Lot Genealogy: How Automatic Trace-Building Works

A genealogy you build during the recall is a reconstruction. A genealogy that builds itself during production is an answer. Here's the mechanics.

Written by George Munguia · Published Jul 19, 2026

Every trace question, recall scope, customer complaint, supplier problem, is a walk through a graph: nodes are lots, edges are 'went into.' Plants differ only in when the graph gets built. Some build it during the incident, from binders and batch sheets, under a deadline. Some build it during production, automatically, one link at a time. The second kind answers in seconds because the answer already exists.

The three link-writing moments

Receiving: inbound lot arrives, gets its node, supplier, material, quantity, COA attached (read by document AI, not typed).
Consumption / transformation: the load-bearing edge. When flour lot F-812 goes into batch B-2201, that event is captured, at the scale, the staging scan, or the station tablet, with quantity and time. Miss this moment and no software saves you; capture it and the genealogy is inevitable.
Pack and ship: batch to case to pallet to customer, closing the graph outward.

Walking the graph

One-back: start at a finished lot, walk edges upstream, every input lot, with quantities and records attached. One-up: start at a suspect input, walk downstream, every batch, lot, and customer it touched, which is recall scope, exact, not padded 'to be safe' because the data was fuzzy. Precision is money: plants over-recall when genealogy is uncertain.

Where the process context multiplies value

A genealogy that also carries process context, which machine, which hour, which operator, which cook profile, answers the question after the trace: not just where the bad lot went, but why it went bad. That is the bridge from compliance to operations: ML ties the defect pattern to the conditions, and the trace layer becomes a quality tool (the category mechanics live in our traceability software guide).

Retrofit reality

Plants assume genealogy requires serializing everything and a year of integration. The additive version digitizes the three link-writing moments in weeks, and backfills history by letting document AI read the paper archive. Serialization gets added only where unit-level identity pays, firearms, appliances, and anywhere warranty claims arrive by serial number (see firearms and windows and doors for what unit-level looks like).

Frequently Asked Questions

What is lot genealogy?

The parent-child map linking supplier lots to batches, batches to finished lots, and finished lots to shipments. With genealogy in place, any trace question is answered by walking the map in either direction, in seconds, with records attached.

How is genealogy built automatically?

By capturing the linking events as production runs: inbound lots at receiving, consumption at staging or the station (which input lots fed which batch), and pack/ship links outward. Each event writes an edge; the graph assembles itself.

What is the difference between one-up and one-back?

One-back walks upstream, everything that went into a lot. One-up walks downstream, everywhere an input went. Regulators, auditors, and customers ask both; precise genealogy answers both without padding the recall scope.

Do we need to serialize products for genealogy?

Only where unit-level identity pays, serialized goods, warranty-driven products, regulated items. Most food and process plants need lot-level genealogy, which requires capturing links at receiving, consumption, and pack, not serializing every unit.

The Operations Brief

The best of what we publish, once a week.

Practical guides on high-production manufacturing, straight from the floor. No spam, unsubscribe anytime.

You're in. The next brief lands Monday morning.

See Harmony on your line

Harmony connects your machines, systems, and paperwork into one real-time operational layer — no rip-and-replace. See what it looks like in a plant like yours: read the CLS case study.

Book a Demo →

← Back to Resources