Grid Health & Outage Command Centres.
We build the visual layer for grid operations and its regulators: outage maps fed by OMS events and AMI last-gasp signals, geospatial network views drawn from GIS connectivity, and reliability dashboards where every SAIDI point traces back to the outage records behind it. For water operators, the same canvas approach renders catchment and DMA views with leakage, pressure, and CSO event-status overlays.
Fragmented Silos
Legacy Utility systems and disconnected feeds.
Storm-Mode Visualisation
One Network Picture
Production Reality
Geospatial Network Canvases
Geospatial Network Canvases
Storm-Mode Outage Views
Drillable Reliability Dashboards
Industry-Specific Friction Points
Storm Views Built for Blue Sky
Outage maps that work on a quiet Tuesday collapse in a storm, when thousands of concurrent events, crew positions, and estimated-restoration clocks must render at once without hiding the feeders that matter most.
Three Rooms, Three Pictures
The control room watches ADMS, field crews see the work-management queue, and executives get yesterday's slide. During a major event there is no shared visual truth of the network — only competing versions of it.
Dashboards Regulators Can't Interrogate
Reliability visuals summarise SAIDI and SAIFI but cannot drill to the underlying interruptions, so every regulator question becomes a manual data pull instead of a click-through to the source records.
How the Utility delivery flow works
This technical flow diagram reveals how Unolabs treats Utility data to deliver governed, production-ready outputs.
Source Layer
Network Canvas
Rendering feeders, transformers, and service-territory geography as the base layer for every operational view.
Industry Logic
Event Overlay
Overlaying live outage events, meter last-gasp clusters, and restoration progress onto the network canvas.
Reliability Views
Linking SAIDI/SAIFI summary visuals to the interruption records and exclusion logic beneath them.
Activation
Ops Publish
Publishing role-tuned views for dispatchers, storm rooms, field tablets, and executive briefings from one visual model.
How the work is engineered for Utility
Storm-Mode Visualisation
We design outage views that degrade gracefully under event load — clustering incidents, prioritising by customers interrupted, and keeping estimated-restoration confidence visible as conditions change.
One Network Picture
We render the GIS network model as the shared canvas, layering AMI signals, outage events, and crew status on top, so operations, field, and leadership all look at the same grid.
Drillable Reliability Reporting
We build regulator-facing dashboards where every aggregate reliability metric opens into its constituent interruptions, exclusion decisions, and data lineage.
Where the Real Work Is
A Control-Room Display Is Not a Dashboard
Displays watched for twelve-hour shifts obey different rules than dashboards opened for ten minutes. Control-room HMI practice favours muted, low-saturation backgrounds where color is reserved exclusively for abnormal states, readability at distance in dim rooms, and zero decorative motion — because an interface that cries wolf trains operators to ignore it. Analyst dashboards, by contrast, reward density and exploration. We build both from the same governed marts but hold them to deliberately different design standards.
Feeder Detail at Territory Scale
Rendering an entire service territory down to conductor and service-point detail is a genuine performance engineering problem: millions of geometries, live event overlays, and the worst traffic arriving mid-storm when every stakeholder opens the map at once. We engineer for that hour — vector tiling, level-of-detail rules that collapse detail gracefully as the viewport widens, GPU-accelerated rendering, and event clustering — and load-test against storm-scale concurrency rather than a quiet demo afternoon.
Field Views Live in the Worst Conditions
The field tablet version of the network view operates in direct sunlight, in rain, in gloves, and in coverage dead zones — conditions that quietly disqualify most dashboard conventions. Touch targets must be large, contrast must survive glare, and the view must degrade usefully offline, caching the crew's assigned area and syncing when signal returns. Designing for the field first, rather than shrinking a desktop view, is what determines whether crews actually consult the tool during restoration.
Visible work products, not vague advice
Each deliverable is designed to be used by Utility architects, engineers, data owners, and operations teams after the engagement ends.
How We Measure Visual Adoption
A view that is not consulted during a real event has failed regardless of its craft. We baseline current tool usage and event behaviour early, then measure whether the new visual layer actually changes it.
Map responsiveness under event load
Interaction latency on the network canvas measured at storm-scale concurrency in load tests and during real events — quiet-day performance is the floor, not the benchmark.
Adoption across control-room shifts
Active use tracked per shift rather than in aggregate, because a tool the day shift likes and the night shift ignores has only half-landed.
Event-to-glass latency
Minutes from an OMS event or meter last-gasp cluster to its appearance on the shared map, tracked as a distribution so tail latency during storms stays visible.
Questions answered in-tool
Share of regulator and executive follow-ups resolved by drill-through during review sessions versus exported to manual data pulls — the practical test of whether drillable reporting works.
Frequently Asked Questions
Can one visual model serve the control room, storm room, and executives?
Yes — that is the design goal. The GIS-derived network canvas and its data layers are built once, and each audience gets a role-tuned view over them: abnormal-state-focused displays for operators, resource-and-restoration views for the storm room, and summarised drillable briefings for leadership. What differs is presentation, never the underlying numbers.
How does the outage map stay usable with thousands of live events?
Through explicit degradation rules decided in advance: incidents cluster spatially as density rises, ranking by customers interrupted decides what stays individually visible, and level-of-detail rendering sheds decoration before it sheds information. The map is load-tested at storm-scale concurrency, so its behaviour in the worst hour is engineered rather than discovered.
What technologies do you build utility visualisation on?
Deck.gl and GPU-accelerated web rendering over GIS layers for the geospatial canvases, and Power BI or Microsoft Fabric for analytical and regulatory reporting where estates already standardise there. The stack follows what your teams can operate and license — the storm-mode behaviour, HMI discipline, and drill-through wiring are the durable part.
How an engagement starts
Bring screenshots of the three screens your operators, field crews, and executives actually used during the last major event — in a working session we annotate where each one failed its audience and propose the first view worth rebuilding.
Interested in the full industry blueprint?
We have deeper technical documentation for Data Visualisation & BI Dashboards for Utility in the Utility sector.
Bring your outage map from the last storm. We'll show what real-time would change.