Photo the packing slip, reels to the ERP

The problem

The foundation of all traceability depends on typing under pressure.

Instead of entering the ERP to type reel by reel -lot, quantity, date, moisture sensitivity- the supervisor spends much of the shift on the unload.

• A single delivery brings dozens of reels, trays and tubes: part number, lot, quantity, date code and moisture sensitivity level on each label, plus a packing slip whose format changes with every distributor.
• Instead of that taking minutes, the receiving supervisor spends much of the shift typing reel by reel into the ERP, with a dock that keeps filling behind them.
• Under that pressure, errors are not the exception. A wrong lot or a missed MSL rating is invisible at goods-in and expensive at reflow, where an improperly baked part delaminates and takes the board with it.
• And when an OEM customer asks months later which reel lot went onto which board, the answer depends on whatever was typed that afternoon. Traceability does not fail at the end of the process; it fails at goods-in, and everything downstream inherits that gap.

How it fits the IRIS system

Connect in photo-of-the-analog mode — the packing slip and the labels, no distributor changes.

Connect photo-of-the-analog mode. Vision AI recognizes the supplier's packing-slip format and extracts tabular structure; for reel labels it reads barcodes/2D codes or plain text via OCR (lot, MSL, date). One-tap tablet validation, then automatic push to the ERP.

Barcode where there is a barcode, optical reading where there is not, and one tap to validate. The reel stops being a line somebody typed and becomes the first link of a genealogy that reaches the finished board — which is the link the flagship inventory case rests on.

See the full IRIS architecture →

Before and after

Today's goods-in versus captured goods-in

Aspect	Today	With iLEAN Connect
Booking in a delivery	Much of a shift, reel by reel	Minutes
Packing slip format	Different per distributor	Structure extracted with no template
Lot and date code per reel	Typed under pressure	Read from barcode or by OCR
MSL rating	Frequently missed	Captured and carried into the ERP
Reel → board traceability	Broken at the first gap	Complete from goods-in
Receiving-module licenses	Two or three	One, the supervisor who validates

Extended blocking of the receiving supervisor typing reel by reel → minutes. Hidden lot/MSL errors → zero. Broken traceability → complete via photo and barcode.

Impact estimate

Impact estimate — to be validated against your actual 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: the receiving module typically needs two or three active seats; with Connect only the supervisor validates: −50 % to −70 %, recurring.
• Estimated payback 3-8 months, counting the license line and the receiving supervisor's shift back.
• From a blocked supervisor typing reel by reel to minutes per delivery, delivery after delivery, with the dock clearing at the pace the trucks arrive instead of the pace someone can type.
• Hidden lot and MSL errors down to zero, and traceability complete via photo and barcode instead of broken at the first gap — which is what makes the reel-to-board genealogy possible at all downstream.

−50 to −70% receiving-module licenses · payback 3-8 months. Estimate to be validated.

And the fair question from the production manager

“Does it recognize every distributor's packing slip?” — yes, and without asking any of them to change it: the table structure is extracted with no per-supplier template, and reel labels are read by standard symbology where it exists and by optical recognition where it does not. It is an anchored task where the best models drop below 1.5% error [1], and the supervisor validates before anything enters the ERP.

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