AI control of Iberian ham salting — the exact point on every leg, not the batch average.
Getting Iberian ham salting right means cross-referencing incoming weight, the temperature and humidity curve of the salting room and actual time in salt, leg by leg. iLEAN brings the three realities together, proposes the optimal removal date to the master salter, and leaves the PDO file signed. The person decides — the room never opens on its own.
One day per kilo is an average. The leg is not the average.
Salting is the silent phase of Iberian ham. It sounds like "put the leg in salt, wait the right number of days, take it out." And underneath, the master salter knows it is not like that: the real weight of the leg is not the one the cutting-room scale recorded eight hours ago, salt from a new batch is absorbed differently, the room behaves differently every week depending on how often the door is opened, and two legs from the same hook end up with different cures two years later.
The data exists, but it is broken across three places:
- Incoming weight — the cutting-room scale, the ERP or the room supervisor's spreadsheet.
- The salting curve — temperature/humidity probes if the room has them, a manual sheet if it does not.
- Actual time in salt — the hook, the marking, the calendar and the memory of whoever comes in on Saturday to empty the first batch.
The day it all gets cross-referenced, it is because someone ran three queries, remembered what happened last winter, and signed. The system works because there are people with a trade — and because nobody called at midnight to open a second room that was not in the plan.
iLEAN does not take the trade away from the master salter — it puts the three data points together in his pocket.
What is missing here is not judgment, it is information that spans three systems at the exact moment. iLEAN is the putty between the cutting room's ERP/spreadsheet, the room probes and the reality of the hook. It does not ask you to change the room, the salt or the way you work. It puts a nervous system on top of all that — one that sees, remembers and alerts.
Edge reads the hook. Connect captures weight, salt and curve wherever they live. The agent proposes a removal date per sub-batch. The person signs — and the line breathes at the salting room's pace, not the other way round.
The three iLEAN pieces applied to Iberian ham salting:
- Edge — a terminal with machine vision (CNN) at the room's entrance and exit. It reads the code/marking on the hook, date-time-leg, without the operator writing anything on paper. It works without a network: if the room is cut off, it keeps recording entries and exits on its own power.
- Connect — captures the cutting-room weight whether it comes from a modern ERP, from the scale with its spreadsheet on an old PC, or from the room supervisor's notebook. And it reads the temperature/humidity curve from the probes, or a photo of the analog chart recorder if there are no probes. Graduated capture adapts to the room — not the other way round.
- Agent — cross-references incoming weight, salt batch, salting-room curve and time in salt. It proposes the optimal removal date per sub-batch, anticipates when a sub-batch will come out under-salted if the room is running warm, and closes the per-leg file for the PDO. It never opens the room on its own — it alerts the salter through whichever channel he uses, and the person signs.
Salting by calendar vs. salting by sub-batch with iLEAN
| Aspect | Classic salting | With iLEAN Edge + Connect + Agent |
|---|---|---|
| Removal date | Calendar per batch, the "one day per kilo" average | Proposed per sub-batch, adjusted to real weight + curve |
| Room curve | Probes if there are any, notebook if not | Captured continuously, anomalies pushed to the salter's phone |
| Leg-by-leg traceability | Per batch, rebuilt in a spreadsheet | File per hook, automatic |
| Over-salting / under-salting | Shows up two years later, in the drying house | Anticipated in the room, still correctable |
| The master's knowledge | In his head, leaves with him | Captured as a pattern in the system, stays put |
| PDO audit | Gathering spreadsheets the night before | Dossier ready per leg, with weight, curve and signature |
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 drying house. We put it forward so the committee has an order of magnitude; we refine it during the on-site diagnostic.
- Mid-sized drying house, 2-3 salting rooms, 30,000-50,000 legs/year, regulated PDO.
- Pilot in one room: Edge terminal at entrance/exit + a complementary temperature/humidity probe if the room falls short + integration with the cutting-room scale and the salt batch sheet. First value expected within a few weeks.
- Indicative payback between 4 and 9 months, depending on the historical percentage of legs that fall outside the optimal salt range and the average cost of rework or commercial downgrading.
- The hard lever is twofold: a reduction of out-of-range legs ≥ 30% and an automatic PDO file (quality hours recovered at every audit).
And the master salter's reasonable doubt
"What if the AI proposes a date and the leg is not ready?" — the AI does not decide here, it proposes. And it proposes on top of data the master does not have in full today: the cross-reference of real weight + real curve + real time, on his phone, without pulling up three screens. The reliability of AI in anchored tasks (reading one data point and cross-referencing it with another) brought error below 1.5% in the best models[1]. And even so, what is critical is never decided alone: the room opens when the salter signs. iLEAN's three safety rings exist for exactly this.
[1] OpenAI paper "Why Language Models Hallucinate", 2025 — on the reliability of AI in anchored tasks.
What people ask about AI control of Iberian ham salting
How many days of salting does an Iberian ham need, and why does it vary so much?
The classic rule is roughly one day in salt per kilo of leg, but the real figure depends on the exact weight, on the degree of fat marbling, on the temperature and humidity of the salting room and on the salt batch. That is why two legs from the same hook can need different times. iLEAN does not replace the master salter — it gives him the real weight, temperature and humidity of the salting room hour by hour and leg by leg, so the batch cut-off matches reality instead of an average.
How does iLEAN avoid over-salting or under-salting?
By cross-referencing what today lives on islands: the incoming weight (the scale in the cutting room), the temperature and humidity curve of the salting room (sensors or a manual sheet), and the photo of the hook marking that iLEAN Edge reads on every batch. The agent alerts the master salter when a sub-batch is close to its optimal point and proposes a removal date. If the salting-room curve drifts, it recalculates. The person decides and signs — the AI never opens the room on its own.
What is the risk if the salting ends up out of range?
Two risks: over-salting (the leg loses commercial value — aggressive flavor, higher waste in post-salting, retailer rejection) and under-salting (microbiological risk over a long cure, possible rework or destruction of the batch). Both cost money and reputation on a product the end customer buys for its brand and its origin traceability. Salting is the point where a small error becomes a big one two years later in the drying house.
Does iLEAN integrate with protected designation of origin traceability?
Yes. The agent closes the file for each leg with incoming weight, salt batch, salting-room curve, actual days in salt and outgoing weight — everything the certification body (PDO Jabugo, Guijuelo, Los Pedroches, Dehesa de Extremadura) asks for at audit. It builds itself, with no scramble to gather spreadsheets the night before the inspection. It also fits IFS/BRC if you export.
How much does an iLEAN salting-control pilot cost?
The order of magnitude of a pilot in an Iberian ham salting room sits in the same range as any food plant pilot: Edge terminals to read the hook, temperature/humidity sensors if the room does not have them built in, and integration with the ERP or spreadsheet where the incoming weight lives. Indicative payback of several months, depending on volume and on the historical percentage of legs that came out of range. We ask for your salting room's data and send you the estimated ROI in 48h.
Related: curing chamber control · traceability in curing · carcass identification
More cases in food
- Meat cutting traceability — holding the carcass → cut link is the foundation of the whole business.iLEAN Vision recognizes every cut as it leaves the cutting room and keeps its link to the original…
- Sobrasada IGP Mallorca — you do not touch the curing curve, you watch it better.How iLEAN Edge governs the curing curve and the traditional stuffing of Sobrasada IGP Mallorca with AI…
- The section lead cannot be in five places at once — their AI earpiece can.The section lead talks to iLEAN without breaking stride along the breadstick line. It listens, records…
- Planning the frozen food plant around the center that eats the most energy: the freezing tunnel.The freezing tunnel is one of the biggest energy consumers in the plant. iLEAN's Planning Agent keeps it…
- A human verification tablet on the line — the person signs, in 2 taps, before the data enters the system.iLEAN Connect puts a tablet on the breadstick line so the operator confirms in seconds what the AI has…
- Nothing enters the system without a human having seen it — the tablet at the line asks for a two-tap signature before the data crosses into the ERP.An industrial tablet at the line. iLEAN shows a visual summary, the operator signs with two taps and…
Tell us your case and in 48h we'll send you the estimated ROI of this AI project for your drying house.
We work on your salting room's real data, not ours. Diagnostic with no commitment.
Request estimated ROI in 48h See food industry