Customer would like to do Batch reports in BI Tool and Annotations contain important information, like Operator comments, reasons for changed values etc.

There are a variety of reasons why being able to handle views in bulk would be nice. If I am cleaning up the UI it would be easier to delete multiple views at once. If I am doing maintenance on a target and I want to stop a subset of views, then I currently have to do them one at a time. Allowing the option to select multiple views would aid in these tasks.

Have to be able in the filtering page of the PII4BA to 'include all columns where any/all of these conditions are true' and then select where a given attribute is greater, lower, etc. than... another attribute present in the preview table. Now I am able only to define constants (e.g. 100, 50).

Other products (such as DataLink) will allow for English PI Time to be utilized regardless of the language settings. For instance, both "*-8h" and "*-8horas" will work in DataLink if Spanish is the selected language. However, only "*-8horas" will work if Spanish is the selected language for the integrator in the UI.

The request is to bring the integrator more inline with how other OSIsoft products handle PI Time in different languages.

Currently the Views drop down in the Admin > Logs page shows all views from AF Configuration. The list can show apparent duplicate view names, since it also includes deleted views. Either allow the option to filter out deleted views, or have the ability to clean up old views from AF Configuration.

Support integration to non-Esri GIS systems such as GE Smallworld

It would be nice if the PI Integrator installation created a Start Menu shortcut to the Integrator User Interface, similar to how a PI Connector installation creates a Start Menu shortcut linked to the PI Connector Administration page.

This request applies to any OSIsoft product that requires a web browser: If certain browsers or browser versions are not supported by an OSIsoft product, the browser version should be detected and an error message shown to the user so they know to use a different browser or version.

The Integrator application server should have a status page ("OK" or "Running") so that clients or load balancers can easily determine whether the site is up or not.  This should be implemented when the Integrator eventually supports high availability.

As a data science user, I would like to be able to include additional PI/AF tag information such as is Questionable so that I may filter out data down stream easier.

Currently, the only available decimal separator is the period (.). In some cases, the customer may want to use other decimal separators such as the comma (,). It would make sense for the decimal separator to follow the convention of the region and language selected in the UI, or have an option in the menu set it to something custom.

Write back to PI from the Integrator for Machine Learning. Currently we are writing back to Excel from Azure Machine Learning Service but would like to do this with less integrations.

Implement Testing methodology for Microsoft Patches
PI System owners need to ensure that their deployed systems both meet internal IT policies regarding recommended OS and other component recommended and critical patches, and that their PI applications are not adversely affected by these patches.

Consider a testing methodology similar to what PI AF employs which is currently to automatically monthly test all critical and recommended patches against the currently released application version.

As a PI admin using the integrator with event frames for reporting, I would like the integrator to calculate summaries across event frames. For example, I'm interested in the average value of an attribute from the beginning of one phase to the end of the operation, where the phase and operation events are generated by a batch interface.

As a PI admin working with the integrator, I would like to be able to limit the amount of data returned for new elements added to existing views. This avoids the situation where a long-running view is unintentionally slowed when the integrator backfills data for the new element over the whole history of the view.