions developed a system to assist vessel operators for predictive maintenance of ship machinery. The partners optimize crew effort using Natural Language Understanding, an AI technology enabling interaction between people and machines using commands in human language.

Start date: 01/06/2021

Duration in months: 18

Problem Description

The maritime industry is addressing rising maintenance costs by integrating virtual assistants into their processes. FIGAL Innova and SREC Solutions have developed a Vessel Predictive Maintenance system, which monitors and assists crew and operators with predictive maintenance of ship machinery.

Goals

New services

Challenges

The virtual assistant must work in noisy environments and require complex command recognition. To reduce vessel operating costs and improve predictive maintenance, a Natural Language Processing system with increased reliability is developed, integrating into the virtual assistant.

Innovation results

The Marine WORDS project developed a Natural Language Processing system for noisy environments, using advanced Deep Learning techniques. The system, trained on a remote HPC infrastructure, achieved 95% accuracy in noisy environments (up to 80 dBm) with only software.

Business impact

FIGAL and SREC new solution could reduce maintenance costs by 30%, SREC estimates a 20% annual income increase in the next five years. The collaboration also aims to create 10 local jobs and increase the number of Spanish SMEs using CESGA's infrastructure and services.

Project page

Follow the external link