Most misoperations are a decision nobody recorded.
Forty-three percent of protection misoperations are human performance — incorrect settings, design errors, test switches left open after commissioning. A third of control-room events never get a root cause. Indagate records the setting, the change and the reason against the asset, so the next event is a query rather than an investigation.
What a protection engineer is actually reading.
- As-left state
- What a relay, test switch or carrier ground was left as after maintenance or commissioning. NERC tracks it as its own error category because it is never on the drawings
- Relay settings
- The numbers that decide whether protection operates. Incorrect settings were 26% of all misoperations in 2024, and a third of the human-performance ones
- Root cause
- The thing 49 of 147 control-room events never got. Without it, in the regulator’s words, “the same problem likely will occur”
- Balance of plant
- Ventilation, service water, standby power and environmental monitoring — the building services around generation
- Availability
- The number the contract is written against, which is consumed by events nobody logged
- Curtailment
- Output withheld for grid reasons, which has to be excluded before underperformance means anything
NERC PRC-004, EMS event analysis, and ISO 20816.
PRC-004 requires a misoperation to be identified, its cause found and corrective action taken. NERC’s own event analysis names the leading root cause as management and organisation — explicitly including change management — and finds a third of control-room events never get a cause at all. The record is built so that the cause is captured when the change is made, not reconstructed after.
Rotating plant is still evaluated against ISO 10816 / 20816 vibration zones, and an alert cites the zone boundary rather than a tuned number, so a threshold can be argued with on its merits. Where a pack models nuclear site services, its scope is stated on the pack itself: balance-of-plant and building services only.
Reactor protection and safety instrumented functions are excluded by design. Those systems are qualified, licensed, and nowhere near a building management platform.
What an unfound root cause costs.
ReliabilityFirst’s 2025 assessment puts it plainly: misoperations whose cause was never found are “at high risk of reoccurring because the system problem was never found,” and “often the entity chooses not to correct or, when root cause is found, does forget to update based on findings.”
The cost is not the first event. It is the second, identical one — the same relay, the same as-left error, found from scratch by a different engineer because the first investigation was a phone call and not a record.
SCADA, historians and distributed assets.
Operate and maintain
Relate alarms and equipment trends to maintenance history and procedures across generation, transmission and distribution assets. Nuclear balance-of-plant and site services are in scope; safety-related systems are not.
Commission and hand over
Review substation and generation equipment records, commissioning evidence and asset handover packages. Track unresolved issues across distributed infrastructure.
Systems in scope: SCADA, RTUs, historians and CMMS. Integration scope depends on the installed system, available interfaces and agreed access. Explore the terminology.
Ninety minutes without alarms, and no alarm for that.
On 14 August 2003, FirstEnergy’s control-room alarm system failed silently at 14:14. The backup failed thirteen minutes later. IT staff knew; the operators were not told until 15:42. In the Task Force’s words, “FE’s EMS did not have such a notification system” — an alarm for the alarm system failing. Fifty million people lost power.
The pattern recurs at smaller scale every day. ReliabilityFirst’s 2025 assessment lists “as-left personnel error” — test switches left open, carrier grounds left in place, settings placed in the wrong relay — as a standing cause of misoperations, and notes that unknown-cause events are “at high risk of reoccurring because the system problem was never found.”
U.S.–Canada Power System Outage Task Force, Final Report, 2004 · ReliabilityFirst, 2025 Misoperation Assessment · NERC EMS Special Assessment, 2021.
Configure the platform around your equipment.
Start with a relevant industry pack, then tailor its equipment points, monitoring rules and alerts to your specifications. Combine it with compatible custom models and map checks to the standards and procedures your team works to.
It reads your SCADA, not a new sensor network.
Start with available historian exports or an approved OPC interface. We will check signal coverage, timestamps and access before proposing additional instrumentation.