TrendMiner 2024.R2.0

29th of July 2024

In TrendMiner 2024.R2.0, event data creation, visualization and analysis are more powerful than ever. TrendMiner now provides a Plug ‘n’ Play connector to SAP Digital Manufacturing to enable yet another source for assets and context items. Users are now also able to write calculations on events to context item fields, making future analysis and event reporting easily reproducible and sharable with colleagues. An expanding ecosystem means more event metrics are now available to be analyzed. In addition, scatterplots may be formatted by layers in TrendHub and by conditions in ContextHub, enabling comparisons of operating modes and correlations between identified groups of events.

To ensure security and future updates, we're transitioning to a new OS version. A new Remote Backup and Restore feature supports this migration, securing all data and settings with backups on AWS S3 or Azure Blob for improved disaster recovery.

Finally, you can now connect with fellow TrendMiner users and experts through the TrendMiner Community. This community provides a platform for public and private discussions, best practices, tips, and success stories. ​Sign up here​​.

Innovation Highlights

• New OS – AlmaLinux - We provide a fundamental update to the TrendMiner application, as we transition from CentOS 7/RedHat 7 Enterprise to AlmaLinux OS 9/RedHat 9 Enterprise. This update ensures that the users continue to operate on a robust, stable, and secure operating system. 
  The New OS will require a new server and a full migration when upgrading to 2024.R2.0. There will also be new minimum system requirements to run the new TrendMiner version. More information here: https://documentation.trendminer.com/en/102710-create-and-configure-the-new-trendminer-server.html
• Full Backup and Restore – To facilitate the migration between CentOS and AlmaLinux, the new remote Backup and Restore feature provides complete data protection and seamless recovery. This solution offers full backup and restore functionality, with all data stored securely on AWS S3 or Azure Blob storage. To use this feature, you will need an AWS S3 storage account or an Azure Blob Storage account.
  More information: https://documentation.trendminer.com/en/101680-introduction.html
• Write calculations on search results to context item fields – Capture the power of ContextHub without an external sync. Users can now write calculations on search results as context item fields, making the values reproducible and accessible by colleagues for tracking and further analysis.
• Chart alias support – Quickly interpret analysis and communicate findings by choosing your chart alias of choice. Users can now create a custom alias or use the synched tag description on the focus chart and associated tables. Chosen aliases are saved with the TrendHub view and are propagated when sharing and using as tiles on dashboards.
• Scatterplot formatting by layer – Reduce the need to export chart data for further visual analysis and expedite root cause analysis. Layers can now be used to differentiate between different periods of operation on your scatterplots, making it easier to identify changes in KPIs, operating parameters, and correlations between them, across different periods of interest.
• Conditional labeling on ContextHub scatterplots – Increase the analytical power of configured Context conditions by using them to format context scatterplots. This new Context View configuration option provides a graphical representation of context field values, providing further detail and attention to events as they occur.
• With the latest Plug ‘n’ Play Connectivity to SAP Digital Manufacturing, you can now seamlessly access shopfloor data and events directly within TrendMiner. It is now easy to combine process time-series data with contextual data from SAP, facilitating enhanced monitoring, prediction, and delivery of product quality. REMARK: the SAP Digital Manufacturing is not available by default, but it can be enabled “on-request” by your Customer Success Manager.

Application Enhancements

• The parallel coordinates plot feature has been enhanced to allow more customization.
• The generic ODBC Connector has been improved to better support native DateTime conversions, resulting in enhanced performance.
• The Microsoft ADX connector now automatically supports streaming data implementation. This enhancement prevents out-of-memory issues by enabling data to be processed in a more resource-efficient manner.
• To avoid confusion regarding the duration of periods, durations will only be presented as multiples of days and no longer as multiples of months or years. The quick selections for the context chart in TrendHub remain unchanged.
• The MLHub Notebooks user interface has been significantly enhanced to offer a cleaner, more intuitive experience, making the processes of publishing and editing smoother and more user-friendly.

As of 2024.R2.1

• Extended Full Backup and Restore – Release 2024.R2.1 introduces support for Custom S3 providers, expanding your full backup storage options beyond the public cloud services AWS S3 and Azure Blob storage.
  
  You can integrate and store your backups with any compatible S3 provider. This includes S3 providers like Min.IO or any AWS S3 API (v4) compatible storage provider, which can be installed locally or configured to store files remotely. This enhancement offers greater flexibility and customization in managing your data backup and recovery processes.
  
  For more information:
  https://documentation.trendminer.com/en/101680-introduction.html
  https://documentation.trendminer.com/en/local-custom-s3-backup-restore.html

Bug Fixes

• Fixed an issue where the export of value based search results failed if the search definition contained a condition to find all states of a digital/string tag, and the tag has a lot of states. The definition of this search was too big to be exported to Excel.
• Fixed an issue where unsubscribing to monitor alerts did not work if capitals were used in the email address configured in the monitoring notification.
• Fixed an issue where importing an asset structure in CSV format, containing empty spaces or lines at the end of the file would result in errors and an unsuccessful import.
• Previously, when a user selected a subfolder as location for publishing notebook pipeline in MLHub, it was published in 'My work' folder instead. This bug is fixed now and pipelines are published in the correct folder.
• Fixed an issue where the CSV import failed when importing context items for an asset that did not have any yet.

As of 2024.R2.1

• Fixed an issue where the indexing of ML tags failed because of special characters in the tag name.
• Fixed an issue where filtered search results were not immediately reflected on the axis in the parallel coordinates plot, requiring tab switching to update.
• Improved scrolling behavior when some notebook outputs are shown or a cell is added.

Known Limitations

• Conditional formatting for the Gantt will not consider any changes to durations propagated through the use of live mode. To see the updated coloring, users have to refresh the chart.
• External content tiles in dashboard need cookies to properly load all content (and be able to login when required).
• Scooter values may change when zooming in after adding the scooter. This is because scooters currently interpolate the data points on the chart.
• When calculated tags are deleted, they cannot be created with the same name again.
• After renaming a calculated tag linked to a context item, the component of the context item is not updated correctly.
• The action displayed on the right-side panel of ConfigHub when a service is restarted, is no longer visible, these are visible in EdgeManager.
• The Jupyter notebook files and PMML models are not backed up automatically by TrendMiner in this release. On-premise migrations are still to be done manually by customers by downloading the PMML models and then uploading them on the new system. This is not the case for customers migrating from Azure to AWS with Cloud Storage where back-up is done automatically with Velero.
• Notebooks can be shared with other users in the Work organizer. Without “Save-As” functionality, only the original author of notebook can make changes to the shared notebook.

Deprecation Notice

• Proficy 4.5/5.5 connectivity: Please note that the integration with Proficy versions 4.5 and 5.5 will be sunset as of the 2024.R4.0 release. We encourage upgrading to the latest versions of the Proficy historian to benefit from enhanced features and support.
• Due to a change of technologies the in-application Help Center will no longer be available in TrendMiner versions 2023.R2.0 and older from October 1st, 2024 onwards. Please upgrade to a more recent TrendMiner version timely in order to continue browsing through the TrendMiner documentation directly in the application.Keeping your TrendMiner installation up to date with the latest versions is advised to stay current with security patches and new functionality.

  Outside of the application, the latest TrendMiner documentation for end-users and admins is available through:

  https://userguide.trendminer.com
  https://documentation.trendminer.com

Synopsis

Chart alias customization

By popular request, it is now possible to specify a personal alias for every tag or attribute. You can set the personal alias in the tag details and it can be used as labels on the focus chart, statistics table, and compare table. In addition to the chart alias or tag name, you can also choose to use the tag or attribute description as a label on the focus chart within the focus chart settings.

Company tag naming conventions often result in long, complex tag names. By providing a chart alias, it is easier to identify your tag of interest and more meaningful names can be used when sharing views with colleagues who are not familiar with historian tag names.

Scatterplot formatting by layer

In earlier TrendMiner versions, scatterplots in TrendHub would only display the data of your base layer, even if additional layers had been added to the focus chart. In the newest release, scatterplots can be configured to display all visualized layers.

By default, the scatter plot will open as a multi-scatter visualization (i.e. multiple tags), showing only the base layer. You can switch to a multi-layer visualization using the top right navigation button.

In the multi-layer visualization, data points belonging to one layer will all have the same color.

The “Customize layers” button opens a popup that allows you to define how all sublayers will be compared to the base layer.

• Compare all layers: Assigns a unique color to each layer.
• Base vs. the rest: Uses one color for the base layer and another color for all other sublayers.

In addition to comparing all layers individually, you can also assign the same color to periods of the same category, such as ‘good’ or ‘bad’ behavior, enabling you to identify deviating patterns between these behaviors.

By default, all data points are visualized as circles. To enhance the visual distinction of layers, the 'Enable markers' toggle can be activated, which will assign a unique marker to each layer.

Parallel coordinates plot enhancements

In this version, the following improvements have been added to the parallel coordinates plot:

• Changing the order of the axes
• Reversing the direction of the axis
• Color results based on calculations on digital/string tags

Clicking the 'Color by' button, will open a popup where you can select a configured calculation on a digital or string tag and the corresponding string values you want to distinguish by color. A unique color will be assigned to each selected value based on a predefined set of colors.

To enhance the readability of the chart, you can hover over the corresponding label in the legend, which will highlight the results on the parallel coordinates plot.

Improved Create, Import and Edit notebook UI

Users can now navigate across two separate flows; 'Files' to create, open or import a notebook and 'Notebook content' to edit a notebook and use code snippets. Previously, the code snippet component was displayed even though no notebook was opened.

Write calculations on search results to context item fields

Users can now save values from calculations performed on search results to context item fields, ensuring persistence and democratization of valuable data within TrendMiner. The new feature enriches data contextualization capabilities, enabling deeper insights, more visualizations and analysis options within ContextHub, while also building a better integration between TrendHub and ContextHub.

Scatterplot by ContextHub condition

Users are now able to change the color and shape of specific points on the context scatterplot based on predefined rules. By setting rules and conditions, you can visually distinguish data points that meet certain criteria, making it easier to identify patterns, outliers, and key insights at a glance. Whether you are highlighting data points within a certain range, marking significant events, or differentiating between categories, this enhanced visualization capability will enable more effective data interpretation and communication. This feature empowers you to create more informative and visually compelling scatterplots, enhancing your overall data analysis experience.

You can explore the new formatting feature for the scatter plot in ContextHub by clicking the formatting option in the top right corner, next to the chart settings button.

Extended Full Backup and Restore (As of 2024.R2.1)

To improve resilience and reduce recovery times, minimizing disruptions from system failures, we introduced a backup and restore solution that stores the full TrendMiner application data. You can choose between remote or local storage options for storing the backup data, providing an extra safeguard in case of local system failures.

Additionally, our full backup option is recommended for migrating from CentOS to the new OS, AlmaLinux.

Full Application Back-up

• Full Back-up: Captures all application and database data, including the index files.

• Highly Self-Serviceable: Designed for ease of use and self-management.

Remote Back-up/Restore

• Remote Storage Options: Supports AWS S3 and Azure Blob storage.

• Requirements:

  • Requires an AWS S3 or Azure Blob storage account.

Local/Custom S3 Back-up/Restore

• • Storage Options: Supports local or remote storage on S3 storage providers.

  • Requirements:

    • Requires setup of S3 storage.

    • S3 storage provider must be compatible with AWS S3 API (v4).

    • Suggested local S3 provider: Min.IO