The 2015 text of EU GMP Annex 15 allows a worst-case product approach to cleaning validation if a scientific rationale is provided; criteria may include solubility, cleanability, toxicity and potency. Put them in a table per product, with the PDE and the residue load, and reopen it when a new product arrives.
A stored value of 22.4 means nothing without its sensor tag, unit, time stamp, status flag, author and audit trail. The MHRA's 2018 data integrity guidance defines metadata as data that describe the attributes of other data and provide context and meaning. Test that backups, exports and migrations carry them.
At 0.45 m/s across a 1 m clean zone the Reynolds number is about 30,000, far beyond laminar flow. ISO 14644-1 and the 2022 EU GMP Annex 1 use the term unidirectional airflow: steady velocity, roughly parallel streams. It is tested by air speed, uniformity and smoke studies, so write it that way in the URS.
Cleanroom testing is often bought late and on price, and QA first meets the tester as a signature on a report. Before the order, hold evidence of tester training, calibrated instruments, method statements to ISO 14644-1 and ISO 14644-3, a sample report and agreed raw data practice, then list the tester as an approved supplier.
PDE, permitted daily exposure, is a substance-specific dose unlikely to cause an adverse effect if a person is exposed at or below it every day for a lifetime. In a shared facility it is the basis for cleaning limits and for how much HVAC separation the design needs. Ask for it at URS stage.
EU GMP Annex 15 (2015) says a formal release for the next qualification stage should be authorised, and that conditional approval to proceed can be given where there is a documented assessment of no significant impact on the next activity. That assessment is made item by item, against the tests that follow.
An out-of-specification result falls outside the specification or acceptance criteria and starts with a laboratory assessment. An out-of-trend result is inside the limits but breaks the pattern of earlier results. EU GMP Chapter 6 says both should be addressed and investigated, which needs data recorded so that trends can be evaluated.
EU GMP Annex 15 (2015) accepts concurrent validation in exceptional circumstances, where there is a strong benefit-risk ratio for the patient. The decision must be justified, documented in the validation master plan and approved by authorised personnel. A project that is running late does not meet that test.
EU GMP Annex 1 (2022) says airflow pattern studies should be performed at rest and in operation, with video recordings retained. A usable study starts as a plan: locations and interventions from the risk assessment, defined runs, fixed camera and smoke positions, a script with acceptance criteria per shot, and a QA review.
Each SKU left on a transfer list has to be covered by validation and stability work. Before the transfer, list the SKUs with volume and market, group them by formulation, process and pack line, retire or merge the tail, choose the extremes where bracketing is justified, and order the work by value and risk.
A protocol line that says 'as per design' leaves the tester to choose the document and the tolerance on test day. EU GMP Annex 15 (2015) says protocols should define the acceptance criteria. Write a number with its unit and tolerance, the method and room state, and the source document with its revision.
An SOP that becomes effective the moment it is signed puts the next shift in deviation before anyone is trained. The 2011 text of EU GMP Chapter 4 says the effective date should be defined; the gap after approval is where training, the swap of controlled copies and the register update take place.