The truck no longer waits while the delivery note is typed

The problem

The heat number is typed in a hurry, and it is the one that matters.

If data is typed fast to avoid demurrage, heat numbers get traced badly — surfacing weeks later as a non-conformity impossible to trace back to the origin coil.

• The truck arrives with tinplate or tin-free steel (TFS) coils and the receiving lead has to type 12 to 20 delivery-note lines into the ERP while the carrier waits.
• Typing fast avoids demurrage but traces heat numbers badly; typing carefully keeps the truck at the dock for 45 minutes.
• A wrong heat number does not show up that day. It surfaces weeks later as a nonconformity that cannot be traced back to the origin coil.
• And without the origin coil, an adhesion or corrosion problem at the packer client has no way back to the steel supplier. The supplier claim then rests on a hunch instead of a heat number.

How it fits the IRIS system

Connect in photo-to-analog mode — delivery note, heat labels and one validation tap.

Connect photo-to-analog mode. Vision AI recognizes the delivery note's format without asking the supplier to change theirs, and reads GS1-128, QR, barcode or plain OCR on heat-number labels. Validation with one tap on an industrial tablet.

Traceability in metal packaging starts at the coil. If the heat number enters correctly at the dock, every run, sheet and pallet downstream inherits it; if it enters wrong, nothing later can fix it. That is why the dock, not the lab, is where coil traceability is won or lost.

See the full IRIS architecture →

Before and after

Coil receiving typed at the dock versus coil receiving from a photo

Aspect	Today	With iLEAN Connect
Truck at the dock	About 45 minutes	About 5 minutes
Heat number entry	Typed under time pressure	Read from GS1-128, QR, barcode or OCR
Hidden heat-number errors	Surface weeks later	Zero, checked before entry
Trace back to the origin coil	Broken when a digit is wrong	Complete, with photo and label attached
Steel mill delivery note format	Typed whatever the layout	Recognized without asking the supplier to change
Receiving module seats	Two or three in use	None for the lead, who validates on a tablet

Truck 45 min → 5 min. Hidden heat-number errors → zero. Broken traceability → complete via photo and GS1 back to the origin coil.

Impact estimate

Impact estimate — to be validated with your 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.

• Estimated payback 3-8 months.
• 💰 CFO angle: the receiving module typically needs 2-3 licenses; with Connect the lead validates without entering the ERP — a typical 50-70% reduction in receiving and warehouse licenses.
• From 45 minutes of truck at the dock to 5, and carrier demurrage that stops accruing.
• Heat-number errors hidden in the ERP drop to zero, so a nonconformity can always be traced to its origin coil. And a claim to the steel supplier can be argued with the heat number and the photo of its label.

CFO angle: the receiving module typically requires 2-3 licenses; with Connect, only the lead validates without entering the ERP. Typical reduction of 50 to 70% in receiving/warehouse module licenses. Estimated payback of 3 to 8 months. Estimate to validate.

And the fair question from the production manager

“Every steel mill has its own delivery note — does it cope?” — yes, without asking any of them to change: vision recognizes the table structure of each format, and heat-number labels are read by GS1-128, QR or barcode where present and by OCR where not. It is an anchored task where the best models drop below 1.5% error [1], and the lead validates with one tap before anything enters. No supplier has to change a document, a label or a habit.

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