GridCORTEX Live · Deregulated Market Special  ·  ← All Demos
On this page What you are watching The business case Run this at your utility Where you see it and how you say yes

The Scarcity Week Synthetic Data · Simulation

A retail energy provider in a deregulated, ERCOT-style market, BrightPlain Energy, fictional, 240,000 customers, heads into a heat-dome week. The hedge book covers 1,560 MW; the AI forecast says the real peak is 1,840 MW. That 280 MW gap, priced at a $5,000/MWh cap, is the difference between a good week and a company-ending one. Watch the desk close the position before the market prices the heat, activate a churn-safe demand response portfolio, and ride the scarcity hour with zero open exposure, while the competitor's customers get the panic email.

MON 06:00
POSITION REVIEW
⏳ DECISION POINT: TIME SLOWED
SYNTHETIC DATA
Baseload hedge Shaped block Forward buy (Mon) DR / VPP AI load forecast Open position / spot price

Same heat wave. Two very different Fridays.

What forecast-driven hedging is worth when the cap is $5,000
,
Week P&L vs plan
,
Open position at peak
,
Customer churn (week)
The Position
WithoutWith GridCORTEXΔ
The Business
WithoutWith GridCORTEXΔ
Illustrative simulation on synthetic data, prices, positions, and P&L are placeholders. In a GridCORTEX pilot, the forecast and hedge optimizer run on YOUR meter data, YOUR book, and YOUR market, backtested against your worst historical week. See UC 21.2 "Demo and Proof Plan."
280 MW
Open position (Thu-Fri peak)
1,840
AI forecast peak MW
$0.0M
Week P&L vs plan
0.2%
Churn risk (week)
Desk Feed, trading · retail · human-in-the-loop
Mon
Tue
Wed
Thu
Fri
The Validated Use Cases Behind This Scenario
UC 21.2
Load Forecast & Hedge Optimizer
AMI-granular, weather-driven forecasting fused with the hedge book, the open position seen days before the market prices it.
UC 21.1
Churn Prediction & Retention
Scarcity weeks are churn events: bill-shock and poaching risk scored per customer, retention offers targeted before the exodus.
UC 6.4
VPP Orchestration
The retail battery and thermostat fleet dispatched as the cheapest peaking resource the REP owns.
187 UCs
One Framework
The Scarcity Week is one of 187 validated use cases across 10 solution areas and 23 utility domains.
Inside the Demo
What you are watching, and what it proves

BrightPlain Energy is a fictional retail energy provider, a company that buys power on the wholesale market and sells it to 240,000 homes and businesses. It operates in a Texas-style deregulated market, where retailers compete for customers and wholesale prices can legally spike to a cap of $5,000 per megawatt-hour, roughly a hundred times a normal price. The simulation runs one extreme-heat week, Monday 06:00 through Friday midnight. The Monday position review sets the stakes. The company has bought 1,560 megawatts of power in advance to cover what its customers should use: 1,200 megawatts of steady around-the-clock supply plus 360 shaped to the daily usage curve. The forecast vendor's weekly email says the week's peak demand will be 1,610 megawatts. The GridCORTEX forecast, built meter by meter from smart-meter data and street-level weather under the building heat dome, says Thursday peaks at 1,840 megawatts, and with 90% confidence no worse than 1,905, the hottest Thursday in nine years. The difference is a 280 megawatt gap with no power bought to cover it, and at the $5,000 cap a single uncovered peak hour costs $1.4 million.

Three decision points follow, each executed by people: the trading desk makes every trade, the portfolio team runs every activation. Monday morning, close the gap now: the heat dome is visible in weather models but not yet in market prices, so power for Thursday and Friday afternoons (13:00 to 21:00) still sells at a $142 per megawatt-hour average. The desk buys 240 megawatts for $0.68M, cutting the uncovered gap from 280 to 40 megawatts. By Tuesday the same block trades at $310, and Monday's purchase is already $1.1M ahead. Wednesday, activate the customer portfolio: 92,000 enrolled smart thermostats (pre-cooling homes, then easing them 2 degrees, with bill credits sized per customer group) plus contracts with large business customers to cut usage on request. Together they take 118 megawatts off the Thursday-Friday peak at an equivalent cost of $71 per megawatt-hour, designed so customers stay comfortable and stay customers. Thursday at 15:40, with the live price at $4,900, the scarcity-hour call: the last 22 uncovered megawatts can be bought at essentially the cap (about $0.33M), or covered by the company's own coordinated fleet of customer batteries and precision usage cuts, thousands of small devices acting together like one power plant, for about $0.02M. The fleet is dispatched; zero megawatts are bought at the cap.

The ending, with all three approvals: Thursday's 16:00 to 19:00 scarcity window and Friday's second extreme afternoon pass with zero uncovered demand, every scarcity hour supplied below an effective $150 per megawatt-hour, week profit at $1.9M above plan, customer losses held to 0.4%, retention offers accepted by 61% of the at-risk customers they reached, and the risk committee receiving a complete decision log on Monday, recorded step by step in Relay, the built-in audit trail. Ignore the recommendations and the same weather produces a $21.4M loss: the desk discovers the gap live on Thursday, buys 180 megawatts at a $4,100 average while the rest rides the cap, and a $14M collateral demand from the market's clearing house arrives Friday morning.

Without GridCORTEX

The vendor forecast lands in one inbox and the record of what power the company has bought lives in the trading system, so the 280 megawatt gap exists for days before anyone computes it. Nothing is bought while the price is low; by Tuesday the block costs $310 and the desk is still watching. On Thursday at 14:00, with live prices at $3,800 and rising, the desk buys what it can, 180 megawatts at an average $4,100, and the rest rides the cap through 10 exposed scarcity hours at an average supply cost of $2,340 per megawatt-hour, against retail prices set assuming a fraction of that. The market's clearing house demands $14M of additional collateral by Friday 10:00. Bill-shock projections leak onto social media, the retention call center queue hits 4,000 calls, and 2.8% of all customers leave, because frightened customers shop around. Week result: a $21.4M loss.

With GridCORTEX

The software compares the meter-level demand forecast against the company's power purchases continuously (use case UC 21.2), so the gap is visible Monday at 07:00, three days before the market prices in the heat, and closing it costs $142 instead of $4,100 per megawatt-hour, 29 times cheaper. The device-fleet coordinator (UC 6.4) turns enrolled thermostats and batteries into the cheapest peak-hour resource the company owns, 140 megawatts of peak reduction. And the customer-loss model (UC 21.1) runs during the event, replacing the mass panic email with targeted credits that 61% of recipients accept. People stay in charge: the desk executes every trade and the portfolio team owns every activation. The week ends $1.9M above plan, a $23.3M swing, with 0.4% customer losses and no collateral demand.

The key numbers (KPIs), side by side
KPIWithout GridCORTEXWith GridCORTEXDelta
Open position into Thu peakexpected customer demand with no power bought to cover it; every uncovered megawatt rides the spike280 MW0 MWfully covered
Gap detectedwhen the company learned its purchases would not cover the heat waveThursday, liveMonday 07:003 days of warning
Forward cover costthe price paid to buy the missing power, in advance versus in a panic180 MW @ $4,100 avg240 MW @ $14229× cheaper
DR / VPP contributionDR (demand response) pays customers to use less at peak; a VPP (virtual power plant) is thousands of small customer devices coordinated by software to act like one power plantnone140 MW peakthe customer base acts as a power plant
Scarcity hours exposedhours of cap-level prices faced with demand still uncovered10 hrs0 hrs−10
Effective supply cost (Thu-Fri peak)the average cost of every megawatt-hour supplied to customers during the peak$2,340/MWh$147/MWh−94%
Forecast basiswhat the demand forecast is built from; AMI means the smart meters on every homevendor weekly emailAMI + km-scale weathermeter-level detail
Week P&L vs planthe week's profit and loss against budget−$21.4M+$1.9M+$23.3M
Margin / collateral callcash the market's clearing house demands upfront to cover a losing trading position$14M scramblenonetreasury slept
Customer churn (week)the share of the 240,000 customers who left during the week2.8% of book0.4%~5,800 customers kept
Save-desk call volumecalls into the retention center from customers threatening to leave4,000 callsbaselineno panic
Retention offer conversionthe share of at-risk customers who accepted a targeted credit and stayedn/a, blast email61% targetedtargeting beat the mass email
Risk committee packagethe decision record the risk committee reviews after the eventreconstructed in ExcelRelay-traced decision logthe audit trail is built in
Next scarcity event posturehow ready the company is for the next price spikesame blindnesssame system, better priorseach event makes it smarter
Live numbers (KPIs) on the dashboard
Open position (Thu-Fri peak)The megawatts of expected customer demand still uncovered heading into the peak. Zero is the goal. It starts at 280 and steps down to 40, then 22, then 0 as the advance purchase, the customer usage-reduction program, and the device-fleet dispatch are each approved. Anything still open when prices hit the cap is pure loss.
AI forecast peak MWThe forecast peak demand of 1,840 megawatts, built from smart-meter data and street-level weather, with 90% confidence of no worse than 1,905. It contradicts the vendor's 1,610 estimate, and measured against the 1,560 megawatts already bought, it reveals the 280 megawatt gap the whole week turns on. A forecast that only matched the vendor's would leave the desk blind.
Week P&L vs planThe running profit and loss against budget. Climbing toward $1.9M above plan means the cover is working; collapsing toward a $21.4M loss is what happens when 280 uncovered megawatts ride $5,000 prices.
Churn risk (week)The share of the 240,000 customers at risk of leaving. Around 0.4% is a normal week, and the targeted credits hold it there; 2.8% is the bill-shock panic, roughly 5,800 customers lost in five days.

The Business Case: Safety, Hours, and Cost

A utility does not buy a demo. It buys a safety exposure that goes away and a cost that goes down. Below is that case for every use case behind The Scarcity Week, written the way a plant manager, a safety lead, and a CFO each need to read it. Every hour and every dollar is a formula you run with your own rates and volumes. There are no vendor benchmarks in here and no invented percentages. If a number is not yours, it is not a number.
UC 6.4 Virtual Power Plant Orchestration and Optimization

What happens today, without this

The VPP, or virtual power plant, program director runs an event by logging into each vendor portal in turn: one for thermostats, one for residential batteries, one for the commercial storage pilot. Each schedule is set separately, the combined delivered megawatts is a guess made by adding the vendors' own optimistic numbers, and opt outs are tracked by refreshing several dashboards during the event. After the event an analyst pulls a comma separated file out of each platform and stitches them together in a spreadsheet to report performance against the demand response commitment, which takes most of a week each month.

What it replaces or shrinks

  • Logging into each vendor platform separately to configure and launch the same event
  • The manual estimate of combined delivered megawatts assembled from each vendor's own numbers
  • Per platform pre event health checks done by clicking through dashboards
  • Spreadsheet stitching of vendor exports into a monthly program performance report
  • Shrinks the staggering decision, when to lead with batteries and when to call thermostats, to a reviewable recommendation rather than an intuition

Why it is safer

The safety effect is indirect. A VPP that actually delivers what it promised on a peak day reduces the emergency operations that follow when it does not: manual load transfers, callouts, and field staff working a hot afternoon into the night.

Counted in units you already track:

  • Switching operations performed for emergency load transfer on peak days
  • Road miles driven for peak day field callouts
  • Night driving hours for staff recalled during and after evening peak events

Man-hours it gives back

Event day hours come back to the program director and the monthly reporting week comes back to the program analyst.

HOURS AVOIDED PER YEAR = events per year x vendor platforms x minutes per platform to configure, monitor, and stand down, plus programs x monthly reporting hours per program x twelve, minus the review time the director spends approving each combined dispatch plan.

The numbers we need from you to run that formula:

  • Events called per year and the number of vendor platforms in the portfolio
  • Minutes spent per platform per event on configuration and monitoring today
  • Monthly hours spent building the performance report per program
  • Enrolled capacity by program and your commitment obligation in megawatts
  • Loaded hourly rate for the program director and for the program analyst

Where the dollars come from

Cost driverHow it is calculated, from a rate you supply
Capacity value deliveredincremental megawatts delivered against commitment x your capacity price or your own avoided capacity cost
Program laborevent and reporting hours avoided x your loaded program staff rate
Underdelivery exposureyour penalty or shortfall charge per megawatt x the shortfall megawatts you currently incur in a typical season
Market revenuemegawatt hours bid into energy or ancillary products x the settled price in your market, for the hours the portfolio was previously idle
Incentive efficiencyyour incentive payment per enrolled device x the devices you no longer need to call because the dispatch is better ordered

Reliability and maintenance

Reliability
This touches peak day reserve margin and the probability of emergency operations rather than SAIDI or SAIFI directly. A portfolio that reliably delivers its commitment is capacity you do not have to buy or build.
Maintenance
The pre event check across all platforms finds dead telemetry, offline devices, and stale enrollments before an event rather than during one. That converts platform housekeeping from a post mortem into scheduled work.

What else it moves

CustomerOrdering the call so that batteries lead and thermostats join later means fewer customers asked to be uncomfortable, which is the single biggest driver of program attrition.
ComplianceOne performance record across all programs, in the form the market operator and your regulator ask for, instead of four vendor formats reconciled by hand.
WorkforceProgram staff spend event day on judgment rather than on operating four user interfaces at once.

What it costs you, stated honestly

You pay for the orchestration platform, for an integration to each vendor DERMS dispatch interface, and for the contract work to get the vendors to expose those interfaces at all, which is often the slow part. Your program staff also need time to validate the combined forecast against a season of real events before they will offer against it.

How to build the payback case

Payback is usually dominated by the capacity value of delivering the commitment plus the penalty exposure you stop carrying. Program labor is real but it is the smaller line.

This is a planning model driven by your event counts, your capacity price, and your penalty terms. It is not a vendor claim. Re run it with the actuals from one full program season.
UC 21.2 Retail Load Forecast and Hedge Position Optimizer

What happens today, without this

At a retail electricity provider, an analyst rebuilds the book's load forecast in a spreadsheet from a single point weather forecast, applies a growth assumption and a flat churn rate across all customers, and nets it against the hedge portfolio once a week or once a month. The open position comes out as one number rather than a distribution, so nobody can say what the tail looks like. Extreme weather scenarios get built by hand off a historical shape, usually only when the board or a lender asks, and the answer takes days.

What it replaces or shrinks

  • The weekly spreadsheet that rebuilds the load forecast from a single point weather forecast
  • Manual netting of forecast load against the hedge portfolio to produce one open position number
  • The flat churn assumption applied across the whole book because cohort level churn is too much work by hand
  • Shrinks the extreme weather stress test built from scratch each time the board or a lender asks for it
  • Shrinks the hedge sizing conversation that starts from a monthly average shape rather than an hourly one
  • The month end reconciliation between forecast load and settled load done by hand in a workbook

Why it is safer

There is no field exposure in this use case and we will not manufacture one. This is a balance sheet risk case. The one genuine human exposure it touches is the risk desk working continuously through an extreme weather event, which your own timesheets already show.

Counted in units you already track:

  • Night driving hours by risk and supply desk staff commuting home after continuous event coverage, which your overtime records let you estimate
  • Road miles driven: unchanged, and no credit should be taken
  • Confined space entries, elevated work hours and energized area entries: zero, because nobody on this desk goes to a plant or a substation

Man-hours it gives back

Rebuild and stress test hours come back to the load analysts, and the open position becomes a daily product instead of a weekly project.

HOURS AVOIDED PER YEAR = position rebuild cycles per year x analyst hours per rebuild, plus stress tests run per year x analyst hours per test, plus month end forecast to settlement reconciliation hours x 12, minus the time the risk manager still spends reviewing each recommendation and the time the desk spends sizing and executing under its own limits.

The numbers we need from you to run that formula:

  • Customers and cohorts in the book, and the churn rate you currently apply
  • Position rebuild frequency and analyst hours per rebuild
  • Stress tests run per year and hours per test
  • Loaded hourly rate for a load analyst and for a risk manager
  • Your own settlement history and realized imbalance prices, for backtesting

Where the dollars come from

Cost driverHow it is calculated, from a rate you supply
Analyst laborrebuild, stress test and reconciliation hours avoided x your loaded load analyst and risk manager rates
Imbalance exposuremegawatt hours of forecast error reduced x your own realized imbalance price spread, taken from your settlement statements and not from any market study we run
Scarcity tailthe change in your tail loss, measured by backtesting your own positions against your own settled prices, against your own board approved loss tolerance
Hedge efficiencypremium paid for block hedges a shaped position would not have bought, priced at the costs you actually executed
Collateralmargin or collateral you can stop posting because the position is better shaped x your own cost of capital

Reliability and maintenance

Reliability
A retailer owns no wires, so SAIDI and SAIFI are not yours to move. The equivalent for you is continuity of service to your own customers: staying solvent and hedged through an extreme weather event is the difference between serving your book and mass transferring it to the provider of last resort.
Maintenance
The maintenance analogue is model maintenance. Forecast error is measured against settled load every month, so a cohort model that has drifted out of calibration is found on a schedule and re-fit, rather than being discovered in the month it costs you money.

What else it moves

CompliancePosition, limits and stress results are produced on a fixed cadence with an audit trail, which is what your risk policy, your lenders and the market operator's credit desk each ask for in their own format.
CustomerA retailer that survives an extreme weather event keeps serving its customers at the price it promised instead of handing them to a default provider.
Insurance and riskA quantified tail, backtested on your own worst month, is a materially better conversation with your board, your credit providers and your counterparties.

What it costs you, stated honestly

You pay for the scoped engagement that builds and runs this, for weather ensemble data, and for integration into your energy trading and risk management system and your settlement data feeds, plus your risk manager's review time. The desk still owns every trade: the recommendation arrives as a draft ticket, and GridCORTEX holds no market access and executes nothing.

How to build the payback case

Payback is dominated by imbalance exposure and tail loss, both computed from your own settlement statements, and the analyst labor line is real but small. Backtest on your own worst month before you size the program, because that is the only number your board will believe.

This is a planning model built from your book, your settlements and your own risk tolerance, not a vendor claim, and no forecast removes tail risk. Re-run it with your actuals after one full season, including a hot month and a cold snap.
UC 21.1 Retail Energy Churn Prediction and Retention Engine

What happens today, without this

The retention team works from a renewal calendar and a monthly churn report that describes what already happened. An analyst pulls the list of contracts expiring in the next sixty days out of the billing system, and the offer is a discount picked off a small grid, because there is no way to tell who was actually leaving. Customers who were going to renew anyway get the discount too, straight out of a margin measured in cents per kilowatt hour. When a customer calls to cancel, the save desk representative has a script and no idea what the account is worth. Nobody connects the customer sitting on the wrong plan for their usage today to the customer who leaves four months from now.

What it replaces or shrinks

  • The renewal calendar list pull as the way at risk accounts are identified
  • The flat discount grid, replaced by an offer priced against the account's own margin and save probability
  • Shrinks the analyst work of building and reconciling monthly churn reporting, because cause attribution arrives with the score
  • The manual review of interval data for plan to usage mismatch, which today only happens after a customer complains
  • Shrinks the save desk representative's guesswork, because the account arrives with its value and its likely reason for leaving

Why it is safer

This is a software product for a retailer that owns no wires, so there is no field exposure to remove and we will not invent one. The only honest physical connection runs through arrears: customers on a plan that fits their usage fall behind less often, and fewer disconnect for nonpayment requests means fewer trips by the wires company's technician to a meter, which is a visit that carries real confrontation risk. That benefit lands on the utility, not on you.

Counted in units you already track:

  • Road miles driven by the wires company technician on disconnect and reconnect for nonpayment orders you would otherwise have requested
  • Energized area entries at the meter for those same disconnect and reconnect actions
  • Beyond that arrears path no standard field exposure unit moves, and we are not going to list ones that do not

Man-hours it gives back

Analyst and save desk hours come back, and the retention team stops building lists and starts running campaigns.

HOURS AVOIDED PER YEAR = renewal cycles per year x analyst hours per cycle building and reconciling the at risk list, plus 12 x analyst hours per monthly churn report, plus inbound cancellation contacts per year x minutes saved per save desk contact x your loaded save desk rate, minus the hours your commercial team spends reviewing and approving each campaign and its offer economics.

The numbers we need from you to run that formula:

  • Active accounts, monthly churn rate by segment, and average customer tenure
  • Gross margin per account per month by product and segment, in cents per kilowatt hour or in dollars
  • Your current retention offer grid, the average discount granted, and how many of those go to accounts that would have renewed
  • All in cost to acquire a replacement customer, including channel, broker, and enrollment costs
  • Analyst hours per renewal cycle and per monthly churn report, and loaded rates for an analyst and a save desk representative

Where the dollars come from

Cost driverHow it is calculated, from a rate you supply
Retained gross marginaccounts saved above your current baseline x your gross margin per account per month x your expected remaining tenure in months
Avoided acquisitionaccounts saved x your all in cost to acquire a replacement customer
Discount efficiencyyour average retention discount x the number of safe renewals that no longer receive one, which is the line that usually pays for the whole thing
Analyst and save desk laborlist building hours, churn reporting hours, and save desk minutes avoided x your loaded rates
Bad debtyour write off rate x the balance that no longer ages on accounts moved onto a plan that fits their usage, using the share you attribute to plan mismatch

Reliability and maintenance

Reliability
There is no reliability metric here, because you do not operate the wires. The equivalent measure is book stability: churn rate, average tenure, and the volatility of your forward load position, which is what your wholesale desk is hedging against every month.
Maintenance
The maintenance burden is model upkeep, not plant. Retention models drift when the market rate environment moves, so this has to be backtested against your own book each quarter and again after any significant move in market offer prices.

What else it moves

CustomerCustomers on a plan that matches their usage complain less and refer more, and the mismatch is found in interval data rather than in a complaint.
ComplianceOffer decisions are logged with the reason behind them, which matters in markets where regulators examine whether retention offers were made evenhandedly across customer groups.
WorkforceSave desk representatives get the account's value and its likely reason for leaving on screen, which turns a scripted call into a real conversation.

What it costs you, stated honestly

You pay for the GridCORTEX predictive service, for integration to your billing system, interval data feed, customer relationship system, and market price data, and for your own analyst time to define margin correctly and run the backtest on your own book before anyone acts on a score. If you cannot state gross margin per account today, that is the first project, not this one.

How to build the payback case

Payback is usually carried by discount efficiency, meaning the offers you stop giving to accounts that were never leaving, rather than by saved accounts, because saves are the number your finance group will argue about. Prove it with a holdout control group, not with a before and after comparison.

This is a planning model built from your margin, churn, and acquisition cost, not a vendor claim. Re-run it on the backtest against your own book, and then again on the holdout results from your first campaign.
Each of these opens in full on the use case page, alongside the integration plan, the data ask, the path to production, and the operator console. Open the use case library.
For Your Architects and Data Owners
Run this at your utility

What is this, exactly? It is AI software: intelligent agents and models built and delivered by SoftServe, running on NVIDIA accelerated computing. It is not a hardware appliance and it does not replace the systems you run today. It deploys in your own cloud or on your premises, connects read-only to your existing systems, and recommends; your people approve every action, starting in shadow mode until it earns trust.

A forecasting and risk service for a competitive retailer's supply and risk desk. It produces cohort-level load forecasts from weather ensembles, projects the book with churn adjustments, measures the open position against hedges daily, recommends shaped hedges, and runs extreme-event stress tests, as a daily position report and recommendations. The demo above uses synthetic data; everything below describes what the real deployment needs from your organization.

Systems it connects to

Your systemTypical productsHow we connect
Energy trading and risk (ETRM)ION Allegro, openlink Endurscheduled file export (CSV or CIM XML)
Customer Information System (CIS) / retail billingretail billing platformsdatabase replica refreshed nightly
Market and grid operator interfacesERCOT or other ISO settlement dataread-only API
Weather and environmentNational Weather Service feeds, commercial servicesread-only API
Customer relationship and service systemsSalesforce, Zendeskdatabase replica refreshed nightly
Retail market transactionsEDI switch, drop, and enrollment recordsscheduled file export (CSV or CIM XML)
Metering (AMI usage data)Smart Meter Texas or utility-provided interval usagescheduled file export (CSV or CIM XML)

Data it needs from you

How it runs on your systems

Runs in the retailer's own cloud account with GPU instances. Position and book data is commercially sensitive and stays in your environment with need-to-know access. All connections are read-only; the desk executes every trade through its existing brokers.

Path to production

Weeks 1-4
Connect the ETRM, book, and settlement data; ETRM extract access is the usual gate.
Weeks 5-10
Backtest the worst recent month: reconstruct the prior book and hedges, run the ensemble and recommended position, settle both in dollars.
Weeks 11-12
The desk reviews the dollar comparison; go or no-go.
Months 4-5
Harden the daily pipeline, add open-position alerting, and train the desk.
Months 5-7 onward
In production: the desk starts each morning with the position report and ensemble forecast; stress tests run each season.

What we need from your team

Full integration, data, and timeline detail for each use case in this scenario: UC 21.2 · UC 21.1 · UC 6.4
For Your Operators and Dispatchers
Where you will see it and how you say yes

The Approve button you just clicked in the demo above is the real workflow. This is what it looks like on the screen of the retailer's supply and risk desk lead in the GridCORTEX console:

GridCORTEX ConsoleSigned in: the retailer's supply and risk desk lead
Notifications
Open position 340 MWh short for August peaks; 5% tail scenario costs $2.7M unhedged; shaped hedge recommended
Daily model refresh complete; all connected feeds healthy
Recommendation
Send the recommended 340 MWh shaped hedge to the desk
  • Daily netting shows 340 MWh open in peak hours
  • Uri-class stress shows $2.7M exposure at the 5% tail
  • Worst-month backtest: recommended position cut losses 64%
✓ Send hedge to deskModifyDecline
After you approve: The recommendation posts as a ticket in the trading and risk system, where the desk executes under its own limits, and an audit entry records who approved it and why.
Computed from data as of 17:42:10 local; every card shows the timestamp of the data behind it.

What happens when you hit approve

Approve posts the recommendation as a draft ticket in the trading and risk system; the desk sizes, executes, or declines it under its own limits and controls. GridCORTEX never trades and holds no market access.

How you tell it what it cannot see

Position measurement is automatic from trading, billing, and market feeds; the desk types nothing. A deal done by phone flows in once the desk books it in the trading system.

Live data, not stale data

The book and hedge position net daily against overnight billing and trading system syncs, with weather ensembles each cycle; every report shows the as-of timestamp of its positions and forecasts.

Where it lives day to day

Lives as a daily position report in the GridCORTEX console beside the trading system; a position beyond limits pushes a mobile notification to the risk officer. The console runs in a browser beside your existing screens on day one; embedding into your own systems is a roadmap step once the read-only phase has earned trust. Approve, Modify, and Decline are all captured in an audit trail your compliance team can pull, and GridCORTEX never blocks or overrides anything in the systems you run today.

The Gap: Why Your Existing Systems Don't Already Do This

The fair question from any REP: "We have an ETRM, a forecasting vendor, and traders who lived through the last one, what's new here?" Here's the honest answer.

What you own keeps doing its job

  • The ETRM, book of record for every hedge, trade, and settlement. Nothing changes.
  • Forecasting services, daily load forecasts keep arriving; good ones become an input.
  • Your traders, every position decision stays human. The desk approved both moves you watched.
  • CRM & billing, customer relationships and invoicing run as always.

The gap GridCORTEX fills, above them, not instead of them

  • The forecast and the book never meet. The vendor forecast lands in one inbox; the hedge book lives in the ETRM. The 280 MW gap existed for days before anyone computed it, fusing them continuously IS the product.
  • AMI-granular beats system-level. A forecast built per-meter from smart-meter history and km-scale weather sees the heat dome's load three days before a top-down model prices it, that's the window where the forward buy was cheap.
  • DR as a hedge, priced like one. The thermostat fleet was dispatched because its cost-per-MW beat the forward curve; a portfolio decision no DR platform makes on its own.
  • Churn and scarcity are the same event. The retention model runs DURING the market event (bill-shock scoring, targeted credits) not in next month's marketing review.
  • Speed is the position. Intra-day re-optimization as prices move: the Thursday VPP-vs-buy call happened at 15:40 for a 16:00 spike.
Accent, don't replace: GridCORTEX reads your meter data, your ETRM book, and the market feeds · computes the position and the cheapest close continuously · and hands orders to your traders and dispatches to your DR platform. The desk decides. It just finally sees the whole board.
Under the Hood: What GridCORTEX Took Into Account in This Scenario

When someone asks "what did it actually calculate?", this is the list. In the simulation these factors drive the storyline; in a pilot they are computed from your meter data, book, and market.

📈 The Forecast

  • Per-meter load models from AMI interval history, 240,000 customers clustered by usage shape, weather sensitivity, and appliance signatures
  • Km-scale weather physics (Earth-2 class) under the heat dome, which neighborhoods hit 41°C and when, not one city-wide temperature
  • Probabilistic peak: P50/P90/P99 bands, so the desk hedges a distribution, not a single number
  • Intra-day re-forecast as actuals arrive, the model corrects itself every hour

💼 The Position

  • Hedge book decomposition: baseload blocks, shaped products, options, mapped hour-by-hour against the forecast bands
  • Open position valued at forward AND at scenario spot prices, what the gap costs if the market goes to cap
  • Optimal close: forward block vs. call options vs. DR activation vs. VPP dispatch, ranked by cost per MW of cover
  • Credit and collateral impact of every path, the margin call that never happened

🌡 The Portfolio as a Power Plant

  • 92,000 enrolled thermostats, retail battery fleet, and C&I curtailment contracts, dispatchable MW with per-customer comfort and fatigue constraints
  • Churn-safe activation: bill credits sized per segment, opt-out friction minimized, message timing tested
  • VPP dispatch against the 16:00 scarcity hour; 22 MW at marginal cost instead of $4,900 spot

🤝 The Customer Book

  • Bill-shock prediction per customer under the event, who sees a doubled invoice next month
  • Churn scoring live during the event: competitor pricing, contract end dates, complaint history
  • Targeted retention: credits and communications to the top-risk decile, not a blast email
  • Runs on the NVIDIA Agent Toolkit, always-on watch over book vs. forecast, every recommendation Relay-traced for the risk committee

Presenter's one-liner: "The forecast saw the heat three days before the market priced it, the desk closed 280 megawatts while it was cheap, the thermostat fleet became the cheapest peaker in the portfolio, and the churn model protected the book while the competitor blasted panic emails. Same weather, plus twenty-three million dollars. That's what you just watched."

GridCORTEX Live Scenario Demo · Synthetic data throughout; BrightPlain Energy is fictional; no REP, market, or price event depicted is real · SoftServe + NVIDIA · Created by Ronnie Mauldin, NVIDIA Solutions Director, Power & Utilities, SoftServe · JUL 2026