This seems to be an interesting topic. Unfortunately its not a topic that I can contribute towards but one that I would like to learn more about.
Within Upstream O&G there is good focus on Energistics being applied within AF, especially when it comes down to Units Of Measure and WITSML.
Geopolitical semantics: I have seen 3166 and GAUL used in AF.
Some clients have well defined design and engineering departments that already map out industry standards to be adhered to within projects regardless of the system being implemented...probably too many to list out together in this thread. Not a huge emphasis on AF within those type of projects.........yet.
Is OSIsoft after particular examples from an industry, or is this more of a general discussion point?
Security within AF is something on the top of my pile right now so interested in others thoughts/discussions on that particular topic.
Hi Rhys @ Wipro,
Thanks for the input. I didn't know about GAUL and will look into it, thanks.
This was more a personal interest and curiosity than anything else. Certainly we would welcome more insight into what our customers are using and modelling against. Please, if you have more/particular examples per industry let's discuss them and study them.
As said, this is mostly a personal curiosity of mine. However, there are practical use cases and one that I'm very interested in:
within the context of the Rubik project, if we could "know" (or at least map to a standard model) what that prospective AF model represents, then we could possibly automatically suggest measures, KPIs, etc.
Of course, there are other more lofty use cases. I think that with an increase in semantic tagging and reliance on standards, more intelligent projections/automatic discoveries could be accomplish. For instance, it could facilitate interoperability between datasets.
Think about projecting PI data (customer specific models & data) onto eGoverment data and even scientific datasets. It would be pretty cool to automatically discover a correlation between instances of say a nominally/suspected unrelated vandalism that are geographically clustered and spikes somewhere in a process nearby (what if to create that graffiti people stomp on pipes that run through the area...). What if you're modeling a cattle ranch and notice that they aren't performing as before (maybe they're getting sick more often). Perhaps if you project this knowledge onto the state data you could notice that upstream there has been drought and there are corn + oats crops.
Hi Rick Davin,
Thanks for the feedback.
I too want to learn about this topic. During the UC, some presentations talked about interoperability, modeling, etc. and I thought it would make for an interesting discussion.