Predictive grid maintenance and outage forecasting.
18%
Op cost savings
60.7%
Grid uptime
10 wks
To production
240k
Endpoints monitored
Where operations were breaking down.
Distribution faults across the network were being detected only after customer outage reports came in. The result: degraded SAIDI metrics, regulator scrutiny, and crews dispatched reactively instead of preventively.
Asset health data was scattered across SCADA, AMI, and field-inspection records. None of it was correlated. Maintenance planning operated on age-based schedules, not condition-based signals, so healthy transformers were being serviced while at-risk ones quietly degraded.
Leadership needed a system that fused telemetry, weather data, and asset history into a single risk-scored view. It also had to trigger work orders into the existing CMMS without operators changing tools.
How Aelix GridIQ was deployed.
Multi-source telemetry fusion
SCADA, AMI, weather, and asset-history data unified into a single feature store keyed by feeder and substation.
Anomaly + risk models
Time-series anomaly detection on transformer load, voltage, and current patterns. Weather correlation flags storm-driven failure risk hours ahead.
Condition-based maintenance
Risk scores feed directly into the existing CMMS as prioritized work orders. Age-based schedules deprecated in favor of signal-driven dispatch.
Regulator-ready reporting
Automated SAIDI/SAIFI reporting with full audit lineage. Every dispatch decision traces back to telemetry that justified it.
From source data to operational action.
Stage 01
SCADA + AMI + weather
Real-time telemetry from the grid combined with NWS feeds and asset registry data.
Stage 02
Unified data lake
Streams normalized into a feature store keyed by feeder, substation, and asset class.
Stage 03
Anomaly + risk models
Anomaly detection plus weather correlation produces forward-looking risk scores per asset.
Stage 04
CMMS dispatch
High-risk assets generate prioritized work orders in the existing CMMS with full context.
Stage 01
SCADA + AMI + weather
Real-time telemetry from the grid combined with NWS feeds and asset registry data.
Stage 02
Unified data lake
Streams normalized into a feature store keyed by feeder, substation, and asset class.
Stage 03
Anomaly + risk models
Anomaly detection plus weather correlation produces forward-looking risk scores per asset.
Stage 04
CMMS dispatch
High-risk assets generate prioritized work orders in the existing CMMS with full context.
Operational gains after going live.
0%
Operational cost savings
Condition-based maintenance replaces age-based servicing of healthy assets.
0.0%
Grid uptime
Sustained over 18 months of production deployment across the service territory.
240k
Endpoints monitored
Transformers, switches, and field assets continuously scored for failure risk.
47 min
Avg lead time on faults
Storm-driven failure risk flagged ahead of customer outage reports.
0%
SAIDI improvement
Year-over-year reduction in system average interruption duration.
10 wks
From kickoff to production
Native CMMS integration meant zero changes to crew dispatch workflows.
See what Aelix GridIQ would do for your operations.
Talk to our team about the bottlenecks slowing you down, and the system we'd deploy to remove them.