Data Engineering for Energy: SCADA, AMI, OMS, and GIS Into One Truth

A utility starts a data platform initiative and discovers that connecting its systems is architecturally harder than anyone expected. SCADA data arrives as high-frequency time series (seconds to minutes) with point names that map to assets through the SCADA configuration database. AMI data arrives as interval readings (15-minute or hourly) with meter IDs that map to premises through the CIS. OMS data tracks outages by device with timestamps that need to be aligned with SCADA events. GIS holds the spatial and connectivity model with asset IDs that may or may not match the asset registry. The CIS holds customer and account data. The work management system tracks maintenance activities against assets using yet another ID scheme. Joining these across systems requires master data alignment for assets, geographic locations, and customers — and that alignment doesn't exist in any single source system.

Utility data engineering done right starts with the asset and geographic master data. An asset registry that maps SCADA points, GIS features, OMS devices, AMI meters, and work order assets to a common identifier. A geographic hierarchy from system to substation to feeder to transformer to meter. With these aligned, SCADA time-series, AMI intervals, OMS outages, GIS spatial data, and work history can all be joined at the asset and feeder level — which is where the operational questions live. Bronze-silver-gold medallion with time-series handling for SCADA and AMI, event processing for OMS, and spatial handling for GIS. Reconciliation against source systems. Done with this discipline, the utility gets the cross-system analytics it needs. Done as generic ETL, it produces a platform that can't answer feeder-level questions.