Indoor air quality monitoring: a practical guide for facility teams.
Milvian AI teamFacilities and infrastructure intelligence
Indoor air quality monitoring is the continuous measurement of CO2, particulates, VOCs, temperature and humidity in occupied spaces, so ventilation can be adjusted before people are affected. Good monitoring reads every zone continuously, ties readings to occupancy, and links each finding to a ventilation action that can be verified.
What indoor air quality monitoring is for
Indoor air quality monitoring answers a simple question: is the air in this space fit for the people in it, right now? A one-off survey cannot answer that, because indoor air changes by the hour. It changes with occupancy, with cleaning schedules, with outdoor conditions, with process activity and with how the ventilation system is actually running.
Continuous monitoring turns air from an assumption into a measurement. It shows when a space goes out of range, how long it stays there, and what was happening at the time. That is what lets a facility team move from reacting to complaints to fixing the cause.
What to measure
Most programs start with a core set of parameters, then add more where the space demands it.
| Parameter | What it tells you | Typical cause when high |
|---|---|---|
| CO2. | Whether ventilation keeps up with occupancy. | Too little outdoor air for the number of people. |
| PM2.5 and particulates. | Fine particles that reach the lungs. | Outdoor pollution, combustion, process dust. |
| VOCs. | Gases from products and materials. | Cleaning products, new furnishings, solvents. |
| Temperature and humidity. | Comfort, and conditions for mold. | Control faults, setpoint drift, poor dehumidification. |
| Carbon monoxide. | Combustion gases indoors. | Vehicles, generators, faulty appliances. |
| Occupancy. | Whether anyone is there to be affected. | The context every other reading needs. |
Why CO2 matters, and what it does not tell you
People exhale CO2, so in an occupied room it rises when ventilation does not bring in enough outdoor air. That makes CO2 a practical proxy for ventilation performance. It is cheap to measure, it responds quickly, and it gives a clear signal to act on: when CO2 climbs in an occupied space, bring in more fresh air.
CO2 is not the whole picture. A well-ventilated room can still carry particulates from outside or VOCs from a cleaning crew. Monitoring programs that only read CO2 miss those events. The better approach is a sensor set matched to the space: CO2 and temperature everywhere, particulates near entrances, loading docks and process areas, VOCs where products and materials are a concern.
Where sensors go, and how to trust them
Each air sensor speaks for a limited area, so the building should be divided into zones that a single sensor can represent. Mount sensors at breathing height in the occupied part of the room, away from supply diffusers, open windows, doors and direct breath, where a draft or a single person can skew the reading. Large open spaces need more than one sensor. Small rooms with high occupancy, such as meeting rooms and classrooms, usually need their own.
Sensor health is part of data quality. A sensor with a weak connection or a failing battery produces gaps and false readings that look like real events. A monitoring program should report each sensor's connection rate and battery or supply voltage, so the team knows which device needs attention before it trusts the data, and knows exactly where that device is without walking the site.
Finally, keep the history. Air quality questions often arrive after the fact: a complaint, an audit, a certification review. A continuous record by zone, with occupancy alongside, answers them with evidence instead of memory.
Duration and occupancy change the answer
For indoor air, how long a level lasts matters as much as how high it goes. A brief spike when a meeting room fills is a different problem from a level that holds for hours. That is why exposure is better tracked in zone-hours, the time each zone spends above a threshold, than as a single peak reading.
Occupancy is the other half. A CO2 event in an empty building does not matter. The same reading in a full classroom or a busy ward does. Overlaying air quality with occupancy shows when a problem actually affects people, in which zone, and starts the search for the cause. Patterns become visible too: a nightly VOC spike when the cleaning crew arrives, or high CO2 at a loading dock when trucks idle with engines running.
Snapshots can mislead. A survey taken on a quiet day, or a weekend with production stopped, can show air that looks fine and is not representative of a working day. Continuous monitoring shows the real curve.
Fixing air without burning energy
The easy answer to poor air is more outdoor air all the time. That costs energy, because every extra unit of outdoor air has to be heated, cooled or dehumidified. Demand-controlled ventilation is the better answer: raise outdoor air when CO2 and occupancy say a zone needs it, and lower it when the zone is empty. Using an economizer when outdoor conditions are mild makes the extra air cheap.
Aqueduct runs this as a loop. Zone sensors and occupancy feed a live view of each site. When a zone breaches, Aqueduct names the cause in plain language, references the standard behind the rating, and recommends the action, such as increasing ventilation or checking the AHU economizer. Where control is enabled and approved, it adjusts ventilation directly. The next readings verify the zone is back in range. Healthy air on Aqueduct is powered by aQ AirSafe.
