The Purdue Model — the hierarchical architecture that has structured industrial control systems for three decades — was designed for deterministic, isolated, hierarchical control. It is excellent at what it was built for. It is poorly suited to the data architecture that AI requires: high-frequency data access across levels, event-driven real-time processing, and the ability to feed AI models with context-rich, normalised data from multiple sources simultaneously.

Most industrial companies have the data their AI needs. The problem is that the data is trapped: in proprietary DCS historians, behind network architectures that do not allow cross-level communication, and in formats that differ between every vendor's platform. Getting that data to AI models — reliably, securely, and at sufficient frequency — is an architectural problem that precedes any AI algorithm selection.

Four principles address this problem systematically. They are not a prescription to replace existing OT systems — most legacy systems will remain in place for fifteen to twenty-five more years. They are a blueprint for the data and network layer that sits between OT systems and AI applications, enabling AI access without disrupting the deterministic control logic that keeps processes safe and stable.

Why Current OT Architecture Limits AI

Diagram 1 — Current Siloed Architecture vs. AI-Centric Target Architecture
❌  Current: Siloed Architecture
flowchart LR C1["ABB 800xA Historian\n(proprietary format)"] C2["Siemens PCS neo\n(OPC DA only)"] C3["Yokogawa Exaquantum\n(custom API)"] C4["CMMS\n(SAP PM export)"] C5["Paper maintenance logs\n(not connected)"] AI1["AI Analytics\nPartial data only\nIncomplete · Stale · Inconsistent"] C1 & C2 & C3 & C4 -.->|"Manual / batch / adapter"| AI1 C5 -.->|"Not connected"| AI1 style C1 fill:#fee2e2,color:#7f1d1d,stroke:#fca5a5 style C2 fill:#fee2e2,color:#7f1d1d,stroke:#fca5a5 style C3 fill:#fee2e2,color:#7f1d1d,stroke:#fca5a5 style C4 fill:#fee2e2,color:#7f1d1d,stroke:#fca5a5 style C5 fill:#f1f5f9,color:#475569,stroke:#94a3b8,stroke-dasharray:4 4 style AI1 fill:#fecaca,color:#7f1d1d,stroke:#dc2626,stroke-width:2px
▼  migrate to
✓  Target: AI-Centric Architecture
flowchart LR ABB["ABB 800xA"] & SIE["Siemens PCS neo"] & YOK["Yokogawa DCS"] & CMMS2["CMMS"] --> T1["OPC-UA / MQTT\nIngestion Layer"] T1 --> T2["DMK Data Lake\nTime-series + Context + Knowledge"] T2 --> T3["AI Model Layer\n(reads from single\nnormalised source)"] style T1 fill:#dbeafe,color:#1e3a5f,stroke:#3b82f6,stroke-width:2px style T2 fill:#e0e7ff,color:#1e3a5f,stroke:#0f172a,stroke-width:2px style T3 fill:#dcfce7,color:#14532d,stroke:#22c55e,stroke-width:2px

Principle 1 — The "Data As-Is" Strategy: The DMK Lake

OT data is inherently messy. Tag naming conventions differ between every DCS and historian installation. Timestamps use different time zones, different resolutions, and different epoch references. Units of measurement are inconsistent. A temperature reading from one system is in Celsius with 1-second resolution; the same type of reading from a neighbouring system is in Kelvin sampled every 30 seconds.

The "Data As-Is" strategy deliberately avoids the temptation to fix this at the source — which would require modifying running production systems and creating extensive validation work. Instead, it captures raw data as it exists, then normalises it once in a unified "Data-Model-Knowledge (DMK) lake." The lake has three layers:

Once the DMK lake exists, AI models consume data from a single, normalised source. Adding a new historian to the system requires only a new OPC-UA connection and source mapping — not a new custom integration for every downstream AI application.

Diagram 2 — DMK Data Lake Architecture
flowchart TD subgraph Sources["OT Data Sources"] S1[DCS Historians] S2[PLC / SCADA systems] S3[Field Instruments - HART/FF] S4[CMMS / ERP] S5[Lab / Quality systems] end subgraph Ingestion["Ingestion Layer"] I1["OPC-UA / MQTT collectors"] I2["Protocol adapters\n(Modbus, DNP3, Profibus)"] I3["API connectors\n(REST, ODBC)"] end subgraph DMK["DMK Data Lake"] D1["Raw Time-Series Store\n(InfluxDB / Azure Data Explorer / PI)"] D2["Context Layer\n(Asset hierarchy · Tag mapping · P&ID links)"] D3["Knowledge Graph\n(Asset relationships · Failure modes · Maintenance history)"] end subgraph Consumers["AI Consumers"] C1[Predictive Maintenance Models] C2[Energy Optimisation AI] C3[Process Anomaly Detection] C4[LLM Agents at Levels 4-5] end Sources --> Ingestion --> DMK D1 --> D2 --> D3 DMK --> Consumers style Sources fill:#e2e8f0,color:#1e3a5f,stroke:#94a3b8 style Ingestion fill:#dbeafe,color:#1e3a5f,stroke:#3b82f6 style DMK fill:#e0e7ff,color:#1e3a5f,stroke:#0f172a,stroke-width:2px style Consumers fill:#dcfce7,color:#14532d,stroke:#22c55e

Principle 2 — Event-Driven Interfaces for Real-Time AI

Traditional OT data integration relies on polling: system A asks system B "what is the current value?" at a fixed interval. If the interval is 60 seconds, the maximum freshness of the data is 60 seconds — and 99% of the polls return "same as before." This is fine for daily reports. It is not adequate for AI models that need to react to process anomalies in real time.

Event-driven architecture inverts the model: systems publish events when their state changes. Consumers subscribe to the events they care about and react immediately when an event arrives. There is no polling loop, no wasted bandwidth, and no artificial latency introduced by a polling interval.

Diagram 3 — Polling vs. Event-Driven: The Latency and Efficiency Difference
graph LR subgraph Poll["Polling Model"] PA["AI System\nasking every 60s"] PH["Historian"] PA -->|"Poll 1: any change?"| PH PH -->|"No change"| PA PA -->|"Poll 2: any change?"| PH PH -->|"No change"| PA PA -->|"Poll 3: any change?"| PH PH -->|"⚠ Anomaly detected — but when?"| PA end subgraph Event["Event-Driven Model"] ES["Sensor / PLC"] EB["Message Broker\n(Kafka / MQTT / Event Hub)"] EA["AI System\n(subscribed)"] ES -->|"Event: value exceeds threshold"| EB EB -->|"Immediate delivery to subscriber"| EA EA -->|"Action triggered in <1s"| EAct["Alert / Work Order / Control signal"] end style Poll fill:#fff1f2,color:#7f1d1d,stroke:#fca5a5 style Event fill:#f0fdf4,color:#14532d,stroke:#bbf7d0

For OT environments, event-driven integration using MQTT (for field devices) and OPC-UA PubSub (for DCS and SCADA systems) is the modern standard. When a sensor detects a condition that exceeds a threshold, it publishes an event to a message broker. The AI model subscribed to that topic receives the event within milliseconds, runs its analysis, and triggers the appropriate response — an alert, a work order, an operator notification, or a control signal — without any polling delay.

Principle 3 — Platform-Based OT Devices and Zero-Touch Provisioning

One of the most persistent security problems in OT environments is the device lifecycle: a controller or HMI is delivered from the manufacturer with default credentials, a specific firmware version, and a default network configuration. It is installed, given a static IP address, and configured manually by an engineer or integrator. The initial setup is rarely documented in enough detail to reproduce. Firmware updates require the same manual process. When the device reaches end-of-life, decommissioning is often not systematic — devices get disconnected from the physical process but remain connected to the network.

Platform-based OT devices change this model through "bare metal onboarding" and zero-touch provisioning:

  1. The device is powered on with no factory configuration loaded
  2. It contacts a secure provisioning server and authenticates using a hardware-based identity (TPM chip)
  3. The provisioning server delivers the operating system, application stack, configuration, and credentials over an encrypted channel
  4. The device comes online in a known, verified state — no default passwords, no undocumented configuration

This approach eliminates the most common source of initial OT device compromise (default credentials). It also enables rapid, repeatable deployment across multiple sites and simplifies patching — update the platform image centrally, redeploy to devices in the next maintenance window.

Principle 4 — Zero Trust and Microsegmentation

The traditional OT security perimeter — a firewall between IT and OT networks, with implicit trust inside the OT network — has two critical weaknesses in an AI-centric architecture. First, AI data pipelines require controlled, cross-boundary communication that a simple IT/OT firewall cannot easily accommodate without becoming a bottleneck. Second, the AI systems themselves — running on cloud infrastructure or enterprise servers — need to query OT data, creating new pathways that the old perimeter model was not designed to manage.

Zero Trust provides the framework: every user, every device, and every workload must be authenticated and authorised before accessing any resource, regardless of where it is on the network. Microsegmentation implements this at the network level: the OT network is divided into small, tightly defined zones with controlled, monitored conduits between them.

Diagram 4 — OT Network Microsegmentation: Zones and Controlled Conduits
flowchart TB ENT["Enterprise Zone — Levels 4 to 5\nERP · Analytics · AI Platform"] DMZ["OT DMZ — Controlled Data Bridge\nOPC-UA proxy · Firewall · Protocol inspection · Logging"] SCADA["SCADA / HMI Zone — Level 2 to 3\nHistorian · HMI Workstations"] CTRL["Control Zone — Level 1 to 2\nDCS Controllers · PLCs"] FIELD["Field Zone — Level 0 to 1\nSensors · Actuators"] SIS["Safety Systems\nAir-gapped — no network connection"] ENT <-->|"Authenticated, read-only\nmonitored conduit"| DMZ DMZ <-->|"Protocol-filtered\nconduit"| SCADA SCADA <-->|"Vendor-specific\ncontrolled conduit"| CTRL CTRL <-->|"Hardwired / certified\nfieldbus only"| FIELD SIS -.-|"Physically isolated"| FIELD style ENT fill:#dbeafe,color:#1e3a5f,stroke:#3b82f6,stroke-width:2px style DMZ fill:#fef9c3,color:#713f12,stroke:#ca8a04,stroke-width:2px style SCADA fill:#e0e7ff,color:#1e3a5f,stroke:#6366f1,stroke-width:2px style CTRL fill:#fff7ed,color:#7c2d12,stroke:#1e3a5f,stroke-width:2px style FIELD fill:#dcfce7,color:#14532d,stroke:#22c55e,stroke-width:2px style SIS fill:#f8fafc,color:#1e3a5f,stroke:#94a3b8,stroke-dasharray:5 5

For AI data access, the DMZ is the key design element. The AI platform at the enterprise level queries the OPC-UA data gateway in the DMZ. The gateway proxies the query to the OT historian — the historian never receives a connection from the enterprise network directly. All queries are logged, filtered by protocol, and rate-limited. An AI model that needs real-time data gets it through this controlled channel; a compromised AI system cannot directly reach OT control systems even if it tries.

Building the Architecture in Stages

These four principles do not require a full OT transformation programme before the first AI model can run. A practical implementation sequence builds capability incrementally:

StageWhat to BuildWhat You EnableTypical Duration
1OPC-UA / historian connection to a central data store (read-only, no OT network changes)Single data source for first AI model; baseline for anomaly detection4–8 weeks
2Context layer — tag-to-asset mapping, asset hierarchyAI models can understand what each data point represents; search and query by asset4–12 weeks
3Event-driven feeds for key assets (MQTT or OPC-UA PubSub)Real-time AI response; anomaly detection with sub-second latency8–16 weeks
4Network microsegmentation and OT DMZSecure, auditable AI data access; dramatically reduced attack surface12–24 weeks (with planned maintenance windows)
5Zero Trust identity controls for all OT accessVendor access controlled, all sessions recorded; full NIS2 Article 21 alignmentOngoing programme

Stage 1 can often be completed without any changes to running OT systems — only read-only connections to existing historians. This is the right starting point: gain data access, prove the AI use case, and then invest in the deeper architectural work with a validated business case behind it.

What These Principles Are Not

These architectural principles are not a case for making OT look like IT. The deterministic, validated, change-averse nature of OT control systems is not a flaw — it is a design requirement for safe industrial operation. A DCS running a refinery process column must behave predictably and safely at all times. Nothing in this architecture changes that.

The DMK lake, event-driven interfaces, platform-based provisioning, and zero-trust network model all operate above and around the OT control logic — enabling the data access and security posture that AI requires, without touching the control systems that cannot afford to be disrupted. The safety, reliability, and determinism of OT control remain fully intact. The architecture simply adds the data and security infrastructure that allows AI capabilities to run alongside it.