GridCORTEX Live · ADMS Advanced Applications  ·  ← 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

Commissioning Day Synthetic Data · Simulation

The industry's open secret: utilities buy ADMS advanced applications, FLISR, Volt/VAR optimization, load flow, and then never turn them on, because the apps are only as good as the network model and the field devices they command. This is the demo of getting them on and keeping them on: a full field-device audit (65 of 198 devices fail verification), a model remediation list ranked by what FLISR actually needs, 340 staged faults replayed in the digital twin before any live enablement, and then the question that decides success: where do the FLISR and VVO pilots go first? Not the feeder the vendor picked. The one the math picks.

WEEK 1
DEVICE AUDIT
⏳ DECISION POINT: TIME SLOWED
SYNTHETIC DATA
Device verified Model mismatch Comms dead / wrong CT Twin replay in progress FLISR pilot loop VVO pilot feeder

The apps you already bought, finally on.

What ADMS advanced applications deliver when the model is clean, the twin has rehearsed, and the pilots go where the math says
,
Model accuracy
,
FLISR pilot CAIDI
,
Misoperations
Commissioning
WithoutWith GridCORTEXΔ
The Investment
WithoutWith GridCORTEXΔ
Illustrative simulation on synthetic data; device counts, scores, and results are placeholders. In a GridCORTEX pilot, the audit runs on YOUR GIS, YOUR ADMS model export, and YOUR device fleet, and the pilot placement is computed for your territory. See UC 5.2 "Demo and Proof Plan."
0 / 198
Field devices audited
71%
Network model accuracy
0 / 340
Twin fault replays passed
OFF
FLISR · VVO status
Commissioning Feed: GIS · ADMS · field test crews · human-in-the-loop
Audit
Fix model
Twin replay
Place pilots
Live
The Validated Use Cases Behind This Scenario
UC 5.2
Network Model Health
The foundation: GIS-vs-field-vs-ADMS reconciliation, continuously, because FLISR can't run on a model that lies.
UC 5.5
Volt/VAR Performance Auditor
Cap bank health, regulator behavior, end-of-line voltage from AMI, VVO configured against reality, then audited forever.
UC 1.1
ADMS Control Room Co-Pilot
The operators' companion once the apps are live, and the reason they stay trusted instead of switched off.
UC 5.7
ADMS Data Readiness & Migration Gatekeeper
The program behind this day: feeders scored against the vendor's model requirements, waves sequenced, gates enforced, so commissioning day arrives ready instead of surprised.
187 UCs
One Framework
Commissioning Day 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

A fictional utility spent $14 million on two advanced software applications for its control room, and 14 months later neither has ever been switched on. The first, called FLISR (fault location, isolation, and service restoration), automatically finds a broken section of power line, seals it off, and reroutes power around it in seconds. The second, Volt/VAR optimization (VVO for short), fine-tunes voltage along each line so customers get stable power while the utility wastes less energy. Both are common purchases across the industry, and both commonly sit unused, because they are only as good as the utility's computer model of its own grid. Here that model is only 71% accurate. The simulation covers a 10-week program to turn the apps on safely. Weeks 1 and 2 are a full census: all 198 field devices (automatic pole-top circuit breakers, voltage-support equipment, and motorized switches) are checked against the utility's map database, its control-room model, its live sensor feeds, and its smart-meter data. The census exposes the industry's open secret in one line: 65 of the 198 devices, one in three, fail verification. There are 47 map and model errors, 11 devices that no longer communicate, and 7 sensors with the wrong calibration settings, so their readings are wrong. Automatic switching software running on this model would be guessing.

Three recommendations go to the engineering team for approval. Week 2: a repair list ranked by consequence, not by discovery order. The 23 errors that would make the automatic-switching app misfire come first, the 18 that would mislead the voltage app come second, and cosmetic errors come last. That ordering turns a 6-month clean-everything project, the kind that never finishes, into an estimated 6 weeks of mapping-technician work. Model accuracy climbs from 71% to 84% by week 3, 92% by week 4, and eventually 98%. Week 5: rehearsal in a digital twin, a working computer model of the grid where decisions can be tested safely. Before anything goes live, 340 simulated line failures are replayed against the corrected model, and the voltage settings are tested against 12 months of real smart-meter voltage readings. Replay 88 shows the payoff: the switching app isolates a simulated failure in 3 moves and restores 2,240 of 2,610 affected customers in 41 seconds. Replay 204 catches a plan that would have opened the wrong switch. In all, 12 replays fail, and 2 of those failures would have been live wrong-switch events that dropped real customers. All 12 are fixed in simulation, and the rehearsal finishes 340 of 340 passing.

Week 7 is the decision that makes or breaks the program's reputation: where to run the first pilot. Not the line the vendor suggests. The software scores every candidate and picks the RIVERSIDE loop (lines F-101 and F-117) for the switching pilot: it has the right layout, healthy equipment, and the territory's worst outage-frequency record, so the improvement will be visible to everyone. The voltage pilot goes to line F-104, which has the widest voltage swings and full smart-meter coverage to prove the savings. Go-live is staged so trust can build: watch-only mode in week 8, then advisory mode, then full automatic control in week 9. That week the first real failure hits the pilot loop, and 2,180 customers get their power back in 47 seconds, before anyone's phone rings. The voltage app follows and delivers a measured 2.3% cut in energy use on its line. The closing scorecard compares this against the industry's usual outcome. Its headline tiles: model accuracy 98% versus 71%, outage duration on the pilot loop down 38% versus no change with the apps off, and wrong-switch events 0 versus 1 followed by the software being shelved.

Without GridCORTEX

The failure is sampling under schedule pressure. The installation contractor spot-checks 40 of the 198 devices, they pass, and the automatic-switching app goes live everywhere on the 71% model. In its first storm it opens the wrong switch and cuts power to 1,800 customers whose lines were healthy. The mistake makes the morning meeting, then the newspaper. The voltage app, misled by two failing pieces of voltage equipment the model did not know about, switches them on and off 30,000 times in six weeks, wearing them out, so operations quietly turns it off. By month 8 both applications are disabled "pending model cleanup," and the $14 million in licenses joins the shelf of software the utility owns but does not use.

With GridCORTEX

The software does three things, each approved by people. It runs a census instead of a sample: every device checked across four data systems at once, with fixes ranked by which application they would break. It runs a rehearsal instead of a gamble: 340 simulated failures replayed in the digital twin, where a wrong switch costs nothing and harms no one. And it places the pilots by scoring, so the first win happens where equipment is healthy and customers will notice. Protection engineers sign off on every switching plan, and operations controls every step up from watch-only to automatic. The result: 98% model accuracy at go-live, zero live wrong-switch events, outages on the pilot loop 38% shorter, 2.3% energy savings on the voltage pilot, and applications that stay on because the checking never stops.

The KPIs, side by side
KPIWithout GridCORTEXWith GridCORTEXDelta
Device censushow many of the 198 field devices were actually checked before go-live40-device sampleall 198 verifieda census, not a sample
Model accuracy at go-livehow well the utility's computer model matches the real grid when automation switches on71%98%the whole game
Fix effortthe work needed to correct the model errors, and the order the fixes happen in6-month cleanup (never done)6 weeks, app-impact orderranked by consequence
Misoperationstimes the automation opened a wrong switch and cut power to healthy customers1 live, 1,800 customers2 caught in twin, 0 livemistakes made where they are free
VVO cap-bank huntingthe voltage app rapidly switching worn voltage equipment on and off, wearing it out30,000 cycles, then OFFsick banks fixed pre-launchaudit first, automate second
Pilot placementwhich power line hosts the first live trial, and who chose itvendor reference feederscored: layout + health + outage historythe win is visible
The $14M licenseswhat actually happens to the software the utility paid forshelfware by month 8closed-loop and trustedthe investment performs
FLISR pilot CAIDICAIDI is average outage length per affected customer; here, on the pilot loopn/a (apps off)−38% on the pilot loopan improvement regulators can see
VVO energy (CVR)CVR means conservation voltage reduction: energy saved by holding voltage slightly lower and steadier on the pilot linen/a (apps off)2.3% on the pilot feedermeasured from real meter data
Operator trustwhether the control-room staff believe the automation enough to leave it onburned by the misopbuilt in monitor modethe apps stay ON
Post-go-live driftthe model slowly going stale again after commissioning daynext year's problemaudit runs continuouslystays commissioned
Commissioning evidencethe proof that the system was tested before it was trustedbinderstraceable sign-offs & replay logsbuilt for auditors
Live KPIs on the dashboard
Field devices auditedThe census counter, 0 to 198. Stopping at a 40-device sample is how the industry usually fails; reaching 198 means every device the automation will command has been verified.
Network model accuracyHow well the computer model matches the real grid. 71% means automation would be guessing; 98% means it can be trusted. Watch it climb as the ranked fixes land.
Twin fault replays passedProgress through the 340 simulated failures. It lags behind the replay count while the 12 caught failures are being fixed, then completes at 340 of 340: a clean rehearsal before opening night.
FLISR · VVO statusThe state of the two applications. OFF, MONITOR, ADVISORY, then CLOSED-LOOP is the healthy path, with trust building at each step. ON at 71% accuracy followed by DISABLED is the industry's usual story.

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 Commissioning Day, 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 5.2 Network Model Health and Maintenance Assistant

What happens today, without this

Nobody audits the distribution network model on purpose. Errors surface when the advanced distribution management system (ADMS) behaves oddly: a dispatcher says fault location pointed at the wrong lateral, an engineer opens a ticket, a geographic information system (GIS) editor traces the circuit by hand to find what is wrong, and sometimes a crew is sent out to put a meter on a transformer and confirm which phase it is actually on. Phase labels and connectivity come from decades of as built drawings and field sketches of varying quality. A full model audit happens as a funded project every few years, then drifts again. In the meantime dispatchers keep an informal list of feeders where they do not trust the automation.

What it replaces or shrinks

  • Reactive, one ticket at a time chasing of model errors after an application misbehaves
  • Manual GIS traces to work out why a connectivity or phase result looks wrong
  • Truck rolls purely to verify the phase of a transformer, a meter, or a lateral
  • The periodic funded model audit project, replaced by a continuous audit against meter and outage evidence
  • Shrinks the dispatcher's informal list of feeders where automated restoration is turned off or overridden
  • Shrinks the guesswork in deciding which model corrections to fund first

Why it is safer

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:

  • Road miles driven on phase and connectivity verification trips, which the evidence based audit replaces for most cases
  • Switching operations executed against model data later found to be wrong, which is the exposure the correction queue is designed to shrink
  • Energized area entries made solely to verify a model attribute in the field rather than to perform work
  • Night driving hours during restorations extended by an incorrect fault location or an incorrect isolation point

Man-hours it gives back

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.

HOURS AVOIDED PER YEAR = model error tickets per year x hours per investigation across the ADMS engineer and the GIS editor, plus field verification trips per year x crew hours per trip including drive time, plus the hours in your periodic model audit project divided across the years between audits, minus the editor time still spent making, checking, and promoting each corrective edit, which does not go away.

The numbers we need from you to run that formula:

  • Model error tickets raised per year and the average investigation hours each consumes
  • Field verification trips per year taken to confirm phase or connectivity, and crew hours per trip
  • Hours and elapsed months in your last full model audit, and how often you repeat it
  • Number of feeders running automated restoration, and how many are currently disabled or overridden for model reasons
  • Loaded hourly rate for a GIS editor, an ADMS support engineer, and a line crew including vehicle

Where the dollars come from

Cost driverHow it is calculated, from a rate you supply
Investigation laborinvestigation hours avoided x your loaded ADMS engineer and GIS editor rates
Field verificationverification trips avoided x crew hours per trip x your loaded crew rate, plus miles avoided x your fleet cost per mile
Restoration performancecustomer 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 projectsthe cost of your periodic audit project x the share the continuous audit displaces, which your mapping supervisor sets
Unrealized ADMS valuethe 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

Reliability and maintenance

Reliability
Fault location, isolation, and service restoration accuracy is a model problem before it is a software problem, so this touches CAIDI and therefore SAIDI directly: a correct model puts the crew at the right lateral the first time and lets automation reclose the right sections. Be honest about the size, though. The benefit is bounded by how many of your long restorations your own outage records trace to model error, and that is a number you already have.
Maintenance
Model corrections stop being emergent work triggered by a misbehaving application and become a ranked, evidence backed queue the mapping team works through in priority order. Because the audit runs continuously, drift introduced by a batch of GIS edits is caught in the next cycle instead of at the next outage.

What else it moves

ComplianceEvery proposed correction carries the meter or SCADA evidence that justified it, which gives you a defensible record of model quality when a reliability report or a restoration event is questioned.
WorkforceEditors and ADMS engineers stop firefighting and start working a prioritized queue, which is the difference between a job people leave and a job people stay in.
CustomerFewer customers left out because the automation isolated the wrong section, and fewer wrong estimated restoration times built on wrong connectivity.

What it costs you, stated honestly

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.

How to build the payback case

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.

This is a planning model built from your ticket counts, your outage records, and your own labor rates, not a vendor claim. Re-run it after the first correction wave, using the errors you actually found and the edit hours they actually took.
UC 5.5 Volt/VAR Optimization Performance Auditor

What happens today, without this

Most utilities find out how their volt/VAR optimization (VVO) is performing from the vendor's own dashboard and from a benefit study done once, at commissioning, by an outside consultant. After that the number is repeated in filings for years. A drifted regulator setpoint or a capacitor bank that has stopped switching is usually found the slow way: a customer voltage complaint, or a settings review that comes around on a calendar. When an engineer wants to check, they export advanced metering infrastructure (AMI) voltage data for a feeder into a spreadsheet and eyeball it. Meanwhile field crews visit regulators and capacitor banks on a calendar cycle whether or not anything is wrong with them.

What it replaces or shrinks

  • The periodic outside consultant study of conservation voltage reduction (CVR) benefit
  • Manual AMI voltage exports and spreadsheet spot checks of a feeder's voltage profile
  • Complaint driven discovery that a regulator setpoint or a capacitor bank control has drifted
  • Hand assembly of the performance evidence for energy efficiency program reporting
  • Shrinks calendar based visits to every regulator and capacitor bank, replaced by visits to the devices the data names
  • Shrinks the argument about whether the VVO is working, which is currently settled by opinion

Why it is safer

There is a direct mechanism here, through targeting. A regulator or a pole mounted capacitor bank visit is an energized area entry, often elevated work, and usually a switching operation to isolate the bank. Sending crews to the devices the performance data actually names, rather than to every device on a calendar, removes the entries that were never going to find anything.

Counted in units you already track:

  • Energized area entries at regulator and capacitor bank locations for routine calendar based settings checks
  • Elevated work hours on pole mounted capacitor banks and regulators visited without cause
  • Switching operations required to isolate a capacitor bank before work, one per avoided visit
  • Road miles driven on calendar based settings rounds across a distribution territory

Man-hours it gives back

Crew hours come back to distribution operations and engineer hours come back to the voltage owner, who reviews a ranked gap report instead of building one.

HOURS AVOIDED PER YEAR = regulators and capacitor banks in the program x calendar visits per device per year x crew hours per visit including drive time, minus the visits still made to the devices the audit flags, plus consultant study hours purchased per year, plus engineer spreadsheet analysis hours per audit cycle x cycles per year, minus the engineer review time still spent approving each recalibration work order.

The numbers we need from you to run that formula:

  • Count of regulators and capacitor banks in the VVO program and the current visit cycle
  • Crew hours and miles per device visit, including drive time
  • Consultant hours and fee for your current CVR benefit study, and how often you buy one
  • Engineer hours per manual VVO spot check and how many you do per year
  • Loaded hourly rate for a distribution engineer and for a line crew including vehicle, plus your fleet cost per mile

Where the dollars come from

Cost driverHow it is calculated, from a rate you supply
Recovered energymegawatt hours recovered by closing the identified gap x your own avoided energy cost per MWh, where the megawatt hour figure comes from the audit against your own feeder measurements and not from any external benchmark
Peak demandkilowatts of peak reduction restored x your own capacity cost per kilowatt year
Field labor and traveldevice visits avoided x crew hours per visit x your loaded crew rate, plus miles avoided x your fleet cost per mile
Consultant studiesCVR study fees displaced per cycle x cycles per year
Energy efficiency program creditrecovered megawatt hours x the incentive or lost revenue mechanism rate in your own jurisdiction, where one exists, and zero where it does not

Reliability and maintenance

Reliability
This is mostly a voltage quality and efficiency story rather than a SAIDI or SAIFI story, and it is more honest to say that. What it does touch is the failure mode where a capacitor bank has stopped switching or a regulator is hunting, both of which show up in the performance data long before they show up as a customer complaint or a failed device.
Maintenance
Performance data names the device, so a stuck regulator or a dead capacitor control is found from the office and scheduled, rather than found on a calendar visit that may be months away. This is the case where a preventive maintenance interval can honestly move from calendar to condition, because the condition is directly observable in the voltage and switching record.

What else it moves

ComplianceVoltage delivered at the customer meter is measured against your service voltage limits continuously, and the CVR benefit you report in energy efficiency filings has measured evidence behind it rather than a commissioning study from years ago.
CustomerVoltage complaints get answered with the feeder's own measured profile, and the energy the VVO was supposed to save actually shows up on bills.
EnvironmentRecovered conservation voltage reduction is energy not generated, and it is one of the few efficiency measures that requires no customer action at all.

What it costs you, stated honestly

You pay for the GridCORTEX audit service, for AMI, SCADA, and VVO telemetry integration, for the connection into your enterprise asset management or work order system and your ADMS change queue, and for the field crew time to actually perform the recalibrations the audit recommends. The recalibration work is a real cost and it lands on a crew that is already busy, so schedule it before you commit to the energy number.

How to build the payback case

Payback is usually dominated by recovered energy and peak, but only if your VVO is genuinely underperforming. If the audit finds a small gap, the case rests on avoided calendar visits and displaced consultant studies, which are smaller but far easier to defend. Run the audit on a few feeders first and let the answer decide the program size.

This is a planning model built from your own feeder measurements, energy costs, and labor rates, not a vendor claim. Re-run it after the first recalibration cycle using the measured voltage reduction you actually achieved, because that is the only number a regulator will accept.
UC 5.7 ADMS Data Readiness and Migration Gatekeeper

What happens today, without this

Advanced distribution management system (ADMS) data readiness is tracked in a spreadsheet that a program manager updates from emails and status calls. Readiness against the vendor's data model requirements is checked by sampling: an engineer opens a handful of feeders, walks the requirements list for one application, and the result is generalized to the region. When the steering committee asks which feeders are ready for fault location, isolation, and service restoration (FLISR) or for volt/VAR optimization, assembling a defensible answer takes a week or two, and the answer is stale by the next model sync. Drift is the part nobody watches: feeders certified in an early wave get edited in the geographic information system (GIS) months later, and nobody rechecks them until an application starts giving bad answers after go live.

What it replaces or shrinks

  • The manually maintained readiness spreadsheet and the status call chasing behind it
  • Sampling based manual feeder checks against the vendor's data model requirements
  • The multi week scramble to assemble a defensible readiness answer before each steering committee
  • Manual re-verification of previously certified feeders after each GIS or model sync
  • Shrinks the post go live root cause investigation when an advanced application underperforms
  • Shrinks the negotiation with the vendor and the integrator about whose problem the data is

Why it is safer

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:

  • Switching operations executed by automation on model data that had not been verified for that application
  • Energized area entries during field verification sampling, reduced because readiness is scored from evidence in your systems rather than from field visits
  • Road miles driven on verification trips across a multi year, multi region rollout

Man-hours it gives back

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.

HOURS AVOIDED PER YEAR = feeders in the program x manual readiness check hours per feeder per wave x waves, plus steering committee reporting cycles per year x hours per report package across the program office and engineering, plus model syncs per year x re-verification hours per sync, plus post go live investigation hours per underperforming application, minus the program office time still spent reviewing and signing each generated gate report.

The numbers we need from you to run that formula:

  • Feeders in the program, number of migration waves, and advanced applications in scope
  • Engineer hours to check one feeder against one application's data requirements today
  • Steering committee reporting cycles per year and hours to assemble each package
  • Model syncs or GIS promotion cycles per year and hours spent re-verifying after each
  • Loaded hourly rate for a program manager, an ADMS engineer, and your systems integrator's contracted rate

Where the dollars come from

Cost driverHow it is calculated, from a rate you supply
Program office and engineering laborreadiness check and reporting hours avoided x your loaded program manager and ADMS engineer rates
Integrator reworkrework 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 slipmonths 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 valuethe 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 verificationverification trips avoided across the rollout x crew hours per trip x your loaded crew rate, plus miles x your fleet cost per mile

Reliability and maintenance

Reliability
The ADMS is your reliability instrument, so this is a reliability use case at one remove. The SAIDI and SAIFI improvement in your ADMS business case only lands on feeders where the advanced applications are actually enabled and actually trusted by dispatchers. Every feeder held back, disabled, or overridden for data reasons is business case benefit you are paying for and not receiving, and this makes that gap countable.
Maintenance
Data corrections stop being an undifferentiated cleanup backlog and become a sequenced queue tied to a wave and a gate date, so the mapping team knows what has to be right by when. Continuous scoring means drift after a GIS edit batch is caught in the next cycle rather than after go live.

What else it moves

ComplianceEach gate decision carries the evidence behind it, which is the record you need when a regulator asks why a modernization program is behind schedule or over budget.
WorkforceThe program office stops spending its week on spreadsheet reconciliation and status chasing, which is the work that burns out program managers on multi year rollouts.
CustomerThe restoration performance the program promised customers arrives on the schedule it was promised, on feeders where it will actually work.

What it costs you, stated honestly

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.

How to build the payback case

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.

This is a planning model built from your program plan, your feeder counts, and your own labor and integrator rates, not a vendor claim. Re-run it after the first gated wave, using the correction hours the wave actually consumed rather than the ones the plan assumed.
UC 1.1 ADMS Assistant (Control Room Co-Pilot)

What happens today, without this

When a device operates and nobody knows why, the operator on shift goes hunting. They pull trends out of the historian, scroll the sequence of events log, open the outage management system (OMS) to see what else happened on that circuit, and then go looking for the right procedure in a document library or a binder. If the answer is not obvious, they phone the on-call planning or protection engineer, sometimes in the middle of the night. This happens several times a shift on a normal day and constantly during an event, and the answer that gets found is usually not written down anywhere the next shift can reach it.

What it replaces or shrinks

  • Manual historian queries and trend building to answer a routine 'what happened here' question
  • After-hours calls to the on-call engineer for history the operator could have retrieved themselves
  • Searching the procedure library for the switching or relay procedure that applies to the condition in front of them
  • Re-typing the finding into the shift log by hand
  • Shrinks the repeat work where the next shift re-derives an answer an earlier shift already found

Why it is safer

The safety effect here is indirect and we will not dress it up: no field task is removed by an answer on a screen. The mechanism is that a sourced diagnosis in the control room reduces the exploratory dispatch, the crew sent to look around a circuit at night because nobody in the room could say what tripped, and it reduces trial switching done to find out rather than to fix.

Counted in units you already track:

  • Road miles driven on exploratory patrols dispatched before the cause is understood
  • Night driving hours for after-hours investigative trips
  • Switching operations performed to identify a problem rather than to isolate a known one
  • Energized area entries by crews sent to a site to gather information the control room already holds

Man-hours it gives back

Search time comes back to the operator on shift, and callout hours come back to the on-call engineer who currently answers history questions by phone.

HOURS AVOIDED PER YEAR = questions asked per shift x shifts per day x 365 x minutes the operator spends searching per question, plus after-hours engineer callouts per year x hours per callout, minus the time the operator still spends reading the cited sources before acting on the answer.

The numbers we need from you to run that formula:

  • Number of operating desks and shifts per day
  • Questions per shift that currently require a historian, log, or procedure search, and average minutes per search
  • After-hours engineer callouts per year for information requests, and average hours per callout
  • Loaded hourly rate for a control room operator and for an on-call engineer
  • Overtime or callout premium that applies to after-hours engineer support

Where the dollars come from

Cost driverHow it is calculated, from a rate you supply
Operator search laborsearch hours avoided x your loaded operator rate, at the overtime rate for any hours worked past a shift
Engineering calloutafter-hours callouts avoided x hours per callout x your loaded engineer rate, plus your callout premium
Avoided exploratory truck rollinvestigative dispatches avoided x your fully loaded cost per truck roll, which you supply, including vehicle, crew, and premium time
Time to competencynew operator ramp weeks reduced x weeks x your loaded rate for the trainer and the trainee, using your own current ramp schedule
Event write-upevent reports per year x hours currently spent reconstructing what happened x your loaded rate

Reliability and maintenance

Reliability
This does not change how many outages you have. It touches how long the diagnosis step takes before a restoration decision is made, which shows up in your customer average interruption duration index (CAIDI) and your mean time to repair (MTTR) on events where the cause was unclear at the start. Ask your operations staff how many events last year stalled on diagnosis, because that is the population this touches and nothing else.
Maintenance
Recurring causes stop hiding in the log. When the same device answers the same question three times in a quarter, that pattern is visible to the person asking, which is how a repeat trip becomes a planned vegetation or asset job instead of another momentary nobody codes.

What else it moves

WorkforceThe operators who can read a system by memory are the ones closest to retirement. This puts what they know in reach of a newer operator on their first difficult night, with the source attached so the newer operator learns why, not just what.
ComplianceEvery answer lands in the shift log with its telemetry point, log entry, or procedure cited and timestamped, which is the record you want when an event gets reviewed months later.
CustomerA faster, better supported cause determination means the estimated restoration time your customers see is built on something firmer than a guess.

What it costs you, stated honestly

You pay for the scoped engagement that builds and runs this, for read integrations into your historian, supervisory control and data acquisition (SCADA) system, OMS, and procedure library, and for your own staff time. The staff time is the part people underestimate: somebody on your side has to clean up point naming and confirm which procedure documents are current, because an assistant that cites a superseded procedure is worse than no assistant. Budget operator time in the first months to check citations before trusting them.

How to build the payback case

Payback is driven by operator search minutes and after-hours engineer callouts, both of which you can count from your own logs today. Avoided truck rolls are real but harder to attribute, so treat them as upside rather than as the basis of the case.

This is a planning model built from your rates, your call volumes, and your desk counts. It is not a vendor claim. Re-run it with your actuals after two or three months of live use, when you can count how many questions actually got answered without a phone call.
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 data quality service for the ADMS and mapping teams: it audits the distribution network model against meter and outage evidence and produces a prioritized correction work queue, each item with evidence and operational impact. 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
Geographic Information System (GIS)Esri ArcGIS Utility Network, GE Smallworldscheduled file export (CSV or CIM XML)
Advanced Distribution Management System (ADMS/DMS)Schneider EcoStruxure ADMS, GE Vernova PowerOn, Oracle NMSscheduled file export (CSV or CIM XML)
Metering (AMI head-end and meter data management)Itron, Landis+Gyr, Aclara; Oracle or Itron meter data systemsdatabase replica refreshed nightly
Customer Information System (CIS) / billingOracle CC&B, SAP IS-Udatabase replica refreshed nightly
SCADA historianAVEVA PI System, AspenTech eDNA, GE Proficyhistorian mirror (one-way feed)
Outage Management System (OMS)GE PowerOn, Oracle NMS, ADMS outage moduleread-only API
Document and knowledge storesSharePoint, procedure libraries, operator logbooksdocument upload

Data it needs from you

How it runs on your systems

Runs in your cloud account on GPU instances or an on-premises NVIDIA server, read-only through your existing data zone. Corrections are made by your GIS editors through their normal update process, never written back automatically; customer data is limited to service point identifiers.

Path to production

Weeks 1-4: data access and connection setup
Model exports and meter data connect; the ADMS export format is the usual gate.
Weeks 5-12: regional pilot
One region is audited, the top 20 connectivity errors are corrected, and ADMS performance is measured before and after.
Weeks 13-14: evaluation and go or no-go
Confirmed error rates and the measured performance gain drive the decision.
Months 4-6: production hardening
Security review, recurring audit schedule, monitoring, and mapping team training.
Month 6 onward: in production
GIS editors work the refreshed correction queue as routine, and the audit reruns after each model update.

What we need from your team

Full integration, data, and timeline detail for each use case in this scenario: UC 5.2 · UC 5.5 · UC 1.1 · UC 5.7
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 ADMS model manager and the mapping supervisor in the GridCORTEX console:

GridCORTEX ConsoleSigned in: the ADMS model manager and the mapping supervisor
Notifications
Model audit found 312 suspected errors; top item: 62 meters on feeder F-77 disagree with mapped phase per AMI voltage
Daily model refresh complete; all connected feeds healthy
Recommendation
Open the correction queue: top 20 model errors ranked by operational impact
  • 62 meters on F-77 contradict the mapped phase in AMI data
  • 8 switches mapped open while SCADA shows flow through them
  • Top fixes affect 6 feeders running automated restoration
✓ Create correction work queueModifyDecline
After you approve: Correction tasks open in the GIS editing queue with each fix's evidence attached, 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 opens correction tasks in your GIS editing queue through its API, each with evidence attached, inside your mapping team's normal workflow; nothing edits the model directly. Your editors make every change, and the ADMS model updates through your standard promotion process.

How you tell it what it cannot see

Data triggered by continuous comparison of the model against AMI and SCADA evidence; nothing to enter. Editors mark a finding as false in one click and it stays suppressed.

Live data, not stale data

Audits run nightly against the latest AMI voltages and SCADA states; each finding shows its evidence dates, and the dashboard shows the last full audit time.

Where it lives day to day

A GridCORTEX model health dashboard with a weekly email; an error blocking an ADMS application 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 Gap: Why Your Existing Systems Don't Already Do This

The fair question from any grid-mod director: "The ADMS vendor has a commissioning methodology and we hired an SI, what's new here?" Here's the honest answer, and it's why so many advanced-app licenses are shelfware.

What you own keeps doing its job

  • The ADMS, runs the apps. This is about making its model and settings worthy of them.
  • GIS, remains the connectivity source of record; it gets a ranked fix list, not a replacement.
  • The SI & vendor methodology, the project plan stands; the audit and twin replay slot into it as the evidence layer.
  • Protection engineering, every FLISR scheme and VVO setpoint is approved by your engineers before anything is enabled.

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

  • The model audit is usually a sample; it needs to be a census. SIs spot-check. The failures hide in the long tail: the recloser that's mapped to the wrong phase, the CT ratio from the as-built that never matched the field. Population-scale reconciliation of GIS vs ADMS vs SCADA telemetry vs AMI voltage finds all of it, and ranks fixes by which app they break.
  • Nobody rehearses FLISR before go-live. The twin replays hundreds of staged faults against the corrected model, every switch action validated against protection coordination, BEFORE the first live enablement. Misoperation risk is retired in simulation, where it's free.
  • VVO dies of cap-bank neglect. Half the failed VVO programs trace to sick cap banks and hunting regulators the system didn't know about. The audit finds them first; the AMI voltage channel (see The Silent Meters) verifies the result forever.
  • Pilot placement decides the program's reputation. The first FLISR feeder must have the loop topology, the device health, the comms, AND the outage history to show a visible win. Scoring every feeder on app-readiness × benefit is an optimization the project plan doesn't contain.
  • Commissioning ends; model drift doesn't. The same audit runs continuously after go-live; the apps stay on because the model stays true.
Accent, don't replace: GridCORTEX reads your GIS, ADMS export, SCADA, and AMI voltage · finds every device the model lies about · rehearses the apps in the twin · and hands your engineers a pilot plan built to win. The licenses you already paid for finally earn their keep, and stay on.
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 GIS, ADMS, SCADA, and AMI data.

🔧 The Device Census

  • Every recloser, cap bank, regulator, and motor-operated switch: comms reachability, firmware, settings vs ADMS database, phase mapping vs AMI-derived phase, CT/PT ratios vs telemetry sanity checks
  • Mismatch classification by consequence: which errors break FLISR (topology, phasing), which break VVO (cap health, regulator response), which break load flow (impedances)

🗺 Model Remediation

  • GIS fix list ranked by app impact per hour of GIS-tech effort, not by discovery order
  • Impedance calibration from AMI voltage physics: measured drops vs modeled drops, feeder by feeder
  • Model accuracy scored per feeder; the readiness heatmap that decides where apps CAN run

🎬 The Twin Rehearsal

  • 340 staged faults replayed against the corrected model: every FLISR isolation/restoration sequence checked against protection coordination and loading limits
  • VVO dry-run: setpoint schedules simulated against a year of AMI voltage history, end-of-line violations counted before they happen
  • Failures found in replay become model fixes or scheme changes, misoperations retired at zero customers

🎯 Pilot Placement & Go-Live

  • FLISR pilot scoring: loop topology, mid-feeder device health, comms latency, SAIFI history, customer density; the feeder where the win will be VISIBLE
  • VVO pilot scoring: voltage spread, cap bank fleet health, regulator headroom, AMI coverage for measurement & verification
  • Staged enablement: monitor mode → advisory mode → closed-loop, gates approved by operations at each step
  • Runs on the NVIDIA Agent Toolkit, RAPIDS for the census-scale reconciliation, PhysicsNeMo-class simulation for the replays, every finding and gate Relay-traced

Presenter's one-liner: "Everyone buys FLISR; almost nobody turns it on, because the model lies and the first misoperation kills the program. We audited every device, fixed the model in app-impact order, replayed 340 faults in the twin until it was boring, and put the pilots where the math said the win would show. The apps went live, stayed live, and the CAIDI drop showed up in the regulator's numbers."

GridCORTEX Live Scenario Demo · Synthetic data throughout, no utility, ADMS vendor, or SI is depicted · Read-only analytics; engineers approve every setting and every enablement gate · SoftServe + NVIDIA · Created by Ronnie Mauldin, NVIDIA Solutions Director, Power & Utilities, SoftServe · JUL 2026