TrendMiner has released a new version of its software: TrendMiner 2.0. This release focuses on extending the capabilities for diagnostics, collaboration and advanced user administration.
TrendMiner software is based on a high-performance analytics engine for process data captured in time series. Process engineers and operators use the software to easily identify trends in their processes to optimize both efficiency and quality. Using multivariate pattern recognition, they can question the data directly without requiring help from a data scientist.
TrendMiner does more than just enable self-service discovery and diagnostics for process data – it also provides self-service monitoring. For instance, users can easily define a golden batch and fingerprint it, then set up multivariate alarms with predefined boundaries. TrendMiner monitors for these fingerprints and sends notifications to engineering or control room staff in the event of deviations or issues.
In predict mode, process engineers can use search patterns or fingerprints to check on any production process while it is running and predict if a deviation will occur in the future. TrendMiner uses past process data from the historian to deliver solid, data-based predictive analytics directly to the end user.
The improved influence factor functionality in the new version points to causes or influences upstream in the production line and improves the capability to configure early warnings. This allows for better prediction of future process deviations and the ability to control production quality, reduce waste and manage energy consumption.
In TrendMiner 2.0, monitoring is further enhanced as the improved influence factor functionality points to causes or influences upstream in the production line, increasing the capability to identify early warning indicators. In addition, the calculation capabilities have been extended to enhance analysis of process behavior. This improves prediction of future process deviations, control over production quality and optimizes energy consumption.
The release also includes scatterplot improvements to better highlight the influence of one process parameter on another. Furthermore, improvements to user administration and view sharing support enhanced collaboration.