Material receiving in five minutes, with the certificate attached to the lot

The problem

The foundation of all traceability depends on typing under pressure.

Incoming material traceability is the foundation of all product traceability. And today it rests on manual typing done under time pressure, with the carrier waiting. Both exits are bad. Either the truck waits forty-five minutes charging detention, or it gets typed in a hurry and the lots end up badly traced. That resurfaces months later, in the 8D where you have to prove which melt the failed part came from. And there is a second, quieter problem: the material certificate ends up in a filing cabinet or a network folder, unlinked from the production batch that consumed it. When the customer asks, days of documentary archaeology begin.

• The truck arrives with polyamide resin, fluoropolymer pellets, brass bar, o-rings or reinforcement braid reels: a twelve-to-twenty-line delivery note, a material certificate per lot and labeled pallets.
• Both ways out are bad. Either the truck waits forty-five minutes racking up demurrage, or it gets typed in a hurry and the lots end up badly traced.
• That reappears months later, in the 8D where you have to prove which heat the failed part came from.
• And the material certificate ends up unlinked in a filing cabinet, when it is exactly the document that closes the genealogy.

How it fits the IRIS system

Connect in photo mode — delivery note, certificate and labels, no supplier changes.

Connect in photo-to-the-analogue mode, writing into the ERP after human validation. No supplier has to change their document. Step 1 — Photo. Of the packing list, the material certificates and the pallet labels. Step 2 — Recognition. Vision AI identifies the layout of each packing list and certificate and extracts the tabular structure. For labels it reads GS1 codes, QR, barcodes or plain text. Step 3 — Linking. The certificate is filed against the material lot, so it is automatically associated with every part that consumes it: melt-to-part genealogy exists from minute one. Step 4 — Validation and write-back. One tap on an industrial tablet and everything enters the ERP through the interface. The receiving supervisor never opens the ERP.

The material certificate stops being a filed sheet of paper and becomes linked to the lot from minute one. That is what turns the heat-to-part genealogy into a query instead of a hand reconstruction.

See the full IRIS architecture →

Before and after

Today's receiving versus captured receiving

Aspect	Today	With iLEAN Connect
Truck at the dock	45 minutes	5 minutes
Material certificate	Unlinked, in a filing cabinet	Linked to the lot from minute one
Heat → part genealogy	Rebuilt during the 8D	Complete and queryable
Badly traced lots	Surface months later	Do not occur
Carrier demurrage	Recurring	Eliminated
Receiving and warehouse licenses	Two or three	One, the supervisor who validates

Impact estimate

Estimated impact — to validate with your own numbers.

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: the receiving and warehouse module typically requires two or three active licenses; with Connect only the supervisor validates: 50-70 % reduction, recurring.
• From 45 minutes of truck at the dock to 5, truck after truck.
• Carrier demurrage eliminated.
• And a complete heat-to-part genealogy, which is what holds up an 8D when the customer asks.

CFO angle — direct ERP license saving. The receiving and warehouse module typically demands two or three active seats (receiving supervisor, backup, supervisor above); with Connect only the supervisor validates, with one tap, without entering the ERP. Typical 50-70% reduction in receiving/warehouse module seats, plus the elimination of the bulk-capture module cost and the dedicated scanners. Estimated payback of 3 to 8 months, plus the recurring saving and the carrier detention that stops being paid. *Estimate to validate*.

And the fair question from the production manager

«Does it recognize every supplier's delivery note?» — yes, and without asking any of them to change theirs: the model extracts the table structure with no need for a per-supplier template. Labels are read by standard symbology where present and by optical recognition where not. It is an anchored task, where the best models drop below 1.5% error [1], and the supervisor validates before anything enters.

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