Material harmonization
Raw materials from several ERPs enriched with properties extracted from safety data sheets and certificates of analysis, grouped into harmonized materials with AI suggestions, so a bill of materials can name an alternative supplier.
One working session with the people in the seat: which decisions, which systems, what a good week looks like.
The app you see above, running on your own exports instead of the simulated world. Every screen, every decision, your numbers.
Wired to your systems with your permissions, agents proposing, people approving, the ledger recording who did what.
Overview
The same chemical sits in two ERPs under three names, and its properties live in supplier PDFs. Harmonizing by hand takes a chemist per plant, so bills of materials keep naming one supplier.
The app extracts properties from the documents with a vision-language model, shows the evidence beside the record, and recommends which harmonized group a material belongs to with a similarity score the steward accepts or rejects. Once grouped, alternative materials appear in the BOMs that depend on a single supplier. New documents run through the same extraction logic on upload.
Key features
- AI entity extraction: material properties from specification documents with the page cited and the reasoning shown.
- Material harmonization application: raw materials, harmonized groups and similarity-scored recommendations, accepted or rejected with feedback.
- Increased production resilience: alternative materials surfaced in existing bills of materials with the transfer cost.
- Document upload: safety data sheets, certificates of analysis and other documents extracted on arrival.
Decisions it produces
- Accept or reject a group recommendation
- Merge or split a harmonized group
- Add an alternative material to a bill of materials
- Approve extracted properties for write-back to the source system
Data sources
| ERP material masters | raw material ids, CAS numbers, source plant |
|---|---|
| Safety data sheets and certificates | appearance, density, form, handling, storage, extracted per page |
| Bills of materials | which products depend on which material and supplier |
| Steward feedback | accepted and rejected recommendations that retrain the model |
Simulated data throughout. Book a call to see it on your own systems.
Related demos
Entity resolution
A configurable pipeline that cleans, pairs and scores records to find the same person, company or product across datasets, with a labelling app where stewards confirm, flag or reject each match and a config manager that defines every step.
Procurement negotiations assistant
A vendor 360 with exposure on a world map and concentration history, feeding a negotiation cockpit: data-backed arguments, a should-cost bill of materials down to the commodity, and vendor claims extracted as they arrive.
Hazardous material compliance
Regulated materials such as PFAS traced from raw material through the bill of materials, production and sales orders, with each reporting decision validated against the published regulation and compliance status by jurisdiction.