TrendMiner announces release of TrendMiner software version 2017 R2

  • December 22, 2017
TrendMiner announces release of TrendMiner software version 2017 R2
TrendMiner announces release of TrendMiner software version 2017 R2

December 22, 2017 - TrendMiner NV, provider of a self-service process and asset analytics solution, announced that it has released a new software version: TrendMiner 2017 R2. This release focuses on leveraging global connected data sources to unlock data insights across the extended enterprise for optimizing overall plant performance.

TrendMiner software is based on an analytics engine for process data captured in time series. Subject matter experts use the software to search through big production data, analyze the results and identify trends in their processes in order to optimize efficiency and quality. TrendMiner indexes data from historians and connected data sources to create real-time interactive analytics. Search algorithms, combined with pattern recognition, help to uncover previously hidden relationships and identify causes of process behavior. With direct access to analytics insights, actionable information becomes available at all levels of the plant.

The latest TrendMiner release delivers enhancements to support companies with globally dispersed sites and teams. The self-service analytics platform delivers extended connectors and a new offering to help users to solve previously unsolvable cases.

TrendMiner 2017 R2 includes a recommendation engine to help subject matter experts solve challenging cases. This recommendation engine can proactively suggest tags that may be relevant to the analysis, including influence factors in upstream process sections with potentially significant time delays. In combination with the search and discovery capabilities, users can now examine a process anomaly of interest, search for more instances of similar behaviours, and have TrendMiner generate hypotheses for diagnostics based on all selected occurrences, all within the same interface.

The automated suggestions guide users on where to find the most relevant results and help them to proactively generate a hypothesis about the root cause of behaviors. Instead of relying only on their individual knowledge, users can gain insights from all the indexed data and solve production issues faster. Any sensor measurements that have been indexed by TrendMiner can be used to inform suggestions, regardless of the originating historian or connected data source.

As an OEM Partner of OSIsoft, TrendMiner has introduced a new integration with OSIsoft PI Asset Framework (PI AF). TrendMiner R2 allows users to choose between using the internal Asset Framework of TrendMiner or to use PI AF as an external provider. When TrendMiner is configured to use PI Asset Framework as an external Asset Provider, the complete asset model is managed in PI AF. Changes in PI AF can be reflected in TrendMiner by manually triggering synchronization.

The structure of PI AF can be synchronized so that instrument tags are presented in TrendMiner and related to the corresponding assets within the plant breakdown structure from PI AF. Tags can then be indexed with automatic relations to the assets and annotations can be attached to any asset in the hierarchy, based on the PI AF structure. This allows users to browse for tags on a plant breakdown structure.

By visualizing the asset structure defined in TrendMiner or from the PI Asset Framework, users now can e navigate the structure of their plant to find relevant attributes to work with.

TrendMiner 2017 R2 is designed to support scaling out across business units from single sites to enterprise level. The multi-node setup version of the software features a fully distributed index store and compute layer. This supports enterprise level scale-out and global installations across multiple servers and multiple historians for thousands of users. 

In TrendMiner R2, integrations have been further extended to include full support for Honeywell PHD, GE Proficy and Wonderware. Mechanisms to handle naming conflicts between tags in different historian sources have also been improved to better support using a large number of data historians at global enterprises.  

Learn More