TrendMiner 2021.R1 gives data science to experts

Also in 2021.R1 are extended capabilities to support multiple asset frameworks and many new user-driven features to help end users improve operational performance and overall profitability.

TrendMiner enables operational experts in process industries to analyse, monitor, and predict operational performance using sensor-generated time-series data. The goal of TrendMiner has always been to empower engineers with analytics for improving operational excellence, without the need to rely on data scientists. In doing so, TrendMiner brought data science to the engineer. In the 2021.R1 release, TrendMiner makes the next step of this journey by integrating notebook functionality into the software so that users can easily jump from looking at data in a TrendMiner view to working with it in a code-based data science environment.

With their data science libraries of choice (e.g. Pandas, NumPy, SciPy, SciKit-Learn), engineers can create and run custom scripts themselves for advanced statistical analyses and use AutoML capabilities to build machine learning models for anomaly detection. On top of that, they can operationalise the resulting notebook visualisations (also created with libraries of their choice such as Matplotlib, Plotly, Seaborn) as dashboard tiles in TrendMiner DashHub.

Thomas Dhollander, CTO at TrendMiner commented: “Classical data science depends on bringing process / asset know-how to the data scientist, while self-service analytics aims at packaging a subset of data science modelling capabilities and bringing these to the subject matter expert as a robust set of features (no technical tuning parameters, no data science training needed). Companies that recognise the potential in interweaving these complementary approaches will be the ones that can accelerate their operational efficiency and competitive advantage.”

To support enterprise rollouts and the increased complexity of existing IT-landscapes, TrendMiner has extended its capabilities for handling multiple plant breakdown structures also known as asset frameworks. OSIsoft PI users can easily connect multiple OSIsoft PI Asset Framework servers and set access permissions.

Besides support for multiple PI AF structures, multiple CSV asset trees can be imported for use as a data source within TrendMiner. As a result, System Administrators can better control accessibility with the ability to publish and unpublish structures, while the users have more flexibility to analyse the operational performance of multiple plants and production lines, each with their separate plant breakdown structures.