A founder-led AI company, Colossus Compute, walks in and says: "We want three AI campuses in your territory. As large as you can serve, as fast as you can serve them." One gigawatt, on a 3,200 MW system. Watch the utility run the full lifecycle with GridCORTEX: portfolio siting study in minutes, a staggered phasing offer, the regulatory filing with an auto-generated evidence pack, onsite-generation terms that turn their batteries into your peak asset, long-lead procurement, construction, and three energizations, every commitment with a human in the loop.
| Without | With GridCORTEX | Δ |
|---|
| Without | With GridCORTEX | Δ |
|---|
Month 0: a fictional utility whose entire system peaks at 3,200 megawatts (MW) receives an inquiry from Colossus Compute, a fictional founder-led AI company: "We want three AI campuses in your territory. As large as you can serve, as fast as you can serve them." The ask is 300 + 450 + 250 MW, a full 1.0 gigawatt (1,000 MW), nearly a third of everything the utility serves today. The simulation clock runs 36 months and covers the whole journey: siting study, regulatory filing, contracts, equipment purchasing, construction, and three switch-ons. Within the first month, GridCORTEX loads the utility's network model, a full year of hour-by-hour power-flow data, land parcels, and water and fiber routes, and answers the core question, where and when can this much new demand plug in, across every candidate connection point in 6 minutes. The utility's normal study process quotes 14 months for the same answer. The answer: the system can serve the full gigawatt if the campuses start in stages, Lakeline 300 MW at month 16, Eastport 450 MW at month 22, South Yard 250 MW at month 28, with two new transformers and one new connection to a 345-kilovolt transmission line. Serving all three at once would cost $214 million and push service out to 2030.
Four decision points carry the story, and a person approves each one, because each commits the utility contractually or financially. First, the portfolio offer: all three campuses, staggered, with upgrades built in phases for $162 million instead of $214 million; the term sheet is signed at month 3 under a supply contract in which Colossus agrees to grow its usage on a schedule the grid can absorb. Second, the regulatory filing: new rate terms for very large customers plus an updated demand forecast, backed by a 214-page evidence pack generated automatically from the same model that sited the campuses, 3 days of work instead of 4 months. The regulator approves at month 6, on the first submission, on condition of quarterly progress reports that GridCORTEX also generates automatically from live meter data. Third, the onsite-generation terms. Colossus will have six 35 MW gas turbines and a 300 MW battery of its own behind the fence. The plan sets the protective settings that keep those generators from pushing power into the grid unexpectedly, prices the utility's obligation to back them up when they fail into the contract, and enrolls the 300 MW customer battery as a resource the utility can call on at peak. Fourth, buying the long-lead equipment: factory slots for two large transformers and high-voltage breaker positions are locked at month 7, ahead of the 24-month industry waiting line.
Then the build: site work starts at all three campuses by month 10, with detailed engineering studies running alongside construction instead of ahead of it; the transformers ship at month 11; the new transmission connection is 40% strung by month 13. Campus 1 switches on at month 16 (Lakeline, 300 MW). Month 19 is the payoff scene: a demand peak triggers a conservation alert, and the Colossus batteries discharge 190 MW into the grid for two hours under the flexibility agreement. The peak is served, no expensive standby power plant has to start, and the data center never notices. Campus 2 follows at month 22, bringing the total to 750 MW, and Campus 3 at month 28: the full 1.0 gigawatt is on line, and the model is already built for the next giant customer who walks in.
The closing comparison: 3 of 3 campuses won, $189 million of revenue captured by month 48, and 300 MW of customer batteries serving the utility's peak, against the slow path's 1 of 3 campuses, $30 million, and no battery help at all. The shaded gap between the two growth curves on the final chart is the point: months of served demand, and the investment returns that come with it, captured or lost.
The request goes into the study queue, the waiting line of engineering studies, as a single-site project handled one step at a time. With no offer on the table by month 5, Colossus threatens to build its own power plants behind the fence. By month 9 the founder is posting about "utility bureaucracy" to 200 million followers while a neighboring cooperative courts campuses 2 and 3. The regulatory filing is rejected at month 14 because it lacks a portfolio demand plan, a growth schedule, and evidence of how the system will carry the load; refiling costs six months. The failure is structural: studies are one-off projects, and nothing ties the engineering to the economics or the filing.
At month 20, campuses 2 and 3 sign with the cooperative. Campus 1's onsite generators, never properly studied, push power into the grid unexpectedly during startup and trip the protection systems. Campus 1 finally switches on at month 34, 18 months late, with the customer's battery invisible to the utility and the cost of backing up their generators never priced in. 700 MW of demand, and the returns it would have carried, went to the neighbor. Revenue captured by month 48: $30 million.
A continuously updated model of the territory turns the gigawatt request into a question software can answer: the 6-minute siting study (checked afterward by the utility's official engineering tools) finds the staggered plan that three separate studies never could, saving $52 million by sharing and phasing the upgrades. The same model writes the regulatory evidence pack, so the filing and the engineering cannot contradict each other. And it prices the customer's own equipment both ways: the risk from their gas turbines is studied and contained, and their 300 MW battery becomes a grid asset as good as a dedicated peak power plant.
Every commitment waits for a person to approve it before anything moves. The winning numbers: 3 of 3 campuses, first switch-on at month 16, the full gigawatt at month 28, $189 million of revenue by month 48, regulatory approval on the first submission at month 6, and $9 million in rush fees avoided because the transformer factory slots were booked at month 7.
| Measure | Without GridCORTEX | With GridCORTEX | The difference |
|---|---|---|---|
| Time to a serviceable answerhow long before the utility could tell Colossus where and when it could connect | 14 months (study queue) | 6 minutes (validated after) | 99.9% faster |
| Campuses securedhow many of the three campuses chose to build in this utility's territory | 1 of 3 | 3 of 3 | +700 MW |
| Regulatory outcomewhether the regulator accepted the request for new rate terms and the updated demand forecast | Rejected month 14 · refiled | Approved month 6, first submission | 8+ months saved |
| Filing preparationthe work of assembling the evidence the regulator requires | 4 months, by hand | 3 days, auto-generated | 97% less work |
| First energizationwhen the first campus actually receives power | Month 34 | Month 16 | 18 months sooner |
| Full 1.0 GW energizedwhen all 1,000 megawatts are on line; a gigawatt (GW) is 1,000 megawatts | Never; 700 MW lost | Month 28 | the whole ask |
| Onsite CT integrationCTs are combustion turbines, the customer's own gas generators; a backfeed is those generators pushing power out into the grid unexpectedly | Backfeed trip at commissioning | Studied, protected, priced | no surprises |
| Interconnection capexthe construction cost of connecting the campuses to the grid | $214M all-at-once path | $162M phased portfolio | $52M saved |
| Revenue captured by M48electricity sales revenue from these campuses through month 48 | $30M | $189M | +$159M |
| Customer BESS as grid assetBESS is a battery energy storage system; here, the customer's own 300 MW battery made available for the utility to call on | 0 MW; invisible | 300 MW the utility can dispatch | as good as a peak plant |
| Peak events covered by BESStimes the customer's battery carried the grid through a demand peak instead of an expensive standby plant | 0 | 190 MW × 2 hours at month 19 | no standby plant started |
| Standby-service exposurethe utility's cost risk of backing up the customer's generators whenever they fail; the ESA is the electric service agreement, the supply contract | Unpriced risk | Priced into the ESA | recovered |
| Transformer expediting premiumsrush fees paid to jump the manufacturing queue for large transformers | Paid under schedule pressure | Avoided; slots locked month 7 | $9M saved |
| Rate base from this loadrate base is the investment a regulated utility is allowed to earn a return on; serving more demand grows it | 300 MW share | 1,000 MW share | +233% |
There is no direct field safety benefit from a faster study, and we will not dress one up. The honest indirect mechanism is schedule compression: when a study takes months, the construction and energization window gets squeezed to hold a date the customer already announced, and squeezed windows are where overtime, short notice switching, and rushed field work come from. Returning the answer earlier gives the schedule its slack back.
Counted in units you already track:
Senior planning engineer hours come back to the study queue, and the engineer's day shifts from case setup to judgment on results.
The numbers we need from you to run that formula:
| Cost driver | How it is calculated, from a rate you supply |
|---|---|
| Planning engineering labor | engineer hours avoided x your loaded planning engineer rate |
| Outside study support | consultant study hours avoided x your contracted consultant rate |
| Load you win on speed | the proposed load in megawatts x your own expected annual revenue per megawatt of served large load x the probability you assign to winning a deal you would otherwise lose on time to answer, a probability you set, not us |
| Right sized upgrades | your own estimated cost of the upgrade set a coarse screen would have assigned, minus the upgrade set the detailed study supports, on the cases where the two differ |
| Queue carrying cost | study weeks removed x your own cost per week of carrying an open request, including customer facing staff time and account management |
You pay for the modeling service and its compute, for the work of getting your planning network model and case library into it and keeping them current, and for planning engineer time to validate the accelerated results against studies you have already run by hand. That validation set is not optional and it is where your engineers will spend real hours in the first quarter.
Payback is normally carried by engineering labor and outside study fees, which you can audit from timesheets and invoices. The deal you win on speed is the larger number, and it is the one your CFO will discount hardest, so present it as upside rather than as the base case.
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 study acceleration service for transmission planning and strategy: it models feasibility for data center and large load requests and returns a screening study, with viable configurations and timelines, in weeks instead of months. The demo above uses synthetic data; everything below describes what the real deployment needs from your organization.
| Your system | Typical products | How we connect |
|---|---|---|
| Planning and study tools | PSS/E, PowerWorld, TARA | scheduled file export (CSV or CIM XML) |
| Geographic Information System (GIS) | Esri ArcGIS Utility Network, GE Smallworld | scheduled file export (CSV or CIM XML) |
| Energy Management System (EMS) / transmission SCADA | GE e-terra, AspenTech OSI monarch | scheduled file export (CSV or CIM XML) |
| Market and grid operator interfaces | PJM, MISO, ERCOT interconnection queues | read-only API |
Runs in your own cloud account on GPU instances or an on-premises NVIDIA server. Connections are read-only through your existing data zone; network model data stays under need-to-know access as critical infrastructure information, and your planning engineers sign every study before it reaches a customer.
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 transmission planning study lead in the GridCORTEX console:
Release files the draft study into your planning study workspace in draft status. A planning engineer reviews, stamps, and releases it to the customer under your normal study process; GridCORTEX sends nothing outside.
Event-driven: declare a new request in the console and enter the MW, the proposed site, and the target in-service date; if the request arrived through your queue system, GridCORTEX pre-fills the entry for confirmation.
Studies run on your latest planning cases and an as-operated EMS snapshot; each study card shows the model vintage it was built from.
Studies live in the GridCORTEX console linked to planning tools; a finished draft or an aging request emails the study lead. 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 fair question from any utility with a planning department: "We have PSS®E/CYME, an interconnection queue process, and planners who do this, what's new here?" Here's the honest answer.
When someone asks "what did it actually calculate?", this is the list. The siting answer, the phasing plan, the regulatory pack, and the DER terms in this demo are each the output of analyses across every category below. In the simulation these factors drive the storyline; in a pilot they are computed from your GIS, network model, EMS/ADMS, asset registry, AMI, tariff book, and regulatory calendar.
Presenter's one-liner: "It analyzed the wires, the transformers, the substations, and every hour of the year of power flow across the territory, then picked the sites that serve this size of load with the least new infrastructure and the most benefit to the utility, priced the customer's own generation as both a risk and an asset, and generated the regulatory evidence to prove all of it. That's what you just watched."