Load and generation forecasting
Forecasts demand and renewable output at the interval the market settles on, so positions and dispatch are priced against reality.
- Moves
- Imbalance cost per MWh
- Twin
- Grid Twin
- Capability
- Forecasting
Industries
05 — Energy & Utilities
Reliability and margin, in the same model.
Grid, asset and trading twins that price reliability, outage risk and flexibility together — with agents operating under NERC CIP and market-conduct rules.
North Star
Margin per MWh delivered reliably
The value engine, end to end
Discover → Model → Twin → Agents → Govern → Compound
The same six moves we run everywhere, expressed in the physics of this industry.
01
Discover
Where does reliability cost the most margin?
Reliability–margin map
02
Model
What drives outage, loss and imbalance?
Causal network model
03
Twin
What if we re-dispatch or defer capex?
Grid and asset twin
04
Agents
Who acts inside the control room boundary?
Advisory + execution agents
05
Govern
Is it CIP-compliant and auditable?
NERC CIP control file
06
Compound
Does SAIDI and margin both improve?
Outcome contract
Where the value is
Value pools and their drivers
Before a twin is built, we agree where the money sits and what moves it.
Value pool
Asset health
Driver
Failure risk, capex deferral
Measured as
Deferred capex per asset class
Value pool
Network losses
Driver
Technical and non-technical loss
Measured as
MWh recovered
Value pool
Trading & flexibility
Driver
Imbalance cost, flex dispatch
Measured as
Margin per MWh
Value pool
Customer ops
Driver
Billing exceptions, contact volume
Measured as
Cost to serve per meter
Use cases
Where value is already sitting in Energy & Utilities
Each one is a number somebody already owns, a twin that models it, and an agent that moves it inside a defined authority. Filter by function.
5 use cases
Outage prediction and crew dispatch
Anticipates where the network will fail under weather and load, and positions crews before the call volume arrives.
- Moves
- Minutes lost per customer
- Twin
- Grid Twin
- Capability
- Geospatial ML
Asset and vegetation inspection from imagery
Reads drone and satellite imagery for the spans that need attention this cycle instead of inspecting everything on a calendar.
- Moves
- Risk reduced per inspection dollar
- Twin
- Asset Twin
- Capability
- Vision models
Commodity position and hedge optimisation
Tests the hedge book against forecast, weather and regime scenarios, and shows what breaks first.
- Moves
- Risk-adjusted margin
- Twin
- Financial Twin
- Capability
- Simulation
Emissions accounting and abatement ranking
Builds an auditable emissions position by source and ranks abatement options by dollars per tonne, not by ambition.
- Moves
- Cost per tonne abated
- Twin
- Compliance Twin
- Capability
- Data lineage + optimisation
Outcome twins
Twins, levers, features and the agents that run them
Each twin owns a set of levers. Each feature moves one lever. Each agent executes inside a defined authority.
Grid Twin
Model the network state and price the next intervention.
Levers
SAIDI / SAIFILoss rateConstraint cost
Features
- — Outage prediction
- — Switching advisory
- — Loss detection
Multi-agent layer
Outage Agent
Pre-positions crews against predicted faults.
Asset Twin
Defer capex without buying reliability risk.
Levers
Failure probabilityReplacement timingInspection cost
Features
- — Health scoring
- — Capex sequencing
- — Inspection routing
Multi-agent layer
Capex Agent
Ranks and sequences asset interventions.
Trading & Flex Twin
Turn flexibility into a priced, dispatchable position.
Levers
Imbalance costFlex utilisationHedge efficiency
Features
- — Forecast ensembles
- — Dispatch optimisation
- — Position monitoring
Multi-agent layer
Dispatch Agent
Proposes flex dispatch inside risk limits.
Customer Ops Twin
Remove exceptions from billing and service.
Levers
Exception rateContact volumeDebt roll-rate
Features
- — Exception triage
- — Consumption anomaly detection
- — Contact deflection
Multi-agent layer
Billing Agent
Resolves exceptions before the bill goes out.
Unit economics
ROI calculator — per feature, per unit
Set your volume, switch features on or off, and move each impact to your own evidence. Everything is expressed per MWh.
MWh delivered per year
9,000,000
Contribution per MWh
$14
Per-MWh margin; substitute per-meter economics for retail supply.
Annual program investment
$3.5m
Grid Twin
Volume — Delivered MWh
2.0% — Fewer unserved hours from predicted faults.
$2.5m
annual value
Grid Twin
Leakage — Non-technical loss
18.0% — Recovers energy currently unbilled.
$10.9m
annual value
Asset Twin
Unit cost — Asset cost per MWh
3.0% — Defers spend where risk allows.
$22.1m
annual value
Trading Twin
Unit revenue — Realised price per MWh
1.5% — Better position against imbalance.
$13.0m
annual value
Customer Ops Twin
Unit cost — Cost to serve
1.5% — Billing exceptions resolved without a human.
$11.1m
annual value
Value by feature
Annual value contributed by each active feature at current settings.
Value and ROI vs volume
How engineered value and ROI move as mwh delivered per year change. The marker is your current setting.
Cumulative value vs investment
Where the curve crosses the investment line is payback.
Value mix by lever
Which lever the engineered value is actually coming from.
Baseline contribution
$126.0m
Engineered annual value
$59.6m
ROI on program
1602%
Payback
0.7 months
Directional model. In an engagement every number here is replaced by your measured baseline, signed off by finance, and written into the outcome contract before any agent goes live.
Scale
How this compounds from one unit to the enterprise
01
One feeder / one asset class
Prove predicted vs realised reliability value.
02
One region
Same twin, regional constraints and crews.
03
Enterprise
Single reliability-to-margin number across the network.
Integrations
Domain platforms we connect
Vendor-agnostic by design. Connectors are added per engagement — no platform lock-in.
Grid & OT
- GE GridOS
- Siemens Spectrum
- OSI monarch
- AVEVA PI
Asset & work
- IBM Maximo
- SAP PM
- Copperleaf
- Esri ArcGIS
Market & trading
- ETRM platforms
- Enverus
- ICE
- EPEX
Data & cloud
- Snowflake
- Databricks
- Azure
- AWS
Regulations and standards
What we are held to in this industry
Compliance is a design input, not a review gate. Every agent action is logged, attributable and reversible.
NERC CIP
Critical infrastructure protection across any agent touching OT.
NIS2
Resilience, incident reporting and supply-chain security.
EU AI Act
Safety-relevant grid AI: oversight, robustness, logging.
REMIT / market conduct
No agent action that could constitute manipulation.
GDPR
Smart-meter consumption data as personal data.
Standards supported
IEC 61850CIM (IEC 61968/61970)IEC 62443ISO 55000ISO 27001NIST AI RMF
Next industry
Public Sector