Calendar maintenance gives every asset the same interval, so when hours run short the critical ones suffer with the rest. Scoring assets by the consequence of failure moves effort toward what holds a grade or a pressure cascade, and is best done once from the project's asset register.
A maintenance contractor's remote view should be read-only, with setpoints changed only under the site's change control. Its alerts watch the plant behind the room, such as fan speed, filter pressure drop and chilled water temperature, and the escalation path to site is written before the contract starts.
Run-to-failure is a valid maintenance strategy only when a failure shows at once. Standby pumps, fire dampers and smoke detectors fail silently and never qualify. For the rest, check GMP impact, replacement cost and restore time, then record each decision in the criticality assessment with its reason.
A single PM compliance percentage hides the overdue task QA cares about. Track PM compliance by criticality, overdue critical tasks listed by name, repeat failures on the same asset, and mean time to restore the pressure cascade from the BMS trend. All four need a CMMS asset register with criticality assessed.
Replacing one failed terminal HEPA in a live Grade C suite starts with change control and an equivalent filter. Stop work, bag the old filter, inspect the housing, then fit and leak-scan the new filter in place. Clean and check airflow, pressures, counts and recovery before release, with the scope agreed in the change record.
A single AHU replacement in a live GMP plant carries every constraint of a large project: a short shutdown window, existing connections, a pressure cascade to re-prove and change control before fabrication or any site work starts. Planned well, requalification stays short. Planned badly, a weekend job becomes a two-week deviation.
The engineering of a retrofit is rarely the hard part. Keeping the rest of the plant in its validated state while the work happens is where these projects are won or lost.