Skip to content

Life Sciences

Anatomy of one OQ test case

  • Life Sciences
  • 1 min read

Anyone can sign off "interlock functions correctly" as an expected result, and the signature proves very little.

The person who executes a test is often not the person who wrote it, and was not in the design meetings where "correctly" was agreed. The test case has to carry that knowledge on the page. For an airlock door interlock it goes like this:

  1. State the objective and cite the requirement: "Verify the interlock between doors D-101 and D-102 of the personnel airlock, URS line 4.2.3." One test, one requirement.
  2. List prerequisites and instruments: IQ complete, both doors closed, stopwatch with its ID and calibration due date.
  3. Number the steps, one action in each. "Open D-101 and hold it open" is a step. "Check the interlock" leaves the method to the tester.
  4. Give the expected result as a value with a tolerance: "D-102 stays locked while D-101 is open, and releases within 2 s of D-101 closing."
  5. Leave space for the actual result, the evidence reference, signature and date. The actual result is a time or a photo ID, not a tick.

EU GMP Annex 15 (2015 text, section 2) says protocols should define the critical systems, attributes and parameters and the associated acceptance criteria, and that results failing pre-defined criteria should be recorded as a deviation. Both depend on a criterion defined before the test began.

To try a test case out, hand it to a colleague from another project and watch where they stop to ask.

Industries

Related

More on this subject

Two ideas in ASTM E2500 shorten qualification. Subject matter experts plan, execute and review verification, and in ISPE's summary of the 2025 edition the quality unit approves the plans and the final acceptance. Vendor documentation may be used as evidence if the vendor has been assessed. EU GMP Annex 15 (2015 text) allows some FAT tests to stand at IQ/OQ where justified.

InsightRead