Predictive maintenance has been seen as a holy grail for cost cutting manufacturing. There are various steps involved in just feasibility study such as problem identification, sensor installation, signal processing, feature extraction and analysis, and finally modeling. Once a reliable and robust model is developed, the model has to be deployed to a manufacturing environment.
Various tools are being used for modeling and deployment such as R, Python, Docker, Kubernetes, JSON, PostgreSQL etc and will be discussing the process and deployment flow in this session.
I will be presenting “Predictive Maintenance: Zero to Deployment in Manufacturing” at ODSC East through virtual conference. The registrations are still open for this conference.