Basira"Can we just see our PI data in Power BI?" is one of the most common requests we get from Omani operators. Power BI is where leadership already lives — but it doesn't speak PI natively, so the connection is always a small engineering decision.
Why can't Power BI read PI directly?
PI stores compressed, irregular time-series data optimised for historians, not the tidy tabular shape Power BI expects. Pulling raw tags straight into Power BI tends to be slow, heavy, and fragile. The fix is a thin layer in between that summarises and shapes the data first.
What are the realistic options?
- PI Web API → a scheduled pipeline that writes summarised data to SQL/cloud, which Power BI reads. Most robust and the one we usually recommend.
- PI Integrator for Business Analytics → AVEVA's own product that publishes PI data to a target Power BI can consume. Less custom code, more licensing.
- Direct PI Web API query from Power BI → quickest to stand up, but fragile and slow at scale; fine for a small proof of concept only.
What makes a PI → Power BI dashboard actually trustworthy?
- Pre-aggregating at the right interval so refreshes are fast and cheap.
- Mapping PI AF assets to clean, business-friendly names.
- A refresh cadence matched to the decision, not the sensor.
- Monitoring on the pipeline so a silent failure doesn't become a wrong number on a board slide.
Done right, the dashboard is the easy part — the durable pipeline underneath is the real work, and it's what we build.