Reference Terms and Industry Standards

A listing for key terms, acronyms, and standards used throughout the Unified Manufacturing Data Architecture.

Key Terms

EIH — Edge Intelligence Hub

On-prem compute stack that ingests, validates, and contextualizes raw signals before publishing to the Unified Namespace.

CDM — Common Data Model

Domain owned schema that standardizes vocabulary, units, and business rules; enforced via JSON data contracts.

UNS — Unified Namespace

Real-time publish/subscribe backbone (e.g. MQTT) that carries CDM-tagged events in a hierarchical topic structure.

Data Contract

Clearly defined agreement governing data structure, units, and timeliness for each data product and the transmission of them between systems.

LLM Router

A UMDA service that directs tasks to the most appropriate language model (lightweight, domain, or general-purpose) based on complexity, latency, and cost.

UDL — Unified Data Layer

Governed local and centralized storage access consisting of harmonized CDM tables, time-series joins, and enterprise KPIs with full lineage.

FDL — Feedback Data Layer

Historical store of AI inferences, human inputs, and production outcome metrics used for model retraining and audits.

AI Agent

An autonomous service that consumes CDM/UNS data, detects events or anomalies, and writes decisions back to the Feedback Data Layer.

Data Product

Domain owned dataset published with a defined schema, quality SLA, lineage, and access policy so it can be reliably reused across edge, UNS, UDL, and AI layers.

RBAC — Role-Based Access Control

Permission model enforced at UNS topic and UDL table levels to support zero-trust security.

Industry Standards Foundational to UMDA

ISA-95 — Hierarchical model used to map Site, Area, Line, Equipment, and Unit IDs in a CDM.
ISA-88 — Batch terminology (Recipe, Procedure, Unit) aligned with CDM production domain.
ISA-99 / IEC 62443 — Cybersecurity framework for industrial control systems and IoT devices.
OPC UA — Platform independent protocol for secure, model-rich data exchange; common UNS broker option.
MQTT — Lightweight publish/subscribe transport protocol widely used for edge-to-cloud messaging.
Sparkplug B — Open payload specification that sits on top of MQTT for IoT tag naming, state, and auto-discovery.
B2MML — XML schemas that implement ISA-95 for enterprise-to-MES data exchange.
MTConnect — Open, read-only standard for collecting CNC and machine tool data via HTTP/JSON or MQTT.
GS1 EPCIS / GTIN — Global trade item and event standards for supply chain traceability used in CDM reference tables.
ISO 8000 — Data quality and master data governance standard referenced in UMDA stewardship policies.
ISO 10303 (STEP) — Neutral 3-D and BOM data format; useful when integrating PLM with the UDL.
ISO 9000 / 9001 — Quality management system standards guiding continuous improvement KPIs in UMDA rollouts.
ISO / IEC 22989 — Foundational concepts and terminology for Artificial Intelligence systems.
ISO / IEC 23894 — Guidance on risk management for AI applications; complements UMDA AI governance policies.
ISO / IEC 42001 (draft) — Management system standard for governing AI within organizations.
NIST Cybersecurity Framework — Identify / Protect / Detect / Respond / Recover model used for UMDA zero-trust design.
NIST SP 800-53 — Catalog of security & privacy controls mapped to UNS and UDL access policies.
NIST AI Risk Management Framework (1.0) — Voluntary guidance for trustworthy AI; informs FDL and model governance checks.
JSON Schema — IETF standard (2020-12) used to structure data contracts.
IEEE 1451 — Smart sensor interface standard enabling plug-and-play transducer data into the EIH.
W3C PROV-DM — Data provenance model leveraged for end-to-end lineage capture in the UDL.
IEEE 7000-Series — Ethical and transparency standards for autonomous and intelligent systems; aligned with UMDA AI agent guardrails.