SOP: Remote and Virtual FAT Witnessing

This is a ready-to-use SOP for witnessing a Factory Acceptance Test remotely, by live video, instead of traveling to the vendor’s site. Replace every <<FILL: ...>> placeholder with your own specifics, set your document numbers and dates, and route it through your normal document control, review, and approval. A worked filled specimen follows the template so you can see how a completed version reads. Verify each cited regulation against the current source before you rely on it.

Document control header

1. Purpose

This procedure defines how <<FILL: COMPANY NAME>> plans, controls, and documents a Factory Acceptance Test (FAT) that is witnessed remotely, in whole or in part, by live video rather than by an owner representative physically present at the vendor’s site. The objective is to make a remotely witnessed FAT (vFAT) produce evidence of the same quality as an on-site FAT, so it can be relied on for the ship decision and, where the conditions of <<FILL: cross-reference FAT/SAT protocol SOP-ID>> are met, reused into qualification.

2. Scope

This procedure applies to any FAT, or any portion of a FAT, witnessed by an owner representative who is not physically present at the point of test, using live video, screen share, or a comparable real-time remote connection. It covers platform and connection requirements, camera control, the rule for what may be witnessed live versus reviewed from a recording, what is deferred to an on-site check, data-integrity confirmation over a remote link, redline handling, and connection-loss contingency. It does not cover Site Acceptance Testing, which happens at the owner’s site by definition, or a fully vendor-self-executed test with no live owner witnessing at all, which is not a vFAT under this procedure and is handled instead as unwitnessed vendor data per <<FILL: cross-reference FAT/SAT protocol SOP-ID>>.

3. Responsibilities

4. Definitions

• Remote or virtual FAT (vFAT): a FAT witnessed by an owner representative connected by live video and screen share rather than physically present at the vendor’s site.
• Live witnessing: the remote witness observes the condition being forced and the result occurring in real time, with the ability to direct the camera and ask for a repeat or a different view.
• Recorded evidence: video or screen-capture footage reviewed after the fact. It supports a live-witnessed result; it does not by itself constitute witnessing.
• Tactile or dimensional check: a verification that depends on physical handling, measurement, or inspection (surface finish, weld inspection, physical fit, materials confirmation by touch) that a camera cannot settle.
• Ground rules record: the document, agreed and signed before the session, that fixes camera control, live-versus-recorded rules, redline handling, and connection-loss contingency for that specific vFAT.

5. Procedure

5.1 Decide whether a vFAT is appropriate

1. Confirm the system’s risk profile against <<FILL: criticality / novelty criteria, e.g. component criticality assessment>>. Highly novel, first-of-kind, or directly product-contact systems default to on-site witnessing for their GxP-critical and tactile test cases; a vFAT is more readily justified for platform equipment, software-heavy control-system testing, or systems the owner already qualifies routinely.
2. Identify, before booking the session, which test categories will be witnessed remotely and which require an on-site step (see section 5.4). Record this split in the ground rules record.
3. Confirm the vendor’s site can support the platform and camera requirements in section 5.2. Do not proceed on a promise; confirm with a technical check call in advance.

5.2 Fix the ground rules in writing, before the session

1. Agree and record, in the ground rules record: the video/screen-share platform, minimum camera resolution and network bandwidth, who controls the camera (the remote witness directs framing; the vendor operates it, but does not choose what is shown without direction), whether the session is recorded in addition to being live, how the tester signs the protocol, and how the remote witness records contemporaneous concurrence with each result.
2. Agree the redline process: how a protocol change identified during the session is captured, by whom, and how the remote witness’s approval of that change is recorded in real time, not reconstructed afterward.
3. Agree the connection-loss contingency: what happens if video, audio, or screen share drops mid-step (pause, wait for reconnection, do not accept “it passed while you were disconnected” as a result).
4. QA approves the ground rules record before the session is scheduled.

5.3 Witness the session

1. Witness, do not narrate: watch the actual condition forced and the actual result occurring on camera, in real time; do not accept a verbal report of a result that happened off-camera or before the session.
2. Direct the camera actively. For an I/O or alarm test, require the camera to show the field device or terminal and the HMI at the same time, or in a sequence you control, not only the screen the vendor chooses to show.
3. For every test case intended for reuse into IQ/OQ, require live witnessing. A recording submitted afterward supports the record; it does not replace live witnessing for a test whose result you intend to credit into qualification.
4. Confirm data-integrity behaviors on the live system, not on slides or a canned demonstration: force an unauthorized action, watch the audit trail entry appear with the correct user ID and timestamp on camera, and confirm the operator cannot edit or disable it.
5. Record actual results contemporaneously as they happen, exactly as in an on-site FAT. Note in the record that the result was witnessed remotely and by whom.
6. Log every observation as a numbered, classified punch item, the same as an on-site FAT.
7. If the connection drops during a step, pause the step, log the interruption with start and resume time, and re-execute the step from its start once reconnected; do not accept a partial or reported result across a connection loss.

5.4 Defer tactile and dimensional checks

1. Identify, before the session, every planned test case that is tactile or dimensional (surface finish, weld and borescope inspection, physical fit, material confirmation by touch, physical damage checks).
2. Do not attempt these over video. Defer them to either a targeted on-site visit before ship, or to re-verification at SAT, and record which path was chosen and why.
3. Where a targeted on-site visit is used, it is a short, scoped visit for the deferred items only; it does not need to repeat the functional and software tests already witnessed live.

5.5 Close out and control the record

1. Reconcile every step and every punch item, with no blanks, exactly as an on-site FAT closeout.
2. The executed protocol with contemporaneous actual results remains the primary record. Attach the session recording, where one exists, as supporting material, not as a substitute for the executed protocol.
3. Attach the signed ground rules record and the remote witness’s signed concurrence to the FAT report.
4. State explicitly, in the FAT report, which test cases were witnessed live remotely, which were deferred to an on-site check, and the outcome of that on-site check.

6. Acceptance criteria

A remotely witnessed FAT is acceptable when all of the following are true:

• The ground rules record was agreed and QA-approved before the session began.
• Every test case proposed for reuse into IQ/OQ was witnessed live, not reviewed only from a recording.
• Data-integrity behaviors (access control, audit trail) were confirmed live on the running system.
• Every tactile or dimensional check was identified in advance and deferred to a stated on-site path, not skipped.
• No connection-loss gap in the record; any interruption is logged and the affected step was re-executed from its start.
• The FAT report states which portions were witnessed remotely and references the ground rules record and the deferred on-site checks.

7. Records generated

8. References

EU GMP Annex 15 (Qualification and Validation), 2015 revision, sections 3.4 to 3.7 (FAT and reuse of qualification work). ASTM E2500, Standard Guide for Specification, Design, and Verification of Pharmaceutical and Biopharmaceutical Manufacturing Systems and Equipment. ISPE Baseline Guide Volume 5, Commissioning and Qualification (second edition, 2019). 21 CFR Part 11 / EU GMP Annex 11, for audit trail and data-integrity behaviors confirmed during the session.

Confirm the current version and clause numbers of each reference before issue.

9. Revision history

10. Approvals

Filled specimen

The following shows the ground rules and the closeout completed for an example vFAT of a single-use bioreactor control skid, so you can see the level of detail an inspector expects. The company, system, and numbers are illustrative; replace them with your own.

Ground rules record extract:

Session outcome:

The specimen shows the two moves that make a vFAT defensible: every test intended for OQ reuse was witnessed live and on camera, not taken from the recording, and the one connection interruption was logged and the step re-executed rather than accepted on trust.

Common inspection findings this SOP prevents

• A vFAT run with no ground rules record, so who controlled the camera and what was actually shown cannot be reconstructed.
• Test results credited toward qualification from a recording reviewed afterward, with no live witnessing on the date claimed.
• Tactile or dimensional checks silently skipped because “we did the FAT remotely,” with no on-site check ever scheduled to close them.
• A connection loss during a critical test with no log of the gap, and the result accepted as if nothing happened.
• Data-integrity behaviors demonstrated on a slide or a pre-recorded clip instead of forced live on the running system.

How to adapt this SOP

1. Set your document number, owner, and effective date in the header.
2. Set your own criteria in section 5.1 for which systems and test categories default to on-site witnessing versus qualify for a vFAT.
3. Name your actual video/screen-share platform and its minimum technical requirements in the ground rules record.
4. Point the cross-references in sections 2 and 6 to your real FAT/SAT protocol and component criticality assessment.
5. Confirm every regulation in section 8 against the current published version before issue.