Your Building Has a Nervous System. Now It Has a Brain.

How it works:

Ghost heating happens when HVAC keeps running in a space nobody is using — typically 15-25% of a commercial building's total HVAC energy is pure waste like this. Aeion cross-references live occupancy signals against HVAC runtime telemetry, and when a zone sits empty for an extended stretch while its HVAC keeps working hard, that gets flagged — with the wasted energy converted into an estimated annual dollar cost.

Example: Zone 3 sits empty for 47 minutes while its air handler runs at 87% capacity — flagged as ghost heating with an estimated annual cost of $14,200.

Every building also gets an A-F efficiency grade — an overall score (here, 72/100, a "C") plus a breakdown across HVAC, lighting, equipment, and plug loads — so facilities teams know exactly where to focus first. Each grade comes with a ranked list of actionable recommendations sorted by savings-to-effort ratio, so the easiest, highest-impact fixes surface first. Top pick in this example: "Reduce HVAC setpoint 15 minutes after last occupancy detected" — low effort, $12K/year.

Weekly Digest — delivered every Monday:

Subject: Smart Spaces Weekly — Building A (Week of May 4) Consumption: 45,000 kWh (-6.2% vs prior week) Savings from automation: $890 this week Efficiency score: 87/100 (↑ from 82 last week) Top opportunity: Zone 3 ghost heating ($14,200/year if fixed) Active automation rules: 12 (saved $890 this week)

How it works:

At full scale — 15 buildings, 200 zones, 50 sensors each — that's 10,000 telemetry readings every second. Legacy building-automation systems commonly drop messages under that kind of load. Aeion buffers incoming readings and flushes them to storage every second (or every 100 messages, whichever comes first), so ingestion keeps pace under heavy load and idles efficiently when things are quiet — no messages dropped.

Every telemetry reading is stored append-only: readings can never be edited or deleted after the fact. That keeps your ESG/sustainability reporting data tamper-proof, and protects the integrity of the historical data your AI models are trained on.

Each reading is published on a topic scoped to its building, zone, sensor, and metric — for example, aeion/building-a/zone-3/temperature/current → 72.4°F or aeion/building-a/hvac/ahu-1/runtime → 87% — so any device you connect maps cleanly to a physical location.

15+ Device Protocol Adapters:

BACnet/IP — Legacy commercial HVAC (Priva, Automated Logic, Trane) KNX — European building automation standard Modbus RTU/TCP — Industrial sensors, power meters, VAV boxes EnOcean — Energy-harvesting wireless sensors Z-Wave — Home automation, specialty devices Zigbee — Philips Hue, SmartThings ecosystem Crestron — Commercial AV and control Lutron — Lighting control (Caseta, RadioRA, Quantum) Honeywell — Spyder, SpyderGL, building controllers Johnson Controls — Metasys, York Siemens — Desigo, APOGEE, FireX Schneider — EcoStruxure, PowerTag Philips Hue — Smart lighting (via Zigbee bridge) Tuya — Consumer IoT devices HTTP/REST — Generic REST devices (webhook-capable sensors)

How it works:

Upload a 2D SVG or AutoCAD (DWG) floor plan and Aeion automatically extracts zone boundaries, door positions, and room labels, building a spatial map of your facility that powers both the digital twin and indoor wayfinding.

A\ Wayfinding:**

Ask for directions between any two zones — say, lobby to Conference Room B on floor 3 — and Aeion calculates the optimal path across the building's zone graph, accounting for elevators, stairs, and doorways. Choose standard or accessible (elevator-first) routing.

Example result:

• Path: lobby → elevator → floor 3 → corridor B → Conference Room B
• Distance: 53m, estimated time: 2.5 minutes
• Turn-by-turn: "Walk to the elevator on your right (8m) → turn left, corridor B is 40m ahead"

The same spatial model powers the 3D digital twin, viewable in WebXR on Apple Vision Pro, HoloLens, or any desktop browser, with live telemetry overlaid directly on the model.

How it works:

Sensor changes reach your dashboard in well under a second — from ingestion straight through to the browser, with no polling and no stale data.

NOC dashboards subscribe to live channels for occupancy, energy, and alerts, and see updates the moment they happen:

typescript ws.onopen = () => { ws.send(JSON.stringify({ action: "subscribe", channel: "occupancy:all" })); ws.send( JSON.stringify({ action: "subscribe", channel: "energy:building_a" }), ); ws.send(JSON.stringify({ action: "subscribe", channel: "alerts:all" })); };

For example, a zone going from 12 people to empty renders in the occupancy heatmap within a quarter-second of the last person leaving.

This same real-time backbone powers cross-module coordination without any custom integration work: a Sentinel security lockdown can isolate floors via elevator control, an occupancy change can trigger Sentinel's camera tracking, and an automation emergency override can trigger HVAC emergency ventilation.