AI nasal spray control — a crimp out of height is a dose out of specification.

The problem

Three plant realities that almost never get cross-checked in time.

On a nasal spray line — solution or suspension, multidose or single-dose — the guarantee that the right dose is delivered rests on three pieces of data that in most plants each live on their own island:

1. The dose filled into the bottle — what volume the filler is calling for and how it varies from bottle to bottle. Verified by IPC sampling, not continuously.
2. The valve crimping — height, symmetry, neck depth, head pressure. If the pressure curve drifts, the valve can end up partially sealed and the delivered dose falls out of range.
3. The plume in the actuation test — spray angle, droplet density, pattern symmetry. Validated under USP <601>/<1601>; in line it is usually checked by sampling, not at 100%.

The quality manager knows everything adds up for 99% of the batch. That 1% is what triggers the OOS investigation into delivered dose, the suspicion over the crimping of the entire run, and the hold on the whole batch. Caught late it is a deviation; caught in line, it is nothing.

How it fits the IRIS system

iLEAN does not add a fourth system — it seals the cracks between the three you already have.

The nasal spray problem is not a lack of information: it is information living on islands that, at the critical moment (crimping drift from head wear, a change of container batch, a suspension settling in the hopper), does not reach the decision-maker in time. iLEAN acts as the putty that fills those gaps, without asking you to change your filler-capper.

Edge sees the level, the crimp and the plume. Connect reads the crimper wherever it is — modern PLC, old panel, forgotten spreadsheet. The agent cross-checks against the recipe and the USP/GMP specification and, if something does not add up, holds the run. The person signs — never the other way round.

The three iLEAN pieces applied to nasal spray:

• Edge — a terminal with machine vision (CNN) over the packaging carousel and over the actuation bench. It verifies bottle level, crimp height/symmetry and plume pattern in milliseconds. It fires the actuator (selective rejector, stack light, stop) before cartoning. It works without a network. If the plant loses Internet, Edge keeps verifying.
• Connect — captures the crimper whether the data comes from the PLC over OPC-UA, from the old terminal by OCR, or from the packaging lead's setpoint spreadsheet. It also captures the head pressure curve, the product temperature and whatever arrives from outside (an email from the validation lead with a change to the crimping range, a WhatsApp from the supplier about a deviation in a batch of valves).
• Agent — cross-checks dose, crimping, plume, the MES recipe and the USP specification. If there is a deviation, it does not send an email at 10 p.m.: it holds the run and alerts the quality manager through whatever channel they use. The person validates and signs; the line does not restart on its own.

See the full IRIS architecture →

Before and after

Nasal spray controlled by manual IPC vs. cross-checked with iLEAN

Aspect	Manual control + IPC by sampling	With iLEAN Edge + Connect + Agent
Dose verification	Periodic sample on a balance	Every bottle verified in line, in milliseconds
Valve crimping	Gauge every N units, assume the rest	Height/symmetry verified bottle by bottle
Plume pattern	USP sampling at batch start and end	Periodic actuation read by CNN vision
Deviation detection	At IPC or, worse, at release	Before cartoning, automatic hold
Operation without a network	n/a	Edge keeps running on the panel's own power
File for the FDA/EMA auditor	Rebuild from paper, days	Automatic per-batch dossier, with images and traces

Impact estimate

Impact estimate for your plant — to be validated with your numbers.

The block below is an estimate to be validated with the specific data of your plant. We put it forward so the committee has an order of magnitude; we refine it during the diagnostic.

• Mid-sized pharma plant, one multidose nasal spray line (corticosteroid suspension or solution), 60-120 bottles per minute, multi-SKU with format changeovers.
• Edge pilot on one line (cameras over the carousel and the actuation bench + actuator + integration with the crimper and the MES). First value expected within a few weeks.
• Reduction of OOS deviations caused by crimping or dose ≥ 30% in the first pilot batch, depending on initial severity and crimper dispersion.
• Indicative payback between 4 and 9 months, depending on the frequency of documented incidents and the average cost of rework/destruction.
• The hard lever is a single OOS deviation avoided: investigation, rework, risk of a scrapped batch, the auditor's attention.

And the quality manager's reasonable doubt

“What if the AI gets it wrong and lets a badly crimped bottle through?” — hallucination is a problem of free generation, not of anchored tasks. In tasks where the AI merely recontextualizes a piece of data from one system to another (measuring crimp height and comparing it with the specification), the best models brought error below 1.5% ^[1]. And even so, what is critical is never decided alone: iLEAN holds the batch and the person signs. The three safety rings exist precisely for this.

^[1] OpenAI paper “Why Language Models Hallucinate”, 2025 — on the reliability of AI in anchored tasks.

← AI suppository manufacturing control · AI MDI inhaler control → · Syrup filling control →