Companion specification · Tier 2

AdditiveManufacturing

Version1.0.0
Published2025-02
SectorAdditive manufacturing

The additive manufacturing model, published in 2025 and implemented by the umati sample server, which makes it one of the newer models you can actually explore for free. Additive machines have long build jobs, which changes what availability and performance mean.

Where it sits in the stack

Built on the Machinery building blocks and the Device Information Model.

The types that matter

TypeWhat it carries
(type inventory not confirmed)We have not verified the full type inventory. The umati sample server exposes this model, so browse it there.

Using it on a real machine

  1. Browse it on the umati serverThis model is one of the four the umati sample server implements, so you can see the real structure without hardware.
  2. Map the job, not the partAn additive build job runs for hours or days and may produce many parts at once. Job state matters more than a part counter.
  3. Check the Machinery layerIdentification, item state and operation mode underneath are the portable part.

What it gives you for OEE

Long build jobs make conventional OEE awkward and worth rethinking rather than forcing. Availability from machine state still works. Performance against an ideal cycle time does not mean much when a job takes 40 hours - build time against planned build time is the more useful ratio. Quality is a build outcome, often assessed after the fact, which means it arrives late and your figure has to tolerate that.

Try it free: The umati sample server at opc.tcp://opcua.umati.app:4840 exposes this model. It is read-only - writes are silently ignored - and you can run it locally with docker run -d -p 4840:4840 ghcr.io/umati/sample-server:main

Gotchas

  • Long jobs break part-cycle OEE assumptions. Agree what performance means before computing it.
  • Quality assessment often happens after the build completes, so quality data arrives late and retrospectively.
  • Published 2025 - implementations outside the sample server are still early.
Verify before you rely on this: We have not verified the full type inventory. The type inventory on this page is deliberately short. We list only type names we are confident of and would rather publish five correct ones than twelve with four wrong. Browse a real server or the released nodeset for the complete list.

Once the machine speaks a companion spec, the OEE mapping is portable across vendors, which is the point at which analytics stops being a per-machine project.

See Jemba →

Last verified 7 September 2026 against OPC UA 1.05.07. Version and publication date read from the released nodeset. OPC UA Connect is an educational resource and is not affiliated with the OPC Foundation.

Free download
OPC UA Machine Data → OEE Mapping Kit
Which nodes to subscribe to, how they become Availability, Performance and Quality, and which companion specification already gives you the answer, so you stop inventing tags.
Get the free mapping kit →
Free, no course, no login