Winner of the Digital Industry Challenge 2017-2018 organised by General Electric and NUMA
Why predictive maintenance?
Our Vision: Build the future industry with Zero Defects and Zero Downtime.
Current maintenance strategies are mostly based on corrective and preventative methods
Preventative maintenance processes and downtime duration are intentionally over-estimated to minimize risks. However failure and production downtime is still a costly risk.
Equipment are getting more and more connected
- 13 billion connected objects in 2016 / 29 billion in 2020. (IDATE)
- Less than 20% of the assets are connected today and 70% of the collected data is still not used (World Economic Forum)
- Predictive maintenance is the #1 B2B application with 250 million connected industrial equipment in 2020 (ABI Research)
- An innovative Automatic Feature Generation algorithm for time series
- Powerful and performant Predictive Maintenance Models
Good to know
30% win on cost over preventative maintenance
50% over corrective maintenance
(Electric Power Research Institute)
What is unsupervised learning? Unsupervised Learning (see wikipedia) is used when no historical data is available that involves past faulty behavior
A l’occasion de la sortie du rapport Villani sur l’Intelligence Artificielle, j’avais donné mes premières impressions en tant que fondatrice