Glossary
Field-accurate definitions for the vocabulary of sensor intelligence — anomaly detection methods, integration protocols, the engine's reasoning techniques, and the operational terms operators actually use.
Foundations
Sensor intelligence is AI that turns raw building, industrial, or environmental sensor data into prioritised, explainable operational decisions.
Read entryGrounded inference means AI answers are traceable to underlying data — never hallucinated, always citable, deterministic by design.
Read entryCausal AI distinguishes correlation from causation in building sensor data — answering not just what changed, but why.
Read entryDetection
A virtual sensor is a calculated metric derived from existing physical sensors — filling data gaps without installing new hardware.
Read entryAnomaly detection in buildings is the practice of finding the signals that don't behave as expected — and ranking them by how much they should worry you.
Read entryA drift signature is the gradual change pattern in a sensor's behaviour that signals equipment degradation or calibration loss before outright failure.
Read entryA prioritised anomaly queue is the inverse of a traditional BMS alarm list — every detected issue is scored, classified, and ranked so the operator knows what to act on first, and why.
Read entryForecasting
Integrations
A BMS — Building Management System — is the control system that runs a building's HVAC, lighting, and energy infrastructure. It's also the largest underused source of operational data in modern facilities.
Read entryA BAS (building automation system) is the network of controllers, sensors and actuators that runs a building's HVAC, lighting and access control automatically. In most of the world, the same system is called a BMS.
Read entryBACnet (Building Automation and Control Network) is the dominant open protocol for building automation. Most modern BMS speak it natively.
Read entryModbus is one of the oldest still-deployed industrial communication protocols. In buildings, it's most common on energy meters, sub-meters, and legacy equipment.
Read entryOPC UA (Unified Architecture) is the modern open standard for industrial data exchange. It replaces the older OPC Classic and is increasingly used in smart buildings as well as manufacturing.
Read entryoBIX (Open Building Information Exchange) is a web-services standard for reading building automation data over HTTP. In buildings, you meet it most often as the service a Tridium Niagara station exposes.
Read entryNiagara is a software framework, built by Tridium, a Honeywell company, that unifies BACnet, Modbus, LonWorks and other building-automation protocols into one supervisory interface. Most buildings meet it as a Niagara station, a JACE controller or a Honeywell WEBs system.
Read entryAn energy management system (EMS) is the combination of metering, software and routine a business uses to monitor its energy use and reduce it. In commercial buildings it reads meter and BMS data and turns them into decisions.
Read entryO&M manuals (operations and maintenance manuals) are the document package delivered at project handover covering how to operate and maintain a building's equipment, from HVAC and electrical systems to fire and life-safety devices.
Read entryMaintenance management is the practice of planning, scheduling, and tracking the work that keeps a building's equipment running safely, reliably, and within budget, from reactive repairs to predictive interventions.
Read entryTell us what you're trying to figure out: energy drift, a BMS you don't trust, compliance you're chasing. We listen first, then tell you straight whether Explore helps. 30 or 60 minutes, your pick. No commitment either way.