product-guide
Do We Need to Rip Out Our MES or ERP to Publish Battery Passports?
No. Your MES and ERP already generate most of the genealogy and master data the passport needs — the missing piece is a publishing layer, not a systems replacement.
The short answer
No. If your MES and ERP are doing their jobs, they already produce most of what the passport needs — genealogy from the shop floor, master data and lot codes from the enterprise side. The gap is at the very end of the chain: none of these systems can publish a per-unit, QR-resolvable, regulator-facing record. You're missing a layer, not a stack.
What each system already contributes
- ERP — the official transactional record: legal entity, lot codes, purchase orders, country of origin, recycled content declarations from suppliers.
- MES — the per-unit truth: bidirectional genealogy tracking cells from incoming material to finished pack, tied to a serial. Platforms in this space (GE Vernova's Proficy, Honeywell Battery MXP, Critical Manufacturing, AVEVA, and integrators like RedViking) exist precisely to make per-unit genealogy real. When MES and ERP are integrated, shop-floor build records and enterprise data reconcile to a single per-unit identity — which is the passport's spine.
- PLM/PDM — the engineering record: BOM, BMS firmware versions, cell supplier specs, version-controlled.
- Test stations — rated capacity, end-of-line results, the performance fields.
None of this needs replacing. All of it was built inward-facing — serving engineering, QC, and finance — and the passport is the first hard requirement to face this data outward.
What the publishing layer actually adds
The part your existing systems genuinely can't do:
- A public, per-unit record at a stable URL that a scanned QR resolves to — for the exact unit, not the SKU.
- Access tiers — public fields for anyone, restricted fields for authorities and recyclers.
- A compliance data shape — the regulation's field list, validated, versioned when corrected.
- Lifecycle appends — state-of-health and repair/repurpose/recycle events added after the unit ships.
The integration reality check
The connection between your systems and a publishing layer doesn't need to be exotic. In rough order of sophistication: manual CSV/spreadsheet export of batch and serial records (fine at low volume), scheduled exports from ERP/MES, or an ingest API receiving records as your line produces them. A useful property to insist on: file uploads and machine ingestion should land in one provenance stream, so a record's origin is traceable whether it arrived from a spreadsheet or an API — your auditors will eventually ask.
Start with the export you can produce this week, not the integration architecture you'd design given a year.
What this means for your team
Ask your MES/ERP vendors one question: "what can you export per finished unit?" — then treat the passport as a destination for that export. The manufacturers in the best 2027 position are not the ones replacing systems; they're the ones who realized their existing systems were already 80% of the answer.