Environmental monitoring in a GMP facility is a programme, not a test. It has a written plan, defined points, grade-based volumes, alert and action limits, a defined response to excursions, and a periodic review that changes the plan when the data says it should. The instrument is one component of that loop.
This article sets out how the loop is built for viable air monitoring, and where a programmable sampler fits. It is written by the engineering team at HENG MEI Intelligent Manufacturing for pharmaceutical quality and microbiology teams.
Start from the grade, then the volume
Cleanroom classification drives the expected bioburden, and the expected bioburden drives the volume that keeps counts in a countable range. A volume that is too small yields zero counts that cannot distinguish a good room from a marginal one; a volume that is too large produces overcrowded plates and unreliable counts.
| Grade | Typical volume per point | Sampling time at 100 L/min | Note |
|---|---|---|---|
| A / B at rest and in operation | 1,000 L | 10 min | Low counts expected; volume improves sensitivity |
| C | 500 L | 5 min | Common compromise |
| D | 200–500 L | 2–5 min | Higher background; watch plate crowding |
| Isolator interior | ≤500 L | ≤5 min | Minimise disturbance of internal airflow |
These volumes are a starting structure. The definitive values belong in the site's validation documentation and should be justified there, not inherited from a template.

Alert and action limits do different jobs
An action limit is a regulatory or validated threshold: exceeding it requires a documented investigation and, in many cases, a batch impact assessment. An alert limit is an internal early-warning level set below the action limit, chosen from the site's own historical data.
- Action limit exceeded — investigate, assess batch impact, evaluate the room's trend, consider re-sampling and additional cleaning.
- Alert limit exceeded — review trend, check whether the point is drifting, increase sampling frequency at that point.
- Both breached repeatedly at one point — treat as a condition rather than an event; the cause is usually airflow, equipment or a cleaning gap.
Setting alert limits from the site's own data, rather than copying another facility's numbers, is what makes them useful. A point that habitually runs at half the alert limit is a candidate for a lower alert level, not a comfortable pass.
The documentation loop
Four documents keep the programme in a validated state, and each answers a different auditor question.
| Document | Answers | Content |
|---|---|---|
| Monitoring plan | What is sampled, where, how often, at what volume | Point map, grades, volumes, frequencies, media |
| Sampling record | Did this run happen as planned | Point, date, time, volume, operator, instrument |
| Incubation and count record | What grew | Medium, temperature, duration, colony count, calculated CFU/m³ |
| Excursion record | What was done about it | Investigation, batch impact, corrective action, re-test result |
Where a sampler prints a slip at the point of sampling, the sampling record and the plate travel together, which is the simplest way to stop plates and paperwork from diverging. Where the instrument stores and exports data, the record can be reconciled against the plan without transcription.
Instrument qualification and calibration
A viable air sampler in a GMP facility is subject to the same thinking as any other measuring instrument: it must be fit for purpose, and that fitness must be shown. Three items are commonly requested at purchase — installation and operational qualification documentation, a flow calibration certificate, and a service arrangement that maintains calibration over the instrument's life.
Flow is the parameter that matters most. Reported concentration is inversely proportional to volume, and volume depends on flow. A sampler drifting 5% low over-reports concentration by about 5%, consistently, across every result — a bias that trend analysis will never reveal because it looks like a slightly dirtier room.
Review frequency and the trap of a static plan
Environmental monitoring plans go stale. A room is reconfigured, a filling line is added, a wall is moved, cleaning chemistry changes. Each of those events changes the airflow and the contamination profile, and a plan that has not been reviewed is now measuring the previous room.
A practical review cycle is annual, plus a triggered review after any change to layout, HVAC, cleaning regime or product. The review should look at the data by point: points that have never exceeded an alert limit for two years may be sampled less often, freeing effort for points that drift. A plan that responds to its own data is evidence of a functioning quality system; one that has not changed since qualification is evidence of the opposite.
How HENG MEI supports GMP monitoring
HENG MEI Intelligent Manufacturing supplies the programmable HM-KF2 sampler used in grade-based monitoring, with 1–9999 L volume setting, a 12 m/s impact velocity, printed sampling slips and USB data export, alongside the basic HM-KF1 and surface and product microbial methods. The instrument-level logic for choosing between the two is set out in our programmable sampler guide.
To plan a programme, tell our sales contact the area grades, points per room, the volumes in your validation documentation, and whether qualification and calibration certificates are required with the instrument.
Contact: Kyra, Sales Contact, Kyra@hmdzkj.cn, +86 13792642016 (Tel / WhatsApp / WeChat).
Article address:https://www.molecularbio.cn/news/gmp-viable-air-monitoring-programme.html




