It is a wide field, and depends of the particular type of machinery you want to maintain. In the best case you have historical data, sensory inputs, along with a classification (good, worse, bad etc.). Then the application of AI is learning classes. The latter can be done with the various pattern recognition algorithms, neural nets, decision trees, nearest neighbor clustering...

Regards,

Joachim

Joachim Pimiskern We are testing different approaches to defining predictive models based on failures data and through event logs, to anticipate preventive maintenance and avoid outages. We are looking for more data sources (sensors or data of reparation) that can contribute to a better detail of the subsystem or component in failure.

Dear António Brandão ,

SmartSignal can detect, diagnose, predict, and prevent critical failures. These analytics are built on unrivaled deep industry expertise and proven across the world’s largest energy organizations. Unlike generic AI/ML solutions, SmartSignal provides users access to powerful Digital Twin blueprints that accelerate time-to-value across your investments.

https://www.ge.com/digital/applications/asset-performance-management-apm-software/smartsignal-predictive-analytics?_source=google_g&&smartsignal-search&&keyword=preventive%20maintenance%20and%20predictive%20maintenance&matchtype=b&ad=618173310641&adposition=&device=c&location=1000998&gclid=EAIaIQobChMI9LHPxevF-wIVb4CDBx2UYgcpEAAYASAAEgIMEPD_BwE

Regards,

Shafagat

Shafagat Mahmudova SmartSignal seems too focused on analytics and not much on learning. In this type of solution, I have already analyzed the solution of SAS (https://www.sas.com/pt_pt/software/viya.html )with little success in the machine learning and AI component.

https://www.automate.org/industry-insights/getting-started-with-ai-based-predictive-maintenance