The JIT Delivery Note Enters ERP Alone

The problem

The JIT window is measured in minutes; typing is measured in mistakes.

The JIT truck has a minute-tight delivery window; if the receiving supervisor types slowly into the ERP the window is missed, and if they type fast, mistraced batches get entered. Those batch errors surface weeks later in a non-conformance that nobody can trace back to the exact container.

• A JIT truck gets a minute-tight unloading window at the dock, and the next truck is already waiting behind it.
• If the receiving supervisor types slowly into the ERP, the window is missed. If they type fast, the wrong batch gets entered against the wrong container.
• Every supplier prints its own delivery note format, and container labels mix GS1-128, QR codes and plain text.
• A batch error made at the dock surfaces weeks later in a non-conformance, and nobody can trace it back to the exact container or the VINs it went into.

How it fits the IRIS system

Connect from photo to ERP — the supplier does not change a thing.

Connect photo-to-analog mode: vision AI recognizes the delivery-note format without requiring supplier changes, and reads GS1-128/QR/barcode/plain OCR on container labels. One validation tap on the dock tablet, automatic push to ERP.

Vision recognizes each supplier's delivery note as it is, and reads whatever is printed on the container label. The supervisor checks one screen and taps once. The window is met, the batch is right, and the container can be followed all the way to the VIN. On a dock that receives a JIT truck every few minutes, that is the difference between a receiving team that keeps up and one that builds a backlog by mid-shift.

See the full IRIS architecture →

Before and after

JIT receiving typed by hand versus received from a photo

Aspect	Today	With iLEAN Connect
Dock time per JIT truck	Blocked by manual typing	Minutes
Supplier delivery note formats	Each one learned by a person	Recognized as they are
Container label reading	Typed from the label	GS1-128, QR, barcode or OCR
Batch entered against the wrong container	Found weeks later	Caught at the dock
Container-to-VIN traceability	Broken at receiving	Complete from the photo
Receiving and warehouse licenses	One per person who types	Typically 50-70% fewer

Unloading blocked by manual typing → minutes. Hidden batch errors → zero. Broken container-to-VIN traceability → complete via photo and GS1.

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, before counting the license saving.
• A typical 50-70% reduction in receiving and warehouse-module licenses.
• Unloading time back to minutes, and the JIT window stops depending on how fast someone types or how many trucks are queuing in the yard.
• And batch errors that used to surface in a non-conformance weeks later are stopped at the dock, with every container traceable to the units it ended up in.

Estimated payback 3-8 months plus a typical 50-70% reduction in receiving/warehouse-module licenses. *Figure to be validated*. (Payback 3-8 months · -50/-70% receiving-module licenses)

And the fair question from the production manager

“What if it reads the wrong batch number off a scuffed label?” — reading a delivery note and a container label whose formats are known is an anchored task, where the best models drop below 1.5% error [1]. When the label carries a GS1-128 code, the barcode and the printed text are cross-checked against each other and against the supplier's advance shipping notice, and any mismatch stops at the dock tablet. Nothing reaches the ERP until the supervisor confirms it.

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