UN/CEFACT Web Vocabulary

This is the documentation site for the UN/CEFACT Web Vocabulary (Vocab). It includes the semantics from the CEFACT Buy-Ship-Pay reference data model (BSP-RDM) as well as the underlying vocabulary for the UN Transparency Protocol (UNTP) . It also includes all UN/CEFACT code lists except LOCODE which is managed separately. It is maintained in the related Gitlab repository as a suite of ontology files in turtle notation. The turtle files are the single source of truth for all published outputs including all the tables and diagrams in the ontology section of this site as well as the normative and multi-format published output .

Understanding the Ontology

The Vocab is organised into seven domains, including a shared Core. Click a domain to explore its classes, properties and code lists. These pages are informational and are designed to help users gain a quick understanding of the ontology so that it can be used in any application.

Ontology architecture, governance, and external vocabularies.

The links below are designed to answer questions like “how does this ontology work?”, “How is it maintained and governed?” and “How does it relate to other ontologies?”

Architecture: how the vocabulary is authored, built and deployed.
Governance: term lifecycle, versioning, change process, and the auditable build trail (sources, defensible fork, traceability, validation).
External Vocabularies: the peers the Vocab reuses, imports and maps to, and the normative mappings (correspondences to schema.org, QUDT, FIBO, GS1, WCO, and more).

Using the Ontology

The whole purpose of the ontology is that it is used. Even in the age of AI supported translations from any format to any other, well defined ontologies like the Vocab play an important role by eliminating risks of AI hallucination in automated mappings and also dramatically simplifying the task of data integration between internal systems and external standards.

Think of the ontology like a dictionary of terms for international trade, finance & transport and supply-chain traceability & transparency. Use the ontology just like you would use a dictionary for writing letters - ensure terms in your structured documents link to corresponding definitions in this vocabulary or other well governed ontologies. There are several ways to manage the linking between your documents and reference vocabularies like this one but a common technology used today is called JSON-LD . In the conceptual example below:

A trade document like an invoice and a value chain document like a product passport are each described by a JSON Schema . That defines (and validates) the structure of the document but not necessarily the meaning of all the terms.
Each instance document (the invoice and the product passport) are just plain JSON but carry a JSON-LD context reference that essentially says “look in that dictionary for the meaning of this term”.
A single context file can link to multiple reference vocabularies so that you can always use the best fit definition for your use case. In the example below we show the Vocab vocabulary (sufficient for the invoice) but also the FAO AGROVOC vocabulary that is also used by the product passport for agricultural product terminology precision.

Take a look through the UN Verifiable Trade Document (UNVTD ) library or the UN Transparency Protocol (UNTP ) credential library to see real examples of structured documents with both schema defined shapes and vocabulary defined meanings.