Closing SCADA Data Gaps with Smart Infilling

Water utility teams rely on SCADA data to keep systems running smoothly, make confident decisions, and report accurately.

But when gaps appear — from sensor faults or telemetry dropouts — even the most capable teams can be left with uncertainty.

These issues don’t always require complex fixes, just smart, careful handling.

With the right approach, gaps can be filled to support infrastructure planning, reliable reporting, and billing for water exported to other systems.

Staying Ahead in Non-Revenue Water Reporting

Accurate Non-Revenue Water (NRW) reporting depends on complete flow data. But things don’t always go to plan.

A failed meter might leave a two-week hole in your inflow data - just when you're preparing reports.

In some teams, these gaps are patched manually.

Others leave them blank - meaning data from that period is excluded from analysis.

Both approaches introduce risks - from delayed reports to increased uncertainty.

SensorClean has targeted logic for short-gap interpolation grounded in real-world system behaviour and is applied in a matter of minutes – not days.

And the impact doesn’t stop at NRW reporting.

Protecting the Integrity of Peak Demand Events

Pipe failures often occur during off-peak hours — when demand is low and pressure is high.

If a burst happens at say 2am just before a real peak day event, the resulting spike can distort important metrics.

Firstly, engineers don’t want to include the lost water as part of the peak day estimate.

Secondly, engineers don’t want to lose the ‘real peak day’ data that occurred following the morning pipe burst.

To address this, SensorClean applies best-evidence data infilling to restore integrity of the peak day demand data – while quantifying the lost water for NRW reporting.  

Peak Flows are reviewed through a QA process to ensure transparency and confidence for infrastructure planning.

As utilities adopt AI to improve forecasting and anomaly detection, data quality becomes even more critical.

Forecasting with AI models

Clean, uninterrupted SCADA data enables smarter decisions about the future.

AI models for demand forecasting, pump scheduling, and anomaly detection rely on consistent, trustworthy input.

But even brief gaps or spikes from sensor dropouts can throw these systems off, distorting predictions or triggering false alarms.

Smart infilling ensures your data is model ready.

It gives AI tools the continuity they need to learn real patterns - and deliver timely, reliable insights you can trust.

Strengthening Your Data Infrastructure

Utilities are looking for data reliability.

Our role is to support your teams by making SCADA data cleaner, more complete, and easier to trust.

Want to strengthen your SCADA data infrastructure?

Reach out to explore how SensorClean can reduce uncertainty, improve reporting, and unlock forecasting — giving your team greater confidence in every decision.