Each product’s annual product quality review, drafted the morning after its period closes, with every number cited
Each product's annual quality review is drafted the morning after its period closes, with every number cited.
See one case, screen by screen ↓Every product needs its review once a year: every batch made and its disposition, every deviation, OOS and change, complaints, returns and recalls, the stability programme, last year’s actions, qualification status and the quality agreements — all in one signed report, due 90 days after the period ends under the site’s review SOP. With 24 products across three plants and a contract manufacturer, a review is always due.
Most of the author’s hours go before the first sentence: exports from ERP, MES, LIMS and the QMS, counts that do not match between two systems, capability worked out in spreadsheets. The trending that the review exists for — a dissolution shift after a supplier change, a complaint cluster from one packing line — gets the hours that are left, and the QP and Head of Quality get the report late in the clock.
Estimated split for one product review, by hand and with the solution. The totals match the 122 h by hand and the 41 h median target.
Specialist agents collect, trend, reconcile and write every review; the product quality lead decides each finding and signs.
Corventa 20 mg film-coated tablets, made at Plant 2 — Shannon: 186 batches between Aug 1, 2025 and Jul 31, 2026, review due Oct 29. Here is what happens, screen by screen, in the working solution.
Priya opens the review year: 24 product reviews due across 3 plants and 1 contract manufacturer, 16 approved and 8 in progress. Five things need her decision — 4 findings and 1 count to reconcile — and one review, Talvexa oral solution, is 9 days late waiting for the QP signature.
Agents at work: “Reconciliation checker — complaint count differs for Corventa 20 mg — 58 vs 57.”
Veltrimab drug substance closed its period on Sep 30. The collectors pull 58 production orders and 58 batch records, 27 deviations, 1 OOS, 18 changes, 290 release results, 3 stability studies, 2 variations and 4 agreements. The trend analyst raises one finding — host-cell protein trending up — the reconciliation checker confirms all 37 counts agree in two systems, and the writer drafts 14 chapters with 388 cited numbers.
“Seven agents pull 58 batches and every linked record from ERP, MES, QMS and LIMS, trend them, reconcile every count and write all 14 chapters.”
Chapter 1 summarises the period: 186 batches, 181 released, 2 rejected and 3 awaiting disposition at period end. Against last period: 41 deviations (6 major) against 47, 4 OOS with 1 confirmed, 23 changes, 58 complaints against 61, no recalls. Three findings wait for her, and the conclusion is left blank on purpose.
“The writer drafts the conclusion once the 3 remaining findings are decided — the conclusion is yours, built from your decisions.”
Six critical quality attributes are trended over 186 batches; five are stable and capable. Dissolution at 30 minutes shifted down from February 2026 — the batch mean fell from 94.1 % to 89.6 %, and capability since the change is Cpk 1.08 against the 1.33 target in SOP-QA-118 §7.3. Every batch except CV25-1188 still met the Q = 80 % limit; the control chart marks the day it started: “Feb 9 · new cellulose supplier”.
She clicks 89.6 %. The evidence pane shows how it was counted — the mean of batch-mean dissolution at 30 min for the 95 batches made from February 2026 — which agent counted it and when, the system it was counted in, and all 95 batch records behind it.
“Every number in this review is counted from records.”
The shift starts with the first batches made with cellulose from the second supplier, Halvern Cellulose, under change CC-26-0412. Halvern’s certificates of analysis show a d50 of 128–141 µm, against 104–112 µm for lots from the original supplier. The trend analyst recommends a CAPA: particle size at the incoming test, dissolution on every batch for 6 months, and MSAT to assess a supplier-specific granulation water amount. Priya chooses “Raise CAPA as recommended” — CAPA-26-041.
The change control’s own risk assessment, cited: “Particle size distribution is comparable (d50 within ±10 %); no impact on dissolution expected.”
The complaint system counts 58 complaints; the QMS product link counts 57. The checker found the difference: CMP-26-0311 (discolouration, May 2026) is linked to Corventa 40 mg in the QMS, but the batch on the complaint photo, CV26-1058, is a 20 mg batch. Priya relinks it — 58 in both systems.
A count that differs between two systems blocks sign-off until a person resolves it.
Impurity B at 18 months, 0.31 %, is above the trend of earlier pulls, but the projection to the 36-month shelf-life is 0.42 % (upper bound 0.47 %), inside the 0.5 % limit: Priya keeps it on watch with a 21-month pull. Five broken-tablet complaints from April to June all came from batches packed on blister line 2 after its sealing temperature was lowered to 185 °C — she raises CAPA-26-042 to verify forming depth and add a start-up check.
“Keep on watch: add a 21-month pull and review again at 24 months. No change to shelf-life.”
With the three findings decided, the writer drafts the conclusion in chapter 1, naming each finding with the decision and the CAPA it raised, and noting that both corrective actions will be checked in next year’s review.
“The manufacturing process for Corventa 20 mg and its controls remain in a state of control: every released batch met its specification, and the specifications remain appropriate.”
Before signing, the solution checks that the findings are decided (3 of 3), the counts are reconciled and chapters 2–13 are ready. Priya signs as author with her user ID and password, and the meaning recorded with it: “I prepared this review; its content and conclusions are accurate.” The review goes to Dr. Omar Haddad as Qualified Person, then to Dana Okafor, Head of Quality.
The same review, seen by the five people who carry it — what their year looked like, and what it looks like now.
Specialist agents collect, trend, reconcile and write every review; the product quality lead decides each finding and signs.
Pulls production orders, usage decisions, yields and batch-record exceptions for the review period from ERP and MES.
Collects deviations, OOS/OOT investigations, CAPA, change controls, complaints, returns and recalls linked to the product, with the links between them.
Pulls release results for every critical quality attribute and all stability pulls, with condition and time point, from LIMS.
Lists the variations filed, approved or refused in the period, open post-approval commitments, and the technical and quality agreements in force for each site.
Computes capability (Cpk), run rules, a Pareto of deviations, stability regressions to shelf-life and complaint clusters, and raises each adverse trend as a finding with a recommended action.
Counts every number in the review from at least two systems — ERP against QMS disposition, complaint system against QMS link, MES against LIMS lots — and opens an item listing the records that explain any difference.
Fills the 14-chapter template with plain statements, charts and a citation behind every number, and drafts the conclusions from the decisions taken.
Decides each finding and signs. The agents propose; a named person decides.
Ask about any product review, trend or number — every answer cites the record it came from. It can also add a trending rule or link findings across products.
Why did dissolution drop after February?
The drop lines up with the second cellulose supplier. Batch-mean dissolution at 30 min fell from 94.1 % (91 batches, Aug–Jan) to 89.6 % (95 batches, Feb–Jul). The first lower batches are the first made with Halvern cellulose under CC-26-0412, whose particle size is larger — d50 128–141 µm against 104–112 µm. Every batch still passed except CV25-1188, which failed for a different reason (over-wetted granulation).
Which other products use the new cellulose supplier?
Two products use Halvern cellulose: Corventa 40 mg — the same shift, dissolution Cpk 1.17 since February, raised as F-020-1 in its draft review — and Corventa 20 mg, finding F1, Cpk 1.08. The contract-made Corventa 20 mg at Bracken uses only the original supplier, and its dissolution Cpk is 1.41 — no shift. I have linked F-020-1 to F1, so one CAPA can cover both strengths.
Add a rule: flag complaint clusters of 3 or more in 60 days
Added for your approval as a trending rule: complaint cluster — 3 or more complaints with the same failure within 60 days, applied to every product from the next agent run. With it, the broken-tablet cluster on Corventa 20 mg would have been flagged on May 21, six weeks before the period closed.
Who signs, and in what order?
Three electronic signatures, in order: you as author, then Dr. Omar Haddad as Qualified Person, then Dana Okafor, Head of Quality. Your signature is held until every finding is decided and the complaint count is reconciled.
13 screens from the working solution, on its sample data. Pick one to see it large.
The template, the trending rules, the clocks and the signers are settings. A new trending rule can be described in plain words.
| Setting | Default | Choose from |
|---|---|---|
| Review template | Combined US annual review + EU quality review (14 chapters) | Combined · EU product quality review only (13 chapters) · US annual product review only (9 chapters) |
| Review period | Rolling 12 months per product | Rolling 12 months per product · Calendar year for all products |
| Capability target (Cpk) | 1.33 | 1.00 · 1.33 · 1.67 |
| Stability out of trend | A pull outside the 95 % prediction interval of earlier pulls | On · Off |
| Draft built after the period closes | Next morning at 06:00 | Next morning at 06:00 · Within 3 days · On request only |
| Due date | 90 days after period end | 90 · 60 days after period end |
| Qualified Person — EU-released products | Dr. Omar Haddad | Dr. Omar Haddad · Lena Ortiz |
| Final approval | Dana Okafor — Head of Quality | Dana Okafor — Head of Quality · Dr. Maya Chen |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample product data · “target” = the design goal, measured in the live solution · “estimated” = our estimate · Sources: 21 CFR 211.180(e) · EU GMP Chapter 1 (product quality review) · ICH Q10. People, companies and products named on this page are fictional — characters and sample data in the working solution.
A yearly review of each product: batches made and their disposition, results and trends, deviations, OOS, changes, variations, stability, complaints, returns and recalls, last year’s actions, qualification status and agreements. The US calls it the annual product review (21 CFR 211.180(e)); the EU calls it the product quality review (EU GMP Chapter 1). The default template covers both in one 14-chapter report.
When a product’s review period closes, four collectors pull the period’s records from ERP, MES, LIMS, the QMS and the complaint, regulatory, engineering and contract systems. The trend analyst computes capability, run rules and stability regressions, the reconciliation checker counts every number in two systems, and the writer fills the template with charts and a citation behind every number.
Every number links to the records it was counted from, and is counted in at least two systems. If two systems disagree — 58 complaints in one, 57 in the other — the checker lists the record that explains it, and sign-off is blocked until a person resolves it.
No. Every connection is read-only and nothing is written back to the source systems. A change to the report after signature needs a new version.
Yes. The trend analyst raises each adverse trend with its evidence and a recommended action, but only the product quality lead decides: raise a CAPA, watch, or no action with a reason. The conclusion is written from those decisions.
Three electronic signatures, in order: the author, the Qualified Person for EU-released products, and the Head of Quality for final approval. Each is recorded with its meaning, date and time, and the signed review is locked.
Yes. The template, the review period, the capability target, run rules, the stability out-of-trend rule, when the draft is built, the due date and the signers are all settings, and a new trending rule can be described in plain words.
The Agentic Solution Engine builds and deploys it from your requirements — your review SOP, your template, your trending rules and a sample of a product’s records — and it goes live once every quality gate has passed. We will walk you through it on one of your own products first.
Any batch-record question answered in seconds, cited to the exact page and step.
1-min filmValidation DocumentsValidation protocols and reports drafted in minutes, every value traced to its source page.
1-min filmAsk the PlantBatch analysis that took three months now takes hours, with the query behind every answer.
1-min filmWe’ll build a product quality review for one of your products, from a sample of its own records.