Include a quality marker from PI on datapoints

Hi Liv, and once again thank you for forwarding this product idea from Celanese.

I have (for now) asked @Glen Sykes to process the idea for consideration. As a heads-up, since the essence of this request spans different product components, it may be somebody other than Glen who responds / follows up.

If there is a data gap in a time window, the end user should be able to configure whether to interpolate the data or have the data as is, coming from the timeseries data source. This is of importance during root cause analysis, in scenarios such as a faulty transmitter leading to subsequent/major events.

FYI: This is an issue raised by AkerBP as well, leading to a lot of confusion on how to trust data. Knut Vidvey, Jørgen Tennøe og Terje Løken informed about this issue as part of how CDF can approach data quality. 

Thanks for the tag @Thomas Sjølshagen .  I am actively investigating the solution options to satisfy this requirement, and it carries a high priority on my product backlog.

@rsiddha and @Benjamin Medbøen, I’ve reached out to Knut already to understand the workflow in more detail, about how the quality scores would be used and the further actions triggered, but I would value first hand feedback from you if you would be happy to spend a little time with me on this?

For sure, schedule a meeting when possible. I can also share with you the Data Qulity assessment done for AkerBP which highlights this issue 

Sure, Please schedule this and include  @ibrahim.alsyed and @Robert from Celanese as well, along with me. 

Hello All,

I am pleased to confirm that this feature is now available in Beta, please refer to our updated introduction page on the developer portal, and our API documentation for further details.  The feature will be formally announced in our release communications on the 5th March.  We hope to bring this feature to full General Availability level of maturity in our subsequent release, depending on good and timely customer feedback.

In order to ingest Time Series data points with status codes, you will need to update your OPC-UA or PI extractor to the latest beta version. 

The API updates should not affect any existing applications or integrations and it is possible to have a mix of old data points (where bad and uncertain quality data is missing) and new data points with status codes.  The API will continue to behave the same by default.

An update to Cognite Charts to support status codes in how Charts are plotted is in the works, and a separate update will come from the Charts team nearer the time when that will be ready.