TrendMiner Updates Analytics Software

Analytics software from TrendMiner, a Software AG company, enables process and asset experts to analyze, monitor, and predict operational performance through trend analysis of time-series data. The TrendMiner 2018.R2 software update includes a new set of capabilities, collectively called ContextHub. Other additions include OSIsoft PI Event Frames integration, a related context item search for fast filtering of time-series data, and further extensions to the recommendation engine that helps operators and engineers speed up root-cause analysis of process anomalies. 

ContextHub is a repository, search engine, and collaboration platform for context items that can be neatly aligned to assets, processes, and events. The platform can be configured for context itself to become a new dataset that can be both visualized and analyzed. It also serves as a starting point for time-series analytics. Context items can be automatically captured, manually entered, or easily imported from other manufacturing solutions, such as batch systems, laboratory information management systems, computerized maintenance management systems, and overall equipment effectiveness systems, thus maximizing the flexibility of contextual information. 

The new ability to search for context items gives users the power to actively employ gathered context as part of the TrendHub analysis itself. It enables users to select saved ContextHub views from which they can visualize, filter, or overlay time periods in TrendHub. Context items can now become a starting point for trend analysis and facilitate filter requirements through all time-series data. This capability also speeds up root cause analysis or can even create fingerprints and monitors that can be used to send notifications to the control room and adjust process parameters when necessary.

The software, which was already integrated with OSIsoft Asset Framework, now provides out-of-the box access to both historical and new event frames for OSIsoft PI System users and others. 

The new release includes improvements based on user ideas and suggestions. For example, fingerprints used to monitor process performance can now be used for early warnings, which will help avoid the trigger of hard alarms and support control room personnel to take appropriate action. For diagnostic analysis, the new version provides a compare table showing the statistical difference between a range of tags of interest that can be exported to be used in other systems. The Recommendation Engine machine learning tool now uses the matching patterns found in the historical data to recommend potential root causes for process anomalies, which increases the likelihood of recommendations.

"As the digitization of organizations continues, more business-critical information sources are available, but often remain in silos,” said Rob Azevedo, product owner at TrendMiner, in a Dec. 6, 2018 press release. “TrendMiner's ContextHub brings all this information together in its self-service analytics platform, enabling true data-driven decisions for improving overall profitability. Engineers can see which plant assets experience the most downtime by easily gathering available data and analyzing the worst performer, allowing decisions to be based on data and reducing the reliance on a best guess."

Source: TrendMiner