Skip to Main Content
AVEVA Product Feedback


Status No status
Created by Guest
Created on Aug 20, 2022

Safeguard against multiple analyses writing to the same PI Point

Safeguard against multiple analyses writing to the same PI Point. Currently it is possible to configure more than one analysis to write to the same PI point. This is a problem since it often goes undetected until someone looks at the data and sees unpredictable results. The road to finding which analyses is usually long and tedious. In my opinion, one improvement would be to either: 1. Safeguard so that it is impossible to configure the system so that more than one analysis cannot write to the same PI Point or 2. develop a mechanism which warns about this configuration (if OSIsoft deems this behavior valid and an acceptable use case)
  • Attach files
  • Guest
    Oct 19, 2026

    The only way I have found to do this is to

    • Define all OUTPUT PI Point attributes with ReadOnly=False.

      • It would be nice if PI System Explorer would define this automatically when it is an output of an Analytic.

      • PSE does show when it is an output, so it should be possible.

    • Then dump all AF PI Point references into Excel.

    • Find just those attributes with the string ReadOnly.

    • Extract just the PI Point Name into a column.

    • Apply the Excel Conditional Formatting to highly duplicate values.


    RE: The Data Archive doesn't keep track of the exact...
    If you apply PI Point level security and allow the Analytics function read/write access to the PI tag, you can prevent two applications from writing to the same tag. e.g. PI Interfaces (or Manual Data Entry) and Analytics.

    There are those situations where you want to schedule updating a PI Point using Event Based AND scheduled. You can't do it to the same attribute in the same element, but you can create two different attributes for the same PI Point.

  • Guest
    Aug 20, 2022
    Unfortunately this is quite difficult to do. The Data Archive doesn't keep track of the exact application that is writing to a PI Point. As such, you can create two applications that write to the same PI Point currently. To complicate matter, the same service (PI Analysis Service) is writing to all the output PI Points from all the analyses so the Data Archive certainly cannot differentiate one from the other. Any prevention of these configuration errors would need to be done elsewhere.