The dirty secret under every ADMS, DERMS, and OMS: the model lies. A quarter of customers mapped to the wrong transformer, one in five on the wrong phase, connectivity that hasn't matched the field since the 90s conversion, and every advanced application you buy inherits all of it. The old fix is a four-year, $30M field-walk that gets scoped, quoted, and cancelled. The new fix: the grid already knows its own truth. Meters that sag together are connected together. Outages reveal who's really behind which device. Meter events identify phase. Watch the truth engine turn that evidence into 54,000 corrections with confidence scores, and a per-feeder Data Quality Index that finally tells you which apps can be trusted where.
| Without | With GridCORTEX | Δ |
|---|
| Without | With GridCORTEX | Δ |
|---|
The scenario is a 12-week data cleanup program at a fictional utility whose grid map has quietly rotted for decades. The master map lives in the geographic information system, or GIS, and every other tool trusts it. The dashboard's starting facts: a territory Data Quality Index of 68 out of 100, only 72 percent of customers mapped to the transformer that actually serves them, and line connections that have not matched reality since a 1990s records conversion. Why it matters: every advanced software product the utility buys inherits these errors and quietly misfires. The traditional fix, walking the entire territory to verify everything by hand, was quoted at 4 years and $30 million and cancelled twice. The demo's premise is that the grid has already recorded its own truth. The engine ingests 14 billion voltage readings from smart meters, 6 years of outage records, meter event logs, sensor feeds from the control system, and the full map export, and cross-examines them against each other. Early finds set the tone. On circuit FDR-108, the voltage of 61 meters sags in lockstep with transformer T-2214, yet the map assigns 19 of them to a different transformer two poles away. On circuit FDR-117, a 2024 automatic breaker operation darkened 412 meters when the map predicted 371; the 41-meter difference is itself a map error, recorded by the event.
Three decisions go to the board of data stewards, the people who guard the master map. Around week 4, the first presents the truth list: 41,300 customers mapped to the wrong transformer (28 percent), 12,800 recorded on the wrong phase, meaning the wrong one of the three wires that share each line, and 340 connection errors, such as switches that do not exist and links the map never knew about. In total, 54,100 proposed corrections, each carrying its evidence and a confidence score. Around week 6, the second decision sets the correction rules: fixes scoring 95 percent confidence or better, which is 94 percent of them, apply automatically in batches that a human steward approves; the 80-to-95-percent band gets a desk review; and only the ambiguous 6 percent requires a truck and a field visit. A field audit of the first 400 automatic fixes finds 98.2 percent were correct, and batches run at about 6,100 corrections a week. Fixing the phase records pays an immediate bonus: a planning tool finds 9 circuits where rebalancing the three wires frees real capacity (use case 5.6).
Around week 9, the third decision makes trust measurable circuit by circuit: a per-circuit Data Quality Index across connections, phase, customer-to-transformer mapping, and wire properties, with each advanced application unlocked only when the data it depends on is good enough. Outage prediction unlocks at a mapping score of 95 or better. Automatic fault rerouting, known as FLISR, unlocks at 97 on connections. Voltage optimization, known as VVO, unlocks at its own thresholds. A standing drift watch then feeds every switching order, meter swap, and outage back into the evidence, so the map stays true. The ending state: outage prediction unlocked territory-wide at 98.5 percent accuracy, automatic rerouting unlocked on 21 of 28 circuits, voltage optimization on 17, and the program finished at a Data Quality Index of 96, up from 68, in nine months for $4.2 million, with all 54,100 corrections traceable to their evidence.
The conventional path is living with the lies. The outage system predicts the wrong extent of a failure and sends crews to a transformer that was never involved: 40 minutes lost, again and again. The automatic rerouting pilot, switched on in good faith, opens the wrong switch because the map's connections were wrong, and the advanced applications get quietly demoted to advisory status while their licenses keep billing. The voltage optimizer pushes settings based on phase records that are 18 percent wrong, and field equipment gets blamed for the resulting complaints. The cleanup project is scoped again at 4 years and $30 million of field walks, loses to storm hardening in the budget, and is cancelled for the third time in a decade. The core failure: the only accepted source of truth is a field walk nobody will ever fund, so zero corrections get applied and the quality score drifts down from 68 forever.
The capability is a truth engine (use case 5.2, fed by use case 17.1) working from three kinds of evidence: meters whose voltage sags together must be connected together; every historical outage is a natural experiment revealing who is really behind which device; and meter events reveal which of the three wires each customer is on. The engine weighs the evidence, writes each correction with its proof attached, and scores its own confidence. The 4-year field walk becomes a 9-month evidence program, with truck visits for only the ambiguous 6 percent. Humans stay in charge by design: map stewards approve every batch, and ongoing field audits keep the engine's confidence honest. The per-circuit quality gates then make "is the data good enough?" a number, so applications unlock only where trust is earned. The winning numbers: 54,100 corrections, mapping accuracy from 72 to 98.5 percent, quality index from 68 to 96, and $4.2 million instead of $30 million.
| Measure | Without GridCORTEX | With GridCORTEX | Delta |
|---|---|---|---|
| Data Quality Indexa 0-to-100 health score for the accuracy of the utility's master grid map | 68: and drifting | 68 → 96 | +28 points |
| Corrections appliedindividual map errors found and fixed during the program | 0 | 54,100 | each fix carries its proof |
| Customer-transformer accuracythe share of customers mapped to the transformer that actually serves them | 72% | 98.5% | 26.5 points better |
| Phase accuracythe share of customers recorded on the correct one of the three wires that share each line | 82% | 99% | 12,800 corrected |
| Connectivity errorsplaces where the map's picture of how lines connect did not match the real grid | 340 unknown | 340 found & fixed | the links the map never knew |
| Correction basiswhat each fix rests on | opinion + site visits | evidence chains + confidence | auditable |
| Field verification burdenhow much of the territory still needs an in-person truck visit to verify | 100% of territory | 6%: the ambiguous tail | the walk you can afford |
| Model after the projectwhether the map stays accurate once the cleanup ends | rots immediately | drift watch: stays true | a diet, not a cleanse |
| Cleanup cost & timethe price and duration of getting to a trustworthy map | $30M · 4 yrs · cancelled | $4.2M · 9 months · done | 86 percent cheaper |
| OMS outage predictionthe outage management system's ability to predict which equipment failed and who lost power | wrong extents, wrong crews | unlocked territory-wide | trucks to the right pole |
| FLISRautomatic switching that isolates a fault and reroutes power around it; a feeder is a main circuit serving a neighborhood | misoperated, demoted | unlocked on 21/28 feeders | see Commissioning Day |
| VVOvoltage optimization software that fine-tunes delivery voltage to cut energy waste | pushed on wrong phases | unlocked on 17/28 | phase truth first |
| Every study & twinevery engineering study and computer model built on top of the map | inherits the lies | stands on the floor | every other tool benefits |
| Prudence & audit posturehow the utility defends its spending to regulators and auditors; Relay is the system's decision log | "trust us" | every fix evidence-linked | provable, not asserted |
Two real exposures move. Troubleshooters stop being sent to hang and retrieve recorders on hunches, and transformers that were going to fail in service get changed out on a planned outage instead of failing hot, which is when oil, fire, and an unplanned energized job all arrive together.
Counted in units you already track:
Truck roll hours come back to troubleshooters and meter techs, and analysis hours come back to the distribution engineer who currently writes queries by hand.
The numbers we need from you to run that formula:
| Cost driver | How it is calculated, from a rate you supply |
|---|---|
| Field labor and vehicle | truck rolls avoided x crew hours per roll x your loaded crew rate, plus miles avoided x your fleet cost per mile |
| Emergent to planned conversion | your own cost premium of an emergent changeout over a planned one x the changeouts you convert |
| Overtime and callout | after hours callouts avoided x your callout minimum and overtime multiplier |
| Customer equipment claims | your average paid claim for equipment damage from sustained voltage problems x the claims avoided by finding the condition first |
| Deferred replacement | your unit cost per distribution transformer x the units you defer once measured loading shows they have headroom |
You pay for the scoped engagement that builds and runs this and the compute to process the full meter population rather than a sample, for the integration into your MDM, your GIS, and your EAM, the enterprise asset management system, and for your engineers to validate the anomaly queue against a season of known outcomes. The honest prerequisite is meter to transformer connectivity data. If your GIS does not know which meters sit under which transformer, fixing that is the first project.
Payback is dominated by avoided truck rolls and by the emergent to planned conversion on transformer changeouts, because both are in records you already keep. Treat customer claims as upside.
This one has a direct mechanism and it is worth stating in plain terms. A switch mapped in the wrong state or a lateral mapped on the wrong phase is a crew hazard, because isolation decisions and clearance boundaries are built on the model. Correcting the errors that matter most, and correcting them from meter and SCADA evidence rather than from a field visit, removes both the bad data and most of the trips taken to confirm it.
Counted in units you already track:
Investigation hours come back to the ADMS support engineers and the mapping team, and editors spend their day making corrections instead of hunting for what to correct.
The numbers we need from you to run that formula:
| Cost driver | How it is calculated, from a rate you supply |
|---|---|
| Investigation labor | investigation hours avoided x your loaded ADMS engineer and GIS editor rates |
| Field verification | verification trips avoided x crew hours per trip x your loaded crew rate, plus miles avoided x your fleet cost per mile |
| Restoration performance | customer minutes attributable to incorrect fault location or isolation x your own cost per customer minute of interruption, using the share your outage records support and not a share we assert |
| Model audit projects | the cost of your periodic audit project x the share the continuous audit displaces, which your mapping supervisor sets |
| Unrealized ADMS value | the annual benefit your ADMS business case assigned to automated restoration x the share of feeders currently running with it disabled or overridden for model reasons |
You pay for the GridCORTEX audit service, for read access and integration into GIS, AMI, SCADA, and your outage records, for the GIS editing queue integration, and for your editors' time to actually work the corrections. The finding is the cheap part. The correcting is real mapping labor and the queue will be long at first, so budget the editor hours before you buy the audit.
Payback is dominated by field verification trips avoided and by whatever restoration performance your own outage cause coding will support. If your dispatchers have automation disabled on a set of feeders today, the unrealized ADMS value line usually turns out to be the largest number on the page and the easiest one to defend internally.
Two mechanisms, both real. Field surveys with a clamp on ammeter are energized area entries made purely to gather data, and detecting imbalance from advanced metering infrastructure (AMI) removes most of them. Separately, sustained imbalance drives neutral current and thermal loading on the heavy phase, and the failures that follow are the ones that put crews on emergency work at night.
Counted in units you already track:
Survey crew hours and engineering design hours come back, and the engineer reviews a ranked transfer plan instead of building options from a map.
The numbers we need from you to run that formula:
| Cost driver | How it is calculated, from a rate you supply |
|---|---|
| Field survey labor and travel | survey trips avoided x crew hours per trip x your loaded crew rate, plus miles avoided x your fleet cost per mile |
| Engineering design labor | design hours avoided per transfer x transfers per year x your loaded distribution engineer rate |
| Losses | megawatt hours of loss reduction from restored balance, computed on your own circuits by the analysis, x your own avoided energy cost per MWh |
| Avoided equipment loss | distribution transformers and conductor sections lost to imbalance driven thermal duty per year x your installed replacement cost, x the share you believe earlier detection would have prevented |
| Outage cost | customer minutes from phase overload fuse operations x your own cost per customer minute of interruption |
You pay for the GridCORTEX analytics service, for AMI and SCADA integration good enough to trust at the meter level, for the connection into your work management system, and for the design and crew time to execute the transfers it recommends. The transfers themselves are the largest cost and they compete with everything else in the switching schedule, so sequence them rather than releasing the whole list at once.
Payback is dominated by avoided survey trips and by avoided equipment loss on the circuits where imbalance is worst. Losses are real but small per feeder, and outage cost depends entirely on how many of your fuse operations your own records attribute to overload rather than to trees or animals.
The mechanism is indirect but specific, and it is the reason gate discipline exists. Automated restoration acts on the model. Letting FLISR go live on a feeder whose connectivity or switch status is wrong means an automated switching scheme operating on a picture that does not match the field, which is a crew and public exposure. Holding those feeders is the safety benefit, and it costs schedule, which is exactly the trade the gate report is meant to make visible.
Counted in units you already track:
Program office and engineering hours come back to the ADMS program, and the director walks into the steering committee with a current number instead of a two week reconstruction.
The numbers we need from you to run that formula:
| Cost driver | How it is calculated, from a rate you supply |
|---|---|
| Program office and engineering labor | readiness check and reporting hours avoided x your loaded program manager and ADMS engineer rates |
| Integrator rework | rework hours attributable to data problems found after an application went live x your systems integrator's contracted hourly rate, from your own change order history |
| Schedule slip | months of program delay avoided x your own carrying cost per month for the ADMS program, which your finance group already computed for the business case |
| Unrealized application value | the annual benefit your ADMS business case assigned to each advanced application x the share of feeders where that application is not live or not trusted because of data |
| Field verification | verification trips avoided across the rollout x crew hours per trip x your loaded crew rate, plus miles x your fleet cost per mile |
You pay for the GridCORTEX readiness service, for integration to GIS, AMI, SCADA, your outage records, and your project schedule, and for the work of encoding your ADMS vendor's data model requirements per application, which is a joint exercise with your vendor and your integrator. Then you pay in mapping editor hours to close the gaps the scoring finds, and there will be more of them than the program plan assumed.
Payback is dominated by schedule slip avoided and by integrator rework, both of which your program already tracks in change orders. Program office labor is real but small. The unrealized application value line is usually the largest, and it is the one your steering committee will find hardest to argue with.
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.
An analytics and planning service for distribution engineering: it detects phase imbalance drift from advanced metering infrastructure (AMI) data and returns a ranked list of load transfer recommendations, each with the customers to move and the expected balance improvement. The demo above uses synthetic data; everything below describes what the real deployment needs from your organization.
| Your system | Typical products | How we connect |
|---|---|---|
| Metering (AMI head-end and meter data management) | Itron, Landis+Gyr, Aclara; Oracle or Itron meter data systems | database replica refreshed nightly |
| Geographic Information System (GIS) | Esri ArcGIS Utility Network, GE Smallworld | read-only API |
| SCADA historian | AVEVA PI System, AspenTech eDNA, GE Proficy | historian mirror (one-way feed) |
| DER management (DERMS) and DER program platforms | Schneider, GE Vernova, or your interconnection registry | scheduled file export (CSV or CIM XML) |
| Customer Information System (CIS) / billing | Oracle CC&B, SAP IS-U | database replica refreshed nightly |
| ADMS vendor requirements | the vendor's data model and application prerequisite documents | document upload |
| Project scheduling | Oracle Primavera P6, Microsoft Project | scheduled file export (CSV or CIM XML) |
Runs in your cloud account on GPU instances or an on-premises NVIDIA server, read-only through your existing data zone, with meter data handled under your customer data rules. Transfers are planned and executed by your engineers and crews through the normal work process; the tool only recommends.
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 distribution engineer watching feeder loading in the GridCORTEX console:
Approve creates the transfer as a draft design job in your work management system through its API; design reviews it, and crews switch under ADMS control per your procedures. GridCORTEX never initiates switching or edits the model.
Data triggered from AMI population analysis; nothing to enter. An engineer can defer a recommendation or flag a customer constraint in one click.
Rescans AMI data nightly and rereads SCADA continuously; each recommendation shows the analysis window and its as-of timestamp, including on pilot replicas.
GridCORTEX imbalance rankings with a monthly email; a phase projected to overload within 30 days pushes a Teams alert. 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 grid-mod or GIS leader: "We have a GIS team, data stewards, and a cleanup project in the roadmap, what's new here?" Here's the honest answer, and it's why the cleanup project keeps getting cancelled.
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 AMI, OMS, SCADA, and GIS.
Presenter's one-liner: "A quarter of the customers were mapped to the wrong transformer, and the fix was a thirty-million-dollar field walk nobody would ever fund. But the grid had been recording its own truth all along, every sag, every outage, every meter event. The engine cross-examined the evidence, wrote fifty-four thousand corrections with confidence scores, and gave every feeder a data quality number that tells you which apps to trust where. Nine months, four million dollars, and everything you've bought finally works."