Tools
Two companion tools — built to use OCO in practice
The ontology itself is only half the solution. Data comes from labs and from the literature — we cover both with custom tools that use OCO as schema and operate against modularly loaded sub-bundles.
OCO-Workbench
GUI tool for data import, process modelling, and ABox population. Lab personnel record samples, process chains, and measurement data — and produce OCO-conforming triples without ever touching the ontology directly.
- Drag&drop import from CSV, Excel, and selected ELN exports
- Pre-configured process templates from the L3 route templates
- SHACL validation of the ABox as you type
- Versioned provenance for every data point (DCL 0–5)
- Export as Turtle or JSON-LD, importable into HermiT and ROBOT
- Works against any OCO sub-bundle — from L1-only to the full distribution
SIPOC-Extractor
Pipeline for extracting structured SIPOC fragments (Supplier / Inputs / Process / Outputs / Customers) from scientific publications and patents. From PDFs to OCO-conforming ABox.
- PDF structure extraction via GROBID (TEI-XML with sentences, tables, references)
- LLM-supported fragment recognition with confidence levels (DCL 0–5)
- Audit trail via
LiteratureExtractionas first-classprov:Activity - Connected to OpenAlex, Crossref, and DPMA / Espacenet for metadata
- Conflict detection across multiple sources per sample
- Output is processed directly in the OCO-Workbench or as a TTL bundle
Both tools are under active development — available through an Early-Access programme for pilot users. Licence terms, scope of support, and release planning are agreed individually. Both use OCO as schema (not a fork); ontology updates flow through automatically.