Every few years an MES vendor’s sales deck gets a new slide, and this year the slide is crowded. FSMA 204’s recordkeeping requirements are now in effect for covered foods, EU due-diligence rules are asking manufacturers to prove their raw materials didn’t come from deforested land, and carbon border adjustment mechanisms are creeping toward requiring embedded-emissions data on products that never used to need any. Three separate regulatory tracks, three separate agencies or trading blocs, and one shop floor that’s supposed to somehow produce all the underlying data on demand.
The instinct in a lot of plants right now is to treat this as an MES problem and start bolting new fields onto the genealogy tables that already track lot consumption and work-order routing. That instinct is understandable and, in my view, wrong more often than it’s right. Not every compliance data point belongs in MES. Some of it structurally can’t live there without breaking the thing MES is actually good at, which is real-time, transactional tracking of what happened on the line, in what order, with what materials, under what conditions.
Three mandates, three very different data shapes
It helps to separate what these regulations are actually asking for, because they are not asking for the same kind of data, even though they all get lumped together as “traceability.”
FSMA 204 is fundamentally an event-based traceability requirement. It wants Critical Tracking Events — receiving, transformation, shipping — tied to Traceability Lot Codes, with the ability to produce a traceback within a defined window during an outbreak investigation. That is squarely genealogy territory. It’s the same shape of data MES has been capturing for lot genealogy and recall support for years, just with a more standardized vocabulary (KDEs, CTEs) and a harder compliance deadline behind it.
EU deforestation-style due-diligence rules ask something structurally different: not “what happened to this lot on my line,” but “where did the raw material originate, geographically, and can the supplier attest to that.” This is closer to a static attribute attached to an incoming material lot than an event. It doesn’t change every time the lot moves through a process step. It’s set once, upstream, usually by a supplier attestation or geolocation data tied to a purchase order, and it just needs to travel with the material.
Carbon and energy reporting — the CBAM-adjacent obligations now reaching discrete manufacturers rather than just primary metals and cement producers — is different again. It wants a calculated attribute: carbon or energy intensity per unit of output, usually derived from utility meter data, process parameters, and sometimes supplier-declared embedded emissions on inputs. That’s an analytics output, not a transactional record. It’s built from MES and historian data, but it isn’t itself a genealogy fact.
The mistake: treating all three as “more genealogy fields”
Genealogy tables in MES are built around a specific job: tracking parent-child material consumption relationships at production speed, transactionally, so that a recall or a yield investigation can walk the tree quickly. That data model is optimized for write-heavy, real-time transactions tied to work orders and equipment states.
Supplier attestations, geolocation certificates, and carbon coefficients are a different animal. They’re reference data — slow-changing, sourced from procurement and supplier onboarding processes, often revised on a supplier-relationship cadence rather than a production cadence. Cramming them into the same tables that track real-time consumption creates real problems:
- Schema bloat that slows down the transactions MES exists to handle. Genealogy queries during an active recall need to be fast. Every extra join to a supplier-attestation table or a carbon-factor lookup adds latency exactly when you can least afford it.
- Data staleness risk. A supplier’s deforestation attestation might be revised or challenged months after the material was consumed. If that fact lives natively in MES genealogy, you now have a production record that needs retroactive editing — which is exactly the kind of mutability that auditors and quality systems are built to distrust.
- Ownership confusion. Who updates a carbon coefficient when the utility changes its grid mix reporting, or when a supplier revises an emissions factor? If that number sits inside MES, it becomes the plant floor’s job to maintain compliance reference data it didn’t generate and can’t independently verify. That’s a governance failure waiting to happen.
A working framework: three questions per data point
Before adding any new field to MES because of these regulations, run it through three questions.
Is it an event or an attribute? If it’s something that happens at a specific point in the production sequence — a transformation, a shipment, a receipt — it’s an event, and MES genealogy is the right home, ideally mapped to the standard vocabulary regulators are converging on rather than a custom internal schema. If it’s a fact that’s true about a material regardless of when or how it’s processed, it’s an attribute, and it likely belongs upstream, in a supplier master or a compliance layer that MES references but doesn’t own.
Does it change at production speed, or supplier/reporting-period speed? MES should carry data that changes as fast as the line runs. Data that changes on a quarterly reporting cycle or a supplier-contract cycle belongs in ERP, a supplier relationship management module, or a dedicated compliance/sustainability platform — with MES exposing the linkage (which lot, which supplier, which PO) rather than the value itself.
Who is legally accountable for the number if it’s wrong? If a carbon intensity figure gets challenged in a CBAM declaration, the accountable party is almost certainly a compliance or sustainability team working from ERP and utility data, not the plant floor. MES should be a data source feeding that calculation — energy consumption per work order, run time, scrap rates — but the calculation and the regulatory submission are not MES’s job. Keep the boundary where the accountability sits.
What actually belongs in MES right now
Given all that, here’s where I’d draw the line for 2026 planning:
- Lot-to-material-source linkage for FSMA-covered items: yes, in MES, mapped to CTE/KDE structures, because this is genuinely transactional genealogy and MES is the only system with visibility into transformation events at the right granularity.
- Supplier attestations on origin (deforestation-style rules): reference only. MES should carry a pointer — supplier ID, lot ID, attestation reference — not the attestation document or its underlying geolocation data. Let procurement systems or a supplier-document repository own the actual content and its lifecycle.
- Carbon/energy-per-unit figures: mostly not in MES as a stored value. MES and the plant historian should supply the raw inputs — energy consumption, cycle time, throughput — to whatever system does the emissions calculation. Storing a final carbon-per-unit number inside genealogy tables invites exactly the staleness and ownership problems described above.
The bolt-on layer isn’t a cop-out
There’s a reflexive suspicion of “bolt-on compliance layers” among MES purists, as if anything outside the core historian and genealogy model is a shortcut. In this case it’s the opposite. A properly scoped compliance layer — whether it’s a module from your MES vendor, a standalone traceability platform, or a well-governed extension of ERP — is what keeps MES doing its actual job well. The plants that will struggle through 2026 audits aren’t the ones with a separate system for supplier attestations. They’re the ones whose genealogy tables have quietly turned into a junk drawer of regulatory fields nobody fully owns, slowing down the one system that was supposed to make traceability fast.
The regulations are real, the deadlines are real, and the data has to exist somewhere auditable. The mistake would be assuming “somewhere” always means MES.
This article was written with the assistance of artificial intelligence. While we aim for accuracy, the information may be incomplete, out of date, or incorrect, and should be independently verified before you rely on it for any decision. It is provided for general information only and does not constitute professional advice.
