Open Specification v2.0

Industrial Asset
Event Standard

IAES standardizes what happens to industrial assets.
Not how you measure it. Not how you store it.

OPC UA moves data. Sparkplug B structures telemetry. IAES names what happened.

A minimal, vendor-neutral event language for industrial assets. IAES names what happens to equipment — measurements, diagnoses, failures, maintenance actions — so any system can produce, route, and consume asset events without custom integration.

Open standard. Vendor-neutral. Reference implementation by Wertek AI.

Who is this for?

For system integrators, OT engineers, and industrial developers who need to connect SCADA, CMMS, AI, and enterprise systems.

Industrial systems cannot share asset events

In every industrial plant, critical systems operate in isolation. Each one generates events about the same assets — but in completely different formats, APIs, and semantics.

⚙SCADA
⚗PLC
🔧CMMS
📊Historian
🤖AI Engine
⚡Energy
📡IoT Sensors
💼ERP
Different formatsJSON, XML, OPC UA, Modbus, CSV, proprietary
Different APIsREST, SOAP, MQTT, gRPC, file drops
Different semantics"alarm" ≠ "alert" ≠ "event" ≠ "notification"
SCADA
alarm_high_vibration
A register change in a PLC tag. No context, no asset identity, no severity model.
≠
SAP PM
TECO notification
A work order closed in a transactional system. Different schema, different lifecycle.
≠
AI Engine
bearing_fault_predicted
A probabilistic diagnosis with confidence score, RUL, and recommended action. Incompatible with both.
Same asset. Same event. Three incompatible representations.

There is no common language for asset events in industry. IAES provides one.

IAES in the Industrial Standards Landscape

IAES doesn't compete with existing standards. It occupies a different layer.

Standard Purpose
ISO 55000 Asset management principles
AAS (I4.0) Digital twin structure
ISA-95 Manufacturing operations architecture
OPC UA Machine-to-machine communication
IAES Event language for industrial assets

See full comparison: IAES vs Sparkplug B vs AAS vs OPC UA →

IAES covers events across all industrial domains:

Energy & Power Quality Maintenance Reliability Diagnostics & AI IoT & Telemetry Condition Monitoring Manufacturing Oil & Gas Mining Utilities Buildings & Facilities

From Sensor to Action

A sensor detects bearing wear. An AI model classifies the fault — or a technician records a diagnosis from a mobile app. Either way, IAES routes one event to three systems simultaneously.

Vibration Sensor

12.4 mm/s RMS
156 Hz

→

AI / Expert / App

bearing_inner_race
confidence: 0.94

→

IAES Event

asset.health
severity: critical
rul_days: 5

→

SAP PM

IW31 notification

MaintainX

Work order assigned

Grafana

Alert panel fired

One event, three consumers. No custom integration per target system. Whether the intelligence is artificial or human — the event format is the same.

How It Works

PLCs and sensors don't know IAES exists. The intelligence layer — AI models, rule engines, gateways — translates raw signals into standardized events.

1

Edge system detects a condition

A vibration sensor reads 12.4 mm/s on compressor C-12. An energy meter detects power factor dropping to 0.87. A thermal camera sees 94°C on a bearing housing.

2

Intelligence layer generates an IAES event

An AI model, rule engine, or gateway interprets the raw data and emits a standardized IAES event with asset identity, severity, and recommended action.

3

Event is consumed by any system

The same event can simultaneously create a work order in SAP, trigger an alert in Grafana, update a digital twin, and feed an AI model for trending.

SAP PM MaintainX Grafana AI Engine Historian
{ "spec_version": "2.0", "event_type": "asset.health", "event_id": "a9e3c4b2-...", "timestamp": "2026-03-09T14:30:00Z", "source": "acme.ai.vibration", "asset": { "asset_id": "COMPRESSOR-12", "asset_name": "Air Compressor C-12", "plant": "Pesqueria", "area": "Utilities" }, "data": { "severity": "medium", "health_index": 0.42, "failure_mode": "bearing_wear", "rul_days": 21, "recommended_action": "Schedule bearing replacement", "iso_13374_status": "unsatisfactory" } }
Key insight

The PLC doesn't know IAES exists. An AI engine reads raw vibration data and emits an IAES event. A technician opens an app, records a diagnosis, and that's also an IAES event. IAES is the language of the intelligence layer — whether the intelligence is artificial or human.

Incident Lifecycle

A complete incident from detection to resolution. Four IAES events tell the full story — no human intervention needed to close the alert.

08:12
asset.health · onset
Power factor drops to 0.91. Rule engine detects violation. New correlation_id is generated. First alert created.
08:45
asset.health · escalation
Power factor drops further to 0.84. Severity escalated from medium to high. Same correlation_id. No duplicate alert — existing alert updated.
10:03
maintenance.work_order_intent
Consumer creates a work order in MaintainX. Linked to the same correlation_id. Technician dispatched to inspect capacitor bank.
14:10
asset.health · recovery
Power factor recovers to 0.97 for 5 consecutive readings. Recovery event emitted with severity: info, condition_trend: improving. Alert auto-resolved.
correlation_id: 3b2f9d8c-1c33-4a8a — all 4 events share the same ID. One incident, one thread, full traceability.

Get Started in 30 Seconds

Python
pip install iaes
TypeScript
npm install @iaes/sdk
Node-RED, Go, and more →

Already implemented in

Python SDK TypeScript SDK Node-RED n8n workflows

Published on PyPI, npm, and Node-RED palette. Production-ready.

The Model

Any producer. One event format. Any consumer.

Producers
Vibration sensor
AI model
Technician app
Energy meter
➜
Standard
IAES
event
asset.measurement asset.health work_order_intent
➜
Consumers
CMMS / SAP PM
Analytics / BI
AI / ML pipeline
Historian / SCADA
Industrial asset → IAES event → Any enterprise system. One format. Zero custom integrations.

Philosophy

V

Vendor Neutral

No vendor lock-in. Any system can produce or consume IAES events without proprietary dependencies.

E

Event-Oriented

Events are first-class citizens. Each event is self-contained, traceable, and independently meaningful.

I

Intent, Not Command

IAES declares what should happen, not how. Consumers decide implementation — preserving autonomy and flexibility.

O

Open Standard

Licensed under CC BY 4.0. Free to use, implement, extend, and distribute. Open development on GitHub.

B

Backward Compatible

New versions add optional fields. Existing producers and consumers continue working without changes.

Open Source

IAES is developed in the open. The specification, JSON schemas, examples, and reference implementations are all available on GitHub. Contributions, issues, and discussions are welcome.

Licensed under Creative Commons Attribution 4.0 (CC BY 4.0)

Official Specification

Published Specification
IAES 2.0
Industrial Asset Event Standard
Version 2.0 · September 2026 · CC BY 4.0
Read the specification DOI: 10.5281/zenodo.22663498

Normative: IAES_SPEC.md, GOVERNANCE.md and the JSON Schemas. The RFCs record why each change was made — rationale, not authority; IAES-RFC-001 is the original 1.x memo.

IAES 2.0: 10.5281/zenodo.22663498 · every release (resolves to the latest): 10.5281/zenodo.18973216

Permanent, citable identifier via Zenodo (CERN). Indexed by OpenAIRE and Software Heritage. Uses IETF conventions (MUST / SHOULD / MAY per RFC 2119).

How to Cite IAES

If you use IAES in research, industrial systems, or procurement documents, please cite:

Plain text

Garza, G. (2026).
Industrial Asset Event Standard (IAES) v2.0.
Zenodo.
https://doi.org/10.5281/zenodo.22663498

BibTeX
@software{garza_iaes_2026,
  author       = {Garza, Gilberto},
  title        = {Industrial Asset Event Standard (IAES)},
  version      = {2.0},
  year         = {2026},
  publisher    = {Zenodo},
  doi          = {10.5281/zenodo.22663498},
  url          = {https://doi.org/10.5281/zenodo.22663498}
}
ResearchGate

DOI resolves permanently via Zenodo (CERN). Licensed under CC BY 4.0.

Explore More