Foundations of advanced analytics: A layered approach fit for purpose

For data quality, the scope of the analytics has to be understood in its usage context: individual sensors, single pieces of equipment ,a functional group (FEG), or an entire process unit or a plant. Similarly, for maintenance, the analytical requirements must be framed around popular maintenance strategies – reactive, preventive, condition-based and predictive. And, for a process, the layers are often described as – descriptive-diagnostic-predictive-prescriptive. A layered approach can encode SME (subject-matter-expert) knowledge in fit-for-purpose analytics, and you can deploy the analytics in a hybrid fashion – both physics/engineering (first principles) based analytics as well as data driven with machine learning via open source or third party libraries.

Join us for this webinar series where we first explore a layered approach to analytics for use cases in data quality, condition-based/predictive maintenance, process diagnostics, and others. We will introduce the data engineering aspects in the analytics workflow and prepare you for what to expect regarding operationalizing the analytics.