3yearplan/schemas/uns/README.md

UNS Subtrees — initial draft

Status: WORK-IN-PROGRESS draft (2026-04-30). Initial structural skeleton seeded from the 3-year-plan source files (current-state.md enterprise layout + goal-state.md UNS naming hierarchy standard) and OpenText facility documentation. Per-site line and equipment detail is not yet populated — that requires the UNS hierarchy snapshot walk (see goal-state.md → "Unified Namespace (UNS) posture → UNS naming hierarchy standard"). See QUESTIONS.md for the open questions that block finalization.

Quick stakeholder view: HIERARCHY.md — high-level tree across all 6 sites with Warsaw expanded as a worked example.

This directory holds one JSON file per integrated site, declaring the canonical UNS subtree (Enterprise → Site → Area → Line) for that site. Equipment (level 5) is configured per-cluster in OtOpcUa, not declared here.

What's drafted

Site	File	Source of truth	Notes
Warsaw West	warsaw-west.json	OpenText Zimmer Warsaw - Building Use Descriptions.docx + verbal confirmation from project owner (numbered = legacy Zimmer = West Campus)	Six production buildings: bldg-2, bldg-5, bldg-7, bldg-10, bldg-19, bldg-20. Non-production buildings (3, 4, 8, 14, 15, 17, 18) deliberately excluded — Areas are production-only per goal-state. Lines pending walk.
Warsaw North	warsaw-north.json	OpenText Zimmer Biomet Buildings folder + verbal confirmation (lettered = legacy Biomet = North Campus)	Six lettered buildings: bldg-a, bldg-b, bldg-c, bldg-d, bldg-e, bldg-i. Production-vs-non-production breakdown per building is not yet known — all included pending walk. Note F/G/H are absent from the building list; verify whether they exist.
Shannon	shannon.json	current-state.md other-integrated-sites + "single server cluster covering the whole site" rule	Area = _default (single-cluster site). Lines pending walk.
Galway	galway.json	same	same
TMT	tmt.json	same	Area = _default. Lines pending walk. TMT shortname unconfirmed — see QUESTIONS.md Q2.
Ponce	ponce.json	same	same

The not-yet-integrated smaller sites (Berlin, Winterthur, Jacksonville, others) are deliberately not seeded — onboarding shape is still TBD and the site list is volatile per current-state.md.

What still needs to happen

1. Walk integrated sites to populate the lines arrays in each file above. Walk inputs: System Platform IO config, Ignition OPC UA connections, ScadaBridge templates. Walk output: per-site list of production lines with their name (kebab-case), displayName, and the equipment instances belonging to each line (with stable UUIDs — assigned during the walk, not derived from existing identifiers).
2. Confirm Warsaw North production-vs-non-production breakdown — the lettered buildings are all in warsaw-north.json pending verification of which are actually production. The walk should resolve this; non-production buildings should be removed from the Areas list once identified.
3. Resolve remaining QUESTIONS.md — Q2–Q4 (naming conventions) before authoring real line names, so the walk doesn't have to be redone.
4. Move to a dedicated schemas repo once one is created and an owner team is named. The temporary location under 3yearplan/schemas/ is documented in ../README.md.

Validation

Each file is validated against ../format/uns-subtree.schema.json. Constraints:

• enterprise must equal zb across every file (matches ServerCluster.Enterprise in OtOpcUa).
• site, area name, and line name must match ^[a-z0-9-]{1,32}$ or be the literal _default.
• _default is reserved as the explicit "this level doesn't apply at this site" placeholder, per the goal-state design goal of uniform path depth.

Worked examples

• example-warsaw-west.json — original schema worked example contributed with the seed (uses placeholder bldg-3/bldg-4 — superseded by the authoritative warsaw-west.json for any actual consumer; kept here for documentation purposes only).