The challenge
The organisation's technical leadership had committed to a £150M national grid reliability improvement programme — but the underlying reliability problem had a structural cause that capital investment alone would not address. A national electricity transmission operator managing 21,000 km of infrastructure and serving 6.2 million customers was experiencing 34 significant grid stability incidents annually, with 7 resulting in unplanned load shedding affecting hundreds of thousands of customers. The combination of reactive SCADA monitoring, manual inspection programmes covering only 61% of the network, and no predictive analytical capability meant that remote assets were routinely failing without warning. The programme needed a strategy that would use the investment to build genuine predictive capability, not simply replace existing reactive monitoring with newer reactive monitoring.
Discovery
An 18-week technical assessment identified the structural cause of the reliability problem: 39% of the transmission network was inspected less than once every two years, meaning assets in remote areas routinely deteriorated and failed without prior detection. SCADA was generating alerts reactively — recording events after they occurred, not predicting them. Twelve years of historical failure records contained the data needed to build a predictive capability; no systematic analytical programme had ever been applied to them.
Our approach
Diagnostic and assessment: completed an 18-week technical assessment of the transmission asset population, SCADA data architecture, field operations processes, and existing condition monitoring capabilities to define the programme strategy
Grid operations intelligence implementation: designed and built the unified grid operations intelligence architecture, integrating with 47 SCADA substations and 2,100 intelligent electronic devices across the transmission network
Asset failure prediction programme: built a transmission asset failure prediction model using 12 years of historical failure, maintenance, and condition data across six critical asset classes — transformers, circuit breakers, overhead lines, cables, protection systems, and substation equipment
Inspection coverage programme: designed and implemented a drone-based visual inspection programme with defect detection — tripling annual inspection coverage from 61% to 98% of the transmission network at lower cost than the previous manual programme
Grid stability monitoring: built real-time power flow analytics detecting stability risks 4–8 hours before conditions that historically led to stability incidents
Executive analytics layer: designed and built the operational analytics layer integrating SCADA, asset management, work order, and financial data — providing leadership with a unified view of network health, maintenance investment, and operational performance
The Outcome
The engagement delivered a two-layer intelligence architecture. The grid operations intelligence layer monitors operations in real time, detecting stability risks and SCADA anomalies and predicting conditions likely to lead to stability incidents hours before they materialise. The asset condition intelligence layer analyses sensor data, inspection reports, and historical failure data to build a continuously updated risk profile of every critical transmission asset. Together, they gave the national grid operator both operational situational awareness and strategic asset management intelligence — enabling the transition from reactive maintenance to a predictive operations model.
The results
“The grid reliability improvement in the first two years of deployment speaks for itself. But what has changed fundamentally is how we think about managing the network. We are not waiting for things to go wrong any more. We can see risk forming and act on it.”
