The delivery note that self-loads

The problem

The queue at the scale is made of typing.

The truck waits in queue while the operator types every line of the delivery note and waste form into the system. Going fast means wrong waste codes or origins slip in, surfacing weeks later in an inspection nobody can trace precisely.

• Every truck brings a carrier's delivery note and a waste identification form, and the operator types every line — waste code, origin, weight, source site — while the next trucks wait.
• Each carrier uses its own layout, so there is no shortcut: it is read line by line and retyped by hand, truck after truck, with drivers waiting at the barrier and the next load already in line.
• To keep the queue moving the operator goes fast, and a wrong waste code or a wrong source site slips in without anyone noticing, because nobody has time to read back what was typed.
• The error surfaces weeks later, in an environmental inspection, where nobody can trace exactly which truck it came from. By then the load is aggregate, and the paper is in a box.

How it fits the IRIS system

Connect in analog-photo mode — no carrier is asked to change its paperwork.

Connect's analog-photo mode: computer vision recognizes the delivery note and waste-form layout (no need for the carrier to change theirs), extracting tabular structure (waste code, origin, weight, source site). Validation with one tap on the scale's industrial tablet.

Vision recognizes each carrier's layout as it is and pulls out the table: waste code, origin, weight and source site. The operator confirms with one tap on the scale's industrial tablet, and the photo stays attached to the case file as proof. If an inspector asks what arrived, the answer is the document itself, not a line retyped from it.

See the full IRIS architecture →

Before and after

The delivery note typed versus the delivery note photographed

Aspect	Today	With iLEAN Connect
Time per truck at the booth	Minutes of typing	Seconds
Carrier note layouts	Retyped one by one	Read as they come
Waste code or origin error	Hidden until inspection	Caught at the tap
Proof of what arrived	The typed line	Photo plus record, by case file
Intake-module licenses	Several active seats	Only the validating supervisor
Trucks waiting	A queue at the barrier	Moving

Minutes of queue per truck typing → seconds. Hidden waste-code or origin errors → full traceability by photo and case file.

Impact estimate

Impact estimate — to be validated with your queue and license figures.

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, depending on daily trucks and on how many seats the intake module carries today.
• CFO angle: typically a 50-70% cut in intake and scale-module ERP licenses, because only the supervisor validates with a tap.
• Minutes of queue per truck become seconds, which carriers notice before anyone else does.
• And hidden waste-code or origin errors give way to full traceability by photo and case file, which is what the inspection actually asks for.

CFO angle — direct ERP license savings: the intake/scale module typically requires several active licenses; with Connect only the supervisor validates with a tap. Typically a 50-70% cut in intake/scale-module licenses. Estimated payback 3-8 months. *Estimate to validate*.

And the fair question from the production manager

“Every carrier's note is different — will it cope?” — locating and reading the table on a commercial document is an anchored task, where the best models drop below 1.5% error [1], and a new layout only has to be seen a few times. The waste code is checked against the official waste list and your authorization, and the operator's tap is still the gate before anything reaches the ERP. The photo stays with the record, so any doubt can be checked against the original.

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