Electronic Batch Manufacturing Record (eBMR)
A practical guide to batch evidence, regulatory scope, controlled execution and human review across regulated manufacturing.

On this page · 12 sections
- 1What is an Electronic Batch Manufacturing Record, and what does it contain?
- 2Paper BMR, scanned copies and eBMR: compare records, instructions and systems
- 3What 21 CFR 211.188 requires: master-record reproduction and execution evidence
- 4Accurate reproduction: preserve the approved master used for the batch
- 5The eBMR lifecycle: prepare, issue, execute, reconcile, review, decide and retain
- 6Part 11 scope and electronic-signature accountability
- 7Yield exceptions, investigations and batch disposition
- 8EU GMP documentation and Annex 11: validation, personnel and batch release
- 9Integrations and exception recovery: reconnect without duplicating manufacturing evidence
- 10Implementation checklist: intended use, challenge tests and operating metrics
- 11Ten eBMR failure modes and checks that expose them
- 12How V5 Ultimate assembles evidence for batch record review
How does eBMR apply to your shop floor?
Pick your industry and scale — Ask V5 rewrites the definition in your context, gives a worked example, and shows what V5 does on day one.
01What is an Electronic Batch Manufacturing Record, and what does it contain?
An Electronic Batch Manufacturing Record (eBMR) is a batch-specific electronic record of manufacturing execution against controlled instructions. It connects materials, completed operations, results, exceptions and approvals so a reviewer can reconstruct what happened. This is an operational definition, not a statutory one.
Like a batch manufacturing record on paper, it typically contains material lot identities and quantities, processing observations, equipment references, operator entries, in-process results and review decisions. Its useful unit of organisation is the batch and its operations, rather than a collection of unrelated documents.
The record may combine manual electronic entries with data captured from equipment or other systems. Capture method matters when interpreting an entry: a measured value, a calculated result and an operator’s confirmation of an activity are different kinds of evidence. Record design should make those differences visible.
Common aliases are eBR, electronic batch record and paperless batch record. Usage varies: eBR may also cover packaging, while “paperless” describes a working approach rather than proving that every attachment originated electronically.
Define the record’s boundaries before selecting software. Does it cover dispensing through bulk manufacture, include packaging, or join separately controlled records? That decision determines which teams create evidence, where manufacturing handoffs occur and what reviewers expect to find within the batch view.
02Paper BMR, scanned copies and eBMR: compare records, instructions and systems
A scanned paper BMR is not an eBMR in this article’s operational sense. It is an electronic copy of evidence first recorded on paper. Indexing, OCR and an electronic approval workflow can assist document handling without changing where the manufacturing entries originated.
| Record | Where execution is first recorded | Main review concern |
|---|---|---|
| Executed paper BMR | Paper entries | Completeness, legibility and cross-references |
| Scanned paper BMR | Paper, before scanning | Whether the copy captures the complete executed record |
| Native eBMR | Electronic entries or captured data | Whether displayed summaries retain access to underlying context |
| Hybrid batch record | Paper and electronic sources | Whether relationships and completeness are reconciled across media |
An approved master manufacturing record describes what to do; the batch record documents what was done. A manufacturing execution system may create or manage that record but is not itself the record. These are three different objects to assess during implementation: instructions, execution evidence and the supporting system.
For a hybrid record system, identify each component’s batch, step, source and controlled location. For example, electronic dispensing entries might refer to a paper equipment-cleaning log and a laboratory result held in another system. The reviewer needs the relevant log entry and result identity, not merely links to a shared folder or laboratory homepage.
Agree how attachments enter the record: who identifies them, checks that they belong to the batch and resolves missing pages or incorrect associations. A document bearing the correct product name can still belong to another batch.
During software evaluation, demonstrate one complete operation with its supporting evidence. Compare the work required to create, locate and review that evidence—not just the appearance of a finished batch-report PDF. Copy fidelity is addressed separately below.
03What 21 CFR 211.188 requires: master-record reproduction and execution evidence
For US finished pharmaceuticals subject to Part 211, the eBMR design must address the batch-record duties below. Begin with product classification and jurisdiction using the product-specific GMP scope overview; do not transfer these requirements to supplements, food, devices or chemicals solely because they use the same software template.
| Binding provision | Record duty | Practical eBMR evidence |
|---|---|---|
| 21 CFR 211.188(a) | An accurate reproduction of the appropriate master production or control record, checked for accuracy, dated and signed. | The issued batch instructions and evidence of the accuracy check, date and signature. |
| 21 CFR 211.188(b) | Documentation that each significant step in manufacture, processing, packing or holding of the batch was accomplished. | Execution evidence organised by significant operation, with supporting records connected to the operation they document. |
These duties are complementary: correctly issued instructions do not demonstrate execution, and completed tasks do not establish which instructions governed the batch. Keep both sides visible. The reproduction controls are examined in the next section.
Reviewer retrieval and challenge path — operational practice
Batch record → issued instructions → significant operation → supporting source records → reviewer questions and their resolution. Test this path with an actual historical batch, rather than checking only that each screen opens.
| Evidence to open | Reviewer check or challenge |
|---|---|
| Issued batch instructions | Retrieve the instructions used for this batch and their checking evidence. Challenge an unavailable version or a link that opens only today's master. |
| Significant operation in the batch execution history | For mixing, open the material additions, equipment identity and process observations. Which instruction-specified observations remain absent despite a completion label? |
| Ingredient-addition evidence | Retrieve each source lot and actual quantity with its units. Can each addition be traced to this batch and operation, rather than merely to the same product? |
| Laboratory records connected through the MES–LIMS relationship | Open the supporting result, not just its summary status. Do the sample, batch association and units match? Can the reviewer retrieve the underlying record? |
| Reviewer questions and resolutions | List missing entries and broken links separately from justified not-applicable entries. Assign each unresolved question to an owner and retain its resolution. |
For bulk-to-pack traceability, follow the manufacturing batch into each resulting packaged lot. A shared product code does not establish that relationship. Where an operation consumes multiple source lots, retrieve every association, not only the most recent one.
Maintain a product-specific mapping from applicable provisions to evidence locations. Identify direct capture, linked controlled records and unresolved gaps. This makes implementation work actionable without treating every field in a vendor template as a legal requirement.
Open the batch record
Confirm that the record and its linked evidence belong to the batch under review.
- Product and batch number
- Manufacturing dates
- Batch quantity and units
- Associated packaged lots
Open the issued instructions
Retrieve the version used for this batch. Challenge a link that opens only the current master.
- Instruction version issued to the batch
- Issue date
- Accuracy check, date and signature
- Batch-specific amendments
Open a significant operation
Compare the execution entries with the issued instructions. Identify missing observations even when the operation is marked complete.
- Instruction and recorded action
- Operator and execution time
- Equipment used
- Required observations and actual readings
Open the material-addition records
Trace every addition to this batch and operation, including each source lot when several lots were consumed.
- Material identity and source lot
- Actual quantity and units
- Addition time and operator
- Quantity differences and explanations
Open the supporting source records
Open the evidence behind summary statuses. Challenge broken links, mismatched samples and unexplained changes.
- Sample identity and batch association
- Test result, units and applicable limits
- Equipment records for the operation
- Changed entries, who changed them and why
- Linked deviations and investigations
Open the review questions and decision
Check how each evidence gap was resolved and who recorded the batch disposition.
- Missing entries and unavailable records
- Reasons for not-applicable entries
- Question owner and resolution evidence
- Outstanding questions
- Batch decision, responsible person and date
04Accurate reproduction: preserve the approved master used for the batch
For medicinal-product manufacturing within the applicable EU GMP context, Chapter 4, paragraph 4.20 sets the guideline expectation that the batch processing record should be based on the relevant parts of the currently approved Manufacturing Formula and Processing Instructions.
Practical implementation: bind each issued eBMR to its approved electronic instructions, parameters and approval references. Preserve that instruction version so later master edits cannot silently change what an earlier batch displays. This is a reconstruction control, not a claim that paragraph 4.20 prescribes snapshot technology.
Worked case: assume a 250 kg total formulation mass in both versions, with component percentages expressed as w/w of that total. Approved version 7 specifies 2.0%: 250 kg × 0.020 = 5.00 kg. Version 8 specifies 2.4%: 250 kg × 0.024 = 6.00 kg. A batch issued under version 7 should retain 5.00 kg as its original target; a later master revision should not silently substitute 6.00 kg. These hypothetical targets are not legal tolerances.
The guideline’s paragraph 4.20(h) calls for details, with signed authorisation, of deviations from the Manufacturing Formula and Processing Instructions. In practice, retain the original instructions, reason, authorisation and affected steps in the batch execution history; track master revisions separately in a recipe supersession log. Record a batch-specific departure rather than invisibly replacing its issued instructions.
05The eBMR lifecycle: prepare, issue, execute, reconcile, review, decide and retain
Use these seven stages as an operating model adapted to site procedures and product scope—not legally prescribed software states. Each handoff should identify the evidence owner and unresolved work.
| Stage | Main activity | Handoff control |
|---|---|---|
| Prepare | Define instructions, batch scope and evidence owners. | Assign readiness gaps before issue. |
| Issue | Assign batch identity and the instruction version. | Confirm prerequisites for execution. |
| Execute | Record actual operations and observations. | Assign incomplete steps and exceptions. |
| Reconcile | Account for expected entries and supporting records. | Resolve missing or inconsistent evidence. |
| Review | Document checks, findings and responses. | Return unresolved findings to their owners. |
| Decide | An authorised person records disposition and its rationale. | Associate the decision with the reviewed evidence. |
| Retain | Preserve the record and linked evidence. | Maintain retrieval and follow-up ownership. |
Operational completion means production tasks are finished; evidence reconciliation means expected records are accounted for; review completion means assigned checks are complete and findings have been assessed. Track these as separate milestones. A pending laboratory result, for example, need not make a completed manufacturing operation appear unperformed, but it remains an open evidence dependency.
Unresolved release-blocking findings prevent batch release. Distinguish these from permissible follow-up: work may remain open only where applicable requirements and site procedures allow it, with a documented rationale showing that batch acceptability is already established. Assign each follow-up an owner and due date. Evidence still needed to establish conformance cannot be reclassified as follow-up.
Return paths matter as much as forward handoffs
When a reviewer finds a quantity mismatch, return the finding to its evidence owner with the affected operation identified. After the response, route the changed evidence and any affected calculations back through the relevant checks. Do not silently replace the earlier review outcome. At a shift change, a shift handover report can identify outstanding findings, responsible people and the next permitted action.
Retention planning should include linked records, not just the assembled report. Assign responsibility for maintaining access when source applications change or are retired. Test retrieval of the complete evidence set for recall readiness, including records held outside the main eBMR application.
For records required under Part 211, 21 CFR 211.180(c) requires records or copies to be readily available for authorised inspection throughout retention at the establishment where the recorded activities occurred. Immediate electronic retrieval from another location meets this paragraph’s availability requirement.
- Issue InstructionsBind the batch to its authorised instruction version.
- Verify readinessCheck materials, equipment, permissions and outstanding prerequisites.
- Execute operationsCapture actions, quantities, results and relevant source references.
- Reconcile evidenceResolve missing links, incomplete entries and interrupted transfers.
- Review exceptionsAssess discrepancies, investigations and supporting execution history.
- Record dispositionDocument the authorised decision and its supporting rationale.
- Retain and retrievePreserve readable evidence, relationships and retrieval capability.
06Part 11 scope and electronic-signature accountability
An eBMR is not automatically Part 11-compliant. Assess the configured system, records and operating procedures together. Start by identifying the manufacturing records required by FDA predicate rules and which of those records you maintain electronically.
For these manufacturing records, 21 CFR 11.1(b) applies to electronic records created, modified, maintained, archived, retrieved or transmitted under agency record requirements. It excludes paper records transmitted electronically. Use the record-origin distinction covered above when describing your arrangement, then assess the maintained records and their use rather than treating the storage format as the whole scope decision.
Make the signature workflow testable
Use a record-control matrix for the Part 11 assessment: identify the record type, applicable provision, configured behaviour, test result and owner. Label binding requirements, official guidance and site practice separately. The following are practical challenge scenarios, not an exhaustive requirements checklist.
| Scenario | Practical acceptance question |
|---|---|
| Role-restricted signing | Can an unassigned operator sign a verification reserved to another role? |
| Record scope | Does the signing screen distinguish a single-step action from whole-batch review? |
| Changed content | What happens when a signed value is corrected, and which further checks are expected? |
| Interrupted signing | Does cancellation or loss of connection leave the signature status unambiguous? |
| Shared terminal | Can one person resume another person’s unattended signing session? |
Define the expected response before testing. Record any manual action still needed under the procedure, rather than treating every observed software response as acceptable. A successful test should establish what was signed and how the workflow handled the challenged condition.
Use a signature meaning statement to distinguish performing an operation, verifying another person’s work and completing a review. As a workflow design choice, identify the evidence each action covers; the same “Approve” label should not conceal materially different responsibilities.
Accountability needs an operating policy
For closed systems, 21 CFR 11.10(j) requires establishing and following written policies holding individuals accountable and responsible for actions initiated under their electronic signatures. In practice, identify who receives reports of suspected misuse, who investigates them and how the response is documented.
FDA’s Scope and Application guidance, section III.A states that FDA will enforce all predicate-rule requirements. Treat its enforcement policy separately from binding requirements. Assess any reliance on enforcement discretion against the conditions for the particular record or system rather than assuming a blanket exemption.
07Yield exceptions, investigations and batch disposition
Worked case 2: yield is not mass balance
Assume 1,000 kg actual input and 1,000 kg theoretical output: 972 kg accepted output, 12 kg samples and 8 kg waste. Masses share one measurement basis; categories do not overlap, and no work-in-process remains. The hypothetical site-defined output-yield range is 98.0–100.0%, not a regulatory limit.
- Output yield: 972 ÷ 1,000 × 100 = 97.2%, below the assumed site range.
- Accounted mass: 972 + 12 + 8 = 992 kg; 992 ÷ 1,000 × 100 = 99.2%.
- Unexplained mass: 1,000 − 992 = 8 kg; 8 ÷ 1,000 × 100 = 0.8%.
The yield comparison identifies the output-yield exception; material reconciliation identifies the remaining 8 kg discrepancy. Samples and waste account for material but do not become accepted output.
As investigation practice, check source weights, tare treatment, transfers, sampling and waste entries against the same process boundary and measurement basis. An omitted transfer or recording error is a possible explanation, not an established cause. Connect implicated test results to out-of-specification handling.
For US drug products subject to 21 CFR 211.192, the quality control unit must review and approve production and control records against established, approved written procedures before release or distribution. Unexplained discrepancies and specification failures require thorough investigation, even after distribution. If the example’s yield limits are established in the master production and control record, its low yield triggers investigation. The investigation must extend to potentially associated batches of the same drug product and other drug products; its written record must include conclusions and follow-up.
For batch disposition, distinguish supported findings from unresolved quality questions. Explaining the missing 8 kg would resolve that mass discrepancy, not establish acceptable product quality by itself.
Invalidated results remain in the batch record
For drug CGMP records, FDA’s nonbinding data-integrity guidance, Question 2, explains that excluding an invalidated test result from the quality unit’s specification-conformance decision requires a valid, documented, scientifically sound justification. Even after legitimate invalidation, the full CGMP batch record provided to the quality unit includes the original invalidated data and the investigation report justifying invalidation. Exclusion from the conformance decision is not exclusion from the batch record.
08EU GMP documentation and Annex 11: validation, personnel and batch release
Chapter 4 and Annex 11 set out EU GMP guideline expectations for medicinal-product operations. Chapter 4, paragraph 4.10 says to identify each manufacturing activity’s record and its location, with secure controls protecting integrity throughout retention and validated where appropriate. Connect the batch-record inventory to the data-integrity assessment so record locations and controls can be examined together.
The Annex 11 Principle distinguishes two expectations: the application should be validated, and the IT infrastructure should be qualified. Section 1 calls for risk management throughout the computerised-system lifecycle, considering patient safety, data integrity and product quality. Decisions on the extent of validation and data-integrity controls should rest on a justified, documented risk assessment.
Annex 11, section 2 calls for close cooperation between relevant personnel, including Process Owners, System Owners, Qualified Persons and IT. Personnel should have appropriate qualifications, access levels and defined responsibilities for their duties. In practice, distinguish application validation evidence from infrastructure qualification evidence, and connect role assignments to relevant training records.
Annex 11, section 15 addresses systems used for recording certification and batch release: the system should allow only Qualified Persons to certify release, identify and record the person releasing or certifying the batches, and use an electronic signature for that action. Preserve that attributable decision in the electronic release record.
| Annex 11 provision | EU GMP guideline expectation |
|---|---|
| Section 9: Audit trails | Consider system-generated records of all GMP-relevant changes and deletions based on risk. Document reasons for those changes or deletions. Trails need to be available, intelligible and regularly reviewed. |
| Section 16: Business continuity | Provide continuity arrangements for critical processes during system breakdown. Activation time should be risk-based and appropriate to the system and process; arrangements should be documented and tested. |
| Section 17: Archiving | Check archived data for accessibility, readability and integrity. Relevant system changes should include ensuring and testing the ability to retrieve archived data. |
Use an audit-trail review workflow to assign reviewers, define scope and record how findings are handled.
Certification, archive and continuity challenges — operational practice
Build on the reviewer retrieval path above with focused challenges:
- In a controlled test environment, attempt certification with an account not authorised as a Qualified Person. Confirm that it is blocked; retain evidence of the attempted action and a permitted signed certification.
- After a relevant system change, retrieve an older archived batch. Check that attachments open and signature associations remain usable; record missing or unreadable items.
- Exercise a simulated outage separately. Record activation time, activities that can continue and how outage entries are reconciled after recovery.
For comparison, FDA’s 2003 Part 11 guidance, III.A, describes broader legacy-system enforcement discretion only under certain circumstances. Its Legacy Systems section specifies all of these conditions: operation before August 20, 1997; applicable predicate-rule compliance before that date and currently; and documented evidence and justification of fitness for intended use, including acceptable record security and integrity where applicable. Age alone is insufficient when assessing a Part 11 system. Keep this US enforcement-policy assessment separate from the EU guideline review.
09Integrations and exception recovery: reconnect without duplicating manufacturing evidence
As an implementation practice, define record ownership across manufacturing, laboratory and equipment systems before connecting them. For each interface, including MES–SCADA integration, specify:
| Interface field | Define and verify |
|---|---|
| Authoritative source | System owning the original evidence |
| Source system, event ID and revision | Unique event identity and transferred version |
| Units | Source units and conversion rule |
| Event time and receipt time | When execution occurred versus transfer arrived |
| Acknowledgement state | Pending, accepted or rejected; define acceptance |
| Reconciliation owner | Person responsible for resolving mismatches |
Worked case 3: one dispense, three deliveries
Assume one physical dispense of 5.000 kg creates event E104, delivered once and retried twice. Recognise retries by source system and event ID; reconcile changed revisions or payloads rather than silently discarding them. The expected posting is one 5.000 kg consumption, not 3 × 5.000 kg = 15.000 kg. Retain delivery history linked to the accepted event. Values and identifiers are hypothetical.
A missing message does not prove that an operation did not occur. If physical state is uncertain, reconcile before repeating work. Test interruption before capture, after capture but before acknowledgement, and during recovery; check both physical execution and consumption postings.
FDA’s nonbinding drug-CGMP data-integrity guidance, Question 10, explains that static records may meet retention requirements when they are originals or true copies. A printout cannot preserve the dynamic, reprocessable FT-IR record. Assess dynamic versus static records as evidence preservation, not merely file format.
Finally, distinguish backup restoration from long-term record retrieval. Test restored relationships between batch, sample, result and source event, and access through the appropriate reader or originating system. A successful backup job alone does not demonstrate that retained batch evidence remains usable.
10Implementation checklist: intended use, challenge tests and operating metrics
For drug-CGMP electronic master-record creation, FDA’s nonbinding data-integrity guidance, Question 3 recommends intended-use validation and studies proportionate to automated-system risk. It also recommends assessing and mitigating non-CGMP functions’ effects on CGMP operations in shared systems. Use this practical computerised-system validation checklist to define implementation and acceptance, not as a universal regulatory test script.
- Define records and intended use. Identify products, jurisdictions, electronic records, workflow boundaries and excluded functions; document the basis for each scope decision.
- Map evidence ownership. Assign the authoritative source, accountable owner and reconciliation responsibility for manufacturing, laboratory and equipment data.
- Configure instructions and roles. Identify approved instruction versions, permitted actions, reviewer responsibilities and configuration dependencies before testing.
- Establish acceptance criteria. Define expected results and blocking defects before testing; connect requirements, risks and tests through a validation traceability matrix.
- Challenge workflows and interfaces. Test calculations and units, instruction-version binding, incomplete execution, role restrictions, corrections, evidence linkage and interrupted or duplicate interface deliveries.
- Test migration and retrieval. Reconcile migrated records against source evidence, including metadata and relationships; retrieve archived batches and verify that reviewers can reconstruct them.
- Train reviewers. Use representative complete and defective records; assess whether reviewers identify missing evidence and explain escalation decisions.
- Control rollout and changes. Define pilot scope, go-live approval, rollback arrangements and support ownership. Assess changes and select regression tests before deployment.
Define cohorts, interfaces and reporting periods before measuring. Review hours per batch = total active reviewer hours for batches whose review is complete, including rework but excluding queue time, divided by their count. Evidence gaps per reviewed batch = distinct missing-evidence findings divided by reviewed batches. Duplicate-delivery fraction = duplicate delivery attempts divided by all attempts on named interfaces. Use matching populations for each ratio. Recovery time per incident = minutes from interruption detection until reconciled operation. Separate measured results from targets; these metrics promise neither efficiency gains nor compliance.
11Ten eBMR failure modes and checks that expose them
Alongside the checks covered above, use these proposed practices to examine data integrity and attributable recording. They are not legal requirements or reported product defects. Rows marked “Test” are system challenges: use an approved protocol in a controlled test environment, never production records.
| Failure mode | Check |
|---|---|
| Copied-forward operator or time values | Review: Compare entries with the authenticated operator and actual execution timestamps; investigate unexplained identical values. |
| Sequence checks enforced only in the screen | Test: Attempt a prohibited transition through each supported API or import route; inspect the saved state against the approved workflow. |
| Ambiguous time-zone ordering | Review: Compare timestamps, UTC offsets, event sequence and clock synchronisation evidence across systems; flag unresolved ordering, including daylight-saving transitions. |
| Excessive rounding | Review: Recalculate from retained precision using the approved rounding rule; check whether display rounding changes the acceptance decision. |
| Default zero hiding an unmeasured value | Test: Leave a measurement unentered; verify that storage, exports and calculations distinguish missing values from measured zero. |
| Unclear specification revision at assessment | Review: Retrieve the specification revision recorded for the assessment, not merely the version currently available. |
| Role changes not reflected in active sessions | Test: Revoke a logged-in test user's permission; retry the restricted action against the defined revocation timing. |
| Attachments inaccessible after closure | Review: After temporary links expire, reopen a closed batch through normal reviewer navigation and retrieve its attachments. |
| Generic correction reasons | Review: Compare the reason with before-and-after values; determine whether it explains the specific change. |
| Closure status attached to the wrong batch identity | Review: Trace the closure event's stored batch identifier and compare it with the batch displayed. |
For review by exception, document which failures generate visible exceptions and which need separate reviewer checks.
12How V5 Ultimate assembles evidence for batch record review
Frequently asked questions
Q.Can an eBMR include handwritten attachments, and how should the official record boundaries be documented?+
Yes. As record-design practice, apply the boundaries described in the paper-versus-electronic section through a hybrid record system procedure. Identify each attachment type, authoritative format, batch identifier, responsible reviewer and storage location. Define who adds an attachment, how its page count is checked, and how an amended or replaced attachment is linked to its predecessor. Test whether a reviewer can retrieve the complete attachment set from the batch record without relying on the operator’s personal files or memory.
Q.What supplier evidence should a buyer request when an eBMR product is described as Part 11-ready?+
Start with the scope assessment in the Part 11 section. As procurement practice, request version-specific control mapping, configuration assumptions, test evidence, known limitations and responsibility assignments. Ask for a demonstration using your proposed roles and representative records, then capture gaps in requirements traceability. EU GMP computerised-system guidance in Annex 11 §3.1 calls for formal third-party agreements with clear statements of third-party responsibilities. Include support access, change notification and evidence retrieval in the supplier discussion.
Q.What batch evidence should a contract manufacturer provide when the product owner cannot access its execution system?+
For disposition responsibilities, see the lifecycle section. As outsourcing practice, agree an indexed evidence package before production: executed instructions, material genealogy, equipment references, results, deviations, relevant change history and approval identities with timestamps. Specify file formats, attachment identifiers, transfer checks and how the recipient acknowledges missing items. Arrange a trial transfer using a representative completed batch. Define response times for source-data requests and review questions in the agreed batch review process, including access arrangements after the manufacturing contract ends.
Q.What should a manufacturer verify before retiring a legacy system after exporting its batch records?+
As decommissioning practice, use the implementation checklist and the “Reviewer retrieval and challenge path” in the 21 CFR 211.188 section. Define archive acceptance criteria, reconcile batch counts, and select corrected records, unusual attachments and batches spanning software versions. Retrieve each selected batch from the archive, follow a result to its source evidence, and compare approval and change histories with the legacy record. Perform this challenge without an active legacy session. Record discrepancies and ownership in the decommissioning plan. Resolve preservation failures before authorising retirement.
Q.How should a reviewer distinguish an invalidated test result from a result that remains usable for a quality decision?+
Use the investigations section for the invalidation principles. As review practice, follow the out-of-specification workflow from the result identifier to the batch, sample, preparation and test run. Match those identities to the investigation’s stated scope and conclusion; a shared batch number alone does not establish that the conclusion covers this result. Compare decision statuses in the laboratory record, investigation and batch summary. If they conflict or applicability remains unclear, refer the discrepancy to the responsible reviewer and resolve it before completing review.
Q.Which changes to recipes, calculations, roles or interfaces should trigger an assessment of additional testing?+
As change-control practice, screen changes to recipe quantities, unit conversions, rounding, approval roles, instrument mappings and interface retry logic. Use the implementation checklist section for the validation approach. For each change, identify affected versions and open batches, then select tests for the altered behaviour and plausible failure paths. Include permission denial, interrupted transfer and recalculation cases where relevant. Use the recipe supersession log to trace recipe versions, and record the reason for adding, reusing or excluding each test.
Q.How can a supplements, food, device or chemical manufacturer select applicable record rules without borrowing drug GMP requirements wholesale?+
As scoping practice, build a product-and-market applicability matrix with qualified regulatory personnel. List product classification, intended use, manufacturing activities, destinations and the records generated at each site. For each record, document the governing requirement, applicability rationale, approval responsibility, retention basis and electronic handling. Use the cGMP overview for orientation. Keep binding requirements, guidance and contractual commitments in separate columns, assign an owner to unresolved classifications, and revisit the assessment when products, claims, activities or markets change.
Q.When is a dedicated eBMR workflow sufficient, and when should a manufacturer evaluate a broader MES?+
As an architecture decision, start with representative manufacturing scenarios rather than a feature count. A dedicated eBMR workflow may suit a scope centred on instructions, execution evidence and review. Evaluate a broader manufacturing execution system when dispatching, equipment coordination, work-in-progress visibility or material movements also need coordinated control. Test the proposed boundaries using an outage, rework and a schedule change. Compare interface ownership, duplicate data entry, support responsibilities and upgrade testing before deciding which architecture your team can operate and maintain.
Primary sources
- 21 CFR 211.188 — Batch production and control records
- 21 CFR 211.192 — Production record review
- 21 CFR 211.180 — Records and reports: general requirements
- 21 CFR 11.1 — Part 11 scope
- 21 CFR 11.10 — Controls for closed systems
- EudraLex Vol 4 Part I Chapter 4 — Documentation
- EudraLex Vol 4 Annex 11 — Computerised Systems
- FDA guidance: Part 11, Electronic Records; Electronic Signatures — Scope and Application (2003)
- FDA guidance: Data Integrity and Compliance With Drug CGMP — Questions and Answers (2018)
Explore this topic
eBMR sits inside this topic cluster in our glossary. Every neighbour is one click away.
Master and executed records that prove a batch or device was made to spec.
Want to see how eBMR could fit into your own records and workflows? Explore the related V5 pages or talk to our team about what applies to your operation.
