Schema

In this section we present the results of an ongoing effort to design a reference database Schema intended to serve as a common language for the collection, exchange, and analysis of animal feed basket data across multiple platforms and research programmes.

  • The first draft schema was produced during a Solutions Group workshop in Edinburgh hosted by Livestock Data for Decisions (LD4D) in January 2026.

    Peter Steward from the Alliance of Bioversity International and CIAT led the design of the schema, leveraging the shared knowledge of the group across different studies and applications.

    Afterwards co-design workshops validated the draft schema, highlighting potential improvement and revisions.

    A more thorough gap analysis of the model was then conducted, to contrast its structure against that of available Feed Baskets databases from Cornell, ILRI and the ERA Livestock Data Dictionary. The main conclusion is that the Edinburgh schema could be fairly robust with some modifications. The report makes recommendations on how to adapt the Edinburgh Schema to create a model that can hold datasets from the major institutions e.g. Cornell, ILRI, FAO.

    This report sets the foundation for further developing a schema that could be mainstreamed across major institutions working with feed basket data. 

  • A flowchart with coloured labels and arrows showing a draft schema

    A snapshot of the proposed draft Feed Baskets Data Schema. Click to download the full-resolution PDF version.