Photo of the scale ticket that reaches the ERP

The problem

In peak season the queue at the scale is really a queue at the keyboard.

In peak season, trucks pile up at the scale while the receiving supervisor types orchard by orchard into the ERP. Rushing to avoid demurrage means lots get entered wrong, which later shows up as a traceability discrepancy hard to resolve in time for export.

• At harvest peak, trucks line up at the scale while the receiving supervisor types orchard by orchard, crate by crate, into the ERP, with the scale ticket in one hand and the crate labels in the other.
• Every minute of queue is fruit sitting warm in a truck and a carrier charging demurrage, while the next orchard's truck waits behind it with the same problem.
• Typing in a hurry to clear the queue means lots entered with the wrong orchard or harvest date, and nobody has time to double-check them.
• The error surfaces later as a traceability discrepancy on an export shipment, exactly when there is no time left to resolve it, with the container booked and the vessel waiting.

How it fits the IRIS system

Connect in photo-of-paper mode — the ticket and the crate labels, one tap to the ERP.

Connect photo-of-paper mode: one photo of the scale ticket and another of the crate labels (barcode, QR or plain text) digitizes both documents in seconds. iLEAN extracts orchard, weight, harvest date and grower; with one validation tap, everything is inserted into the ERP.

A truck of avocado waiting in the sun is not just a demurrage invoice: it is field heat the cold room will have to take out later. Freeing the scale in minutes protects the fruit as much as the schedule. In peak season, the scale is often the real bottleneck of the whole plant, not the sizer.

See the full IRIS architecture →

Before and after

Receiving at the keyboard versus receiving posted from a photo

Aspect	Today	With iLEAN Connect
Truck at the scale	Up to an hour	Minutes
Crate labels	Read and retyped	Barcode, QR or text read from the photo
Orchard and harvest date per lot	Typed in a hurry	Extracted and validated
Demurrage in peak season	Recurring	Eliminated
Export traceability discrepancy	Found at shipping	Does not arise
ERP users at receiving	Several active users	The supervisor who validates

One-hour truck unload → minutes. Hidden orchard/lot errors → zero. Broken orchard-to-lot traceability → complete via photo.

Impact estimate

Impact estimate — to be validated with your numbers and your ERP invoice.

The block below is an estimate to be validated against your plant's actual data. We put it forward so the committee has an order of magnitude; we refine it during the assessment.

• 💰 CFO angle: only the receiving supervisor validates, without logging into the ERP — a typical 50-70% reduction in receiving and warehouse module licenses.
• Estimated payback 3-8 months, including the truck demurrage that disappears.
• Truck unloading goes from an hour to minutes, so the fruit reaches the cold room sooner and carries less field heat into storage.
• And orchard-to-lot errors hidden at receiving go to zero, so traceability arrives complete at export.

CFO angle: only the receiving supervisor validates without logging into the ERP. Typical 50-70% reduction in receiving/warehouse module licenses, plus eliminated truck demurrage. Estimated payback 3-8 months. *Estimate to be validated.*

And the fair question from the production manager

“Crate labels are torn, wet and printed by dozens of growers — will it read them?” — it reads barcodes and QR codes where they exist and printed text where they do not, with no template per grower and no new label printers at the orchards. It is an anchored task, where the best models drop below 1.5% error [1], and any crate the model is unsure of is flagged in the summary before the supervisor posts the receipt. Nothing is posted with a crate left unread.

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