Visualize Questionable Flag

1. What specific problem are you trying to solve with this enhancement ?

Over short or long periods, acquired and archived data may appear frozen at a seemingly consistent value, and we do not have a simple way within PI Vision to visualize the status of the “Questionable” flag in order to verify the state of this data, even though this information is available in the PI System.

2. How would you like this feature to work?

For an attribute with Data Reference “PI Point,” make the “Questionable” flag available as a readen property (within a template and/or an AF element), so that it can be exposed in PI Vision within the available graphic objects (trends, values, time series tables). This information should, if possible, reflect the status as identified in the acquisition interface (“Uncertain,” “Suspicious,” “Bad,” etc.).

3. Who will benefit the most from this enhancement?

All PI Vision end users.

The “Questionable” flag is just as important as the value of a PI Point. It must be exposed to end users so it can be easily used across the different PI clients. Data acquired and archived in the Historian are often "frozen" due to upstream connection losses in the data collection chain or gateway failures, and the end user does not have an easy way to verify this from PI Vision...

In ProcessBook, the quality of data was indicated with a "?" at the top of trends indicating that the values was suspect from the attributes stored with the archived data point sample. This made is easy to see when data may have flatlined or be incorrect because the underlying data was suspect.

In PI Vision, this functionality does not exist and trended value that appears flatlined may or may not be suspect. While it is possible to create and AF analysis that looks at the BadVal() function of a point, trending it in addition to the value does not make the trend easily understandable and requires a lot of additional analysis to be created to see this information.

I have attached 2 screenshot. The first is the original representation in ProcessBook. The second is an example of trying to re-produce the quality using analysis trends for a value. Whether the indicator is directly within a table or graph or at the top works either way and would be used by most users in the organization (Energy Industry). The quality of data is important for determining the state of a site or an issue as part of troubleshooting.

The questionable flag should be visible not only in the trends but in the Time Series Table as well.

The visualization of the questionable bit in PI vision could also be improved upon over what we had in ProcessBook. In ProcessBook, for trends the questionable bit (?) marks did show up in the legend for each tag's current value which is good, but unfortunately for the graph/historical portion, the (?) marks were shoved to the top of the graph, forcing the user to hover over each (?) mark to figure out which trend marker was questionable at the time. It would be better to enlarge the regular marker on the trend line a little and put a (?) inside the marker whenever the bit is true. Then the user could instantly tell which trends/markers were questionable at a glance.

In addition, there is a bug in AF Analyses where the questionable bit on an input will break the output unless a workaround is employed to scrape the bit off. AVEVA should not only fix this bug but embrace the questionable bit by adding the option of "questionable bit inheritance", such that if any input has a questionable bit, the output will inherit the questionable bit. Inheritance of quality in calculations is a fairly standard feature in SCADA and should exist here also because it allows the user to know when an output of an analysis is questionable due to a questionable input.

The only way our users can see the questionable flag now is with PI Datalink Compressed Data.

We use the questionable bit to indicate when value (from EMS, PI DNP3 and other interfaces) PI tags have questionable quality, and to indicate when companion quality PI tags (most EMS value PI tags have these) are sourced from our alternate EMS site. Both are very valuable for troubleshooting and situational awareness, as well as for SQL queries against PI to find, count or exclude PI tags with values that are questionable.

Quality status needs to be added in Vision, similar to datalink.

Engineers need to be able to see the SCADA quality status on displays, this functionality is still unavailable.

Really need to be able to see the questionability of data for PI Tags on trends and points. This is crucial for my industry (energy) and third party options or work arounds do not provide the functionality that is already present in ProcessBook. Since these attributes are already stored as part of the data, is there some way to just implement this option into the graphs and data points to emulate what already exists?

We would like to see the comments in the events directly the table.

No feedback on this request? A simple addition to the 'value' symbol would be helpful in the 'visibility' pane.

The visualization and functionality of PI Vision leads a lot to be desired. There is a lot that ProcessBook can do that PI Vision can't and renders PI Vision useless.

The same goes for the other 2 flags: Annotations and Substituted data. There are business cases to know via visualization if the data has been Substituted and/or if there is an Annotation (note) on the data. None of these 3 flags (Questionable, Annotated, Substituted) are are currently supported/visualized in Vision as of March 2022.

What's the state to this request? It's already very old and has a lot of votes. Especially as PI Processbook is soon end of life, it's important to have this kind of use cases also supported by PI Vision. Also it's important to show other flags like Substituted and maybe Annotated.

In PI Processbook I was able to see directly in a display if a value was questionable etc. We need this functionality also in PI Vision because otherwise you are not able to see if something is wrong with the data. In my opinion this is a development back if the new client tool doesn't support this simple functionality of the "old" PI Processbook.

This is a critical feature for my client. It is highly utilized in their ProcessBook displays and extremely helpful in identifying bad data which otherwise could create unnecessary work and alarms. Users will not migrate from PB to Vision without this functionality.