"The line must produce 100,000 vials per day." It appears in the project brief, it is repeated in the URS, and it is quoted to three suppliers, who each interpret it differently. One quotes 100,000 divided by 24 hours. One quotes for a single shift. One quotes nominal mechanical speed with no allowance for anything at all. The resulting proposals differ by a factor of three, and the comparison is meaningless.
A daily output is a business target. Machine speed is an engineering parameter. Converting one into the other requires six numbers, and every one of them is a decision somebody must own.
The six numbers
- Operating pattern — shifts per day, hours per shift, days per week.
- Planned downtime within the shift — breaks, line clearance, environmental monitoring, in-process checks.
- Changeovers — how many per operating period, and how long each takes including line clearance.
- Overall equipment effectiveness for the aseptic line, or at minimum an availability and performance allowance based on comparable installed equipment.
- Reject and sampling rate — vials removed for IPC, weight checks, media fill exclusions, and the machine's own reject rate at speed.
- Contingency — the margin the site wants for demand growth or campaign compression.
Worked example
Take the 100,000 vials/day target on a two-shift pattern.
| Step | Basis | Result |
|---|---|---|
| Gross target | Business requirement | 100,000 vials/day |
| Sampling and rejects | 2.5% removed for IPC and weight check | 102,564 vials/day must be filled |
| Operating pattern | 2 shifts × 8 h | 16 h/day gross |
| In-shift planned stops | 45 min/shift (breaks, EM, clearance) | 14.5 h/day |
| Changeover | 1 per day at 90 min | 13.0 h/day available |
| Line efficiency | 80% availability × performance at speed | 10.4 h/day effective |
| Required nominal speed | 102,564 ÷ 10.4 h ÷ 60 | ≈ 164 vials/min |
| Contingency | +15% for demand growth | ≈ 189 vials/min nominal |
The figure that goes in the URS is 189 vials per minute nominal, not 100,000 per day — with the derivation retained so that any assumption can be challenged. Note how far it is from the naive answer: 100,000 over 24 hours is 69 vials per minute, less than 40% of what the line actually needs.
Specify the speed at a condition, not in isolation
A speed with no operating condition attached is still ambiguous. State it as a triplet:
REQ-CAP-001 The filling machine shall achieve a sustained output of not less than
189 vials per minute for the 10 mL format, at the fill tolerance stated
in REQ-FIL-004, over a continuous 4-hour run, with a machine-attributable
reject rate not exceeding 0.5%.
That requirement is testable at Performance Qualification with a stopwatch and a reject count. "100,000 vials per day" is not testable at all — a PQ run is not a day, and a day is not a batch.
Format matters more than people expect
Speed is not a single number across a product family. A 2 mL vial and a 50 mL vial do not run at the same rate, and the difference is rarely proportional. If the line will run several formats, the URS states a speed per format, with the worst case identified. The worst case drives the capacity calculation; the best case drives nothing at all, and quoting it flatters the line.
Where the number usually goes wrong
- Efficiency taken from the supplier. Suppliers quote mechanical speed. Effective output on an aseptic line with intervention rules, environmental monitoring and gowning breaks is a different number, and it belongs to the site, not the supplier.
- Changeover ignored. A line running six products in a week loses more time to changeover than to any mechanical fault. If changeover time is a capacity driver, it is a requirement in its own right, with an acceptance criterion measured from last good vial to first good vial.
- Isolator or RABS transfer time forgotten. Decontamination cycle time is unavailable production time and belongs in the operating pattern, not in a footnote.
- Upstream and downstream not checked. A 189 vpm filler behind a 150 vpm lyophiliser loading system produces 150 vpm and a queue. Line capacity is set by the slowest constrained step.
What to write down
Three artefacts survive the project: the target, the derivation, and the specified speed. Keep the derivation as an appendix to the URS with each assumption attributed to a named owner. When the line under-delivers in year two — and someone will ask why — the argument is settled in ten minutes rather than becoming a six-week investigation.