OSIsoft PI System

Introduction

OSIsoft PI System is an industrial data infrastructure platform and Historian solution for real-time storage, processing and distribution of operational data within industrial environments. The platform was originally developed by OSIsoft and is now part of AVEVA.

PI System is used worldwide in:

• process industry
• power supply
• oil and gas
• water treatment
• production automation
• pharmaceutical industry
• utilities
• critical infrastructure

In modern OT environments, PI System often acts as the central data layer between:

• SCADA
• PLC
• DCS
• MES
• ERP
• analytics platforms
• cloud environments
• AI systems

The platform plays an important role in IT OT Convergence by making real-time OT data accessible for enterprise analytics and decision making.

🏗️ Basic architecture

PI System consists of several integrated components.

Key elements:

The architecture supports:

• real-time data acquisition
• long-term storage
• event processing
• analytics
• dashboards
• enterprise integration

PI System often acts as a central OT data hub within industrial enterprises.

⚙️ PI Data Archive

The core of PI System is the PI Data Archive.

This component processes:

• real-time process values
• event data
• alarms
• machine telemetry
• energy metrics
• status information

Key characteristics:

PI Archive is optimised for time series data with very high data rates.

🧠 Asset Framework

PI AF (Asset Framework) is the semantic layer of PI System.

AF structures industrial data around:

• assets
• processes
• installations
• production lines
• locations

Examples:

Benefits:

• standardised data models
• contextual analysis
• easier dashboards
• better scalability

PI AF plays an important role in digital transformation projects.

📡 Data collection

PI System supports extensive data collection from OT systems.

Key sources:

• PLC
• SCADA
• DCS
• industrial sensors
• energy platforms
• edge gateways

Supported protocols include:

• OPC UA
• OPC DA
• Modbus
• Modbus TCP
• Ethernet IP
• MQTT
• REST APIs

PI Interfaces and PI Connectors collect data from a wide range of OT sources.

🌐 PI in OT architectures

PI System generally sits between OT and enterprise IT.

Typical positioning within the Purdue Model:

PI often acts as the central OT data layer for:

• MES
• ERP
• cloud analytics
• predictive maintenance
• AI systems

This creates a single central source for operational data.

⚡ Compression and storage optimisation

PI System uses advanced compression techniques for efficient storage.

Key techniques:

In industrial environments, millions of data points per second can be processed.

PI is designed to:

• limit storage costs
• reduce network load
• maintain real-time performance

📊 Real-time analytics

PI System supports real-time data analysis.

Key applications:

• KPI monitoring
• energy analysis
• process optimisation
• predictive maintenance
• alarm analysis
• trend analysis

PI Analytics supports:

• calculations
• event detection
• condition monitoring
• real-time correlations

This enables faster detection of anomalies.

🖥️ PI Vision and dashboards

PI Vision provides web-based visualisation of industrial data.

Functionality:

PI Vision is used by:

• operators
• engineers
• maintenance teams
• management
• data analysts

The visualisation layer plays an important role in digital factories.

☁️ Cloud and IIoT

PI System supports integrations with modern Industrial Internet of Things architectures.

Key integrations:

• cloud platforms
• edge computing
• AI analytics
• digital twins
• data lakes

Supported platforms:

PI is therefore often the OT data source for enterprise AI projects.

🔄 Event Frames

PI Event Frames structure time-bound events.

Examples:

• batch processes
• alarm periods
• production runs
• faults
• maintenance activities

Benefits:

Event-based analysis is becoming increasingly important within smart manufacturing.

🧩 PI versus traditional Historians

PI is often considered the industry standard within Historian platforms.

Comparison:

PI stands out mainly through:

• scalability
• ecosystem
• data modelling
• enterprise integration

🔐 OT cybersecurity

PI System often contains business-critical OT data and requires strong security.

Key threats:

• Ransomware
• credential misuse
• unauthorised access
• lateral movement
• API abuse
• supply-chain attacks

Important security measures:

PI System is typically placed within:

• DMZ
• IDMZ
• data platform zones

🛡️ High availability

PI supports extensive high availability functionality.

Key techniques:

In critical infrastructure, continuous availability of process history is essential.

⚡ Performance and scalability

PI System supports very large industrial environments.

Key scalability factors:

PI is used worldwide in installations with:

• millions of tags
• thousands of assets
• global infrastructure

🔄 Lifecycle Management

PI System requires active Lifecycle Management.

Important points of attention:

• interface management
• patch management
• certificate management
• AF structures
• retention management
• backup validation

In OT environments, upgrades often require extensive validation due to dependencies on production processes.

🧪 Practical example: power plant

A power plant uses PI System as the central OT data hub.

Architecture

Data flows

Benefits

• central data hub
• scalable historisation
• real-time analytics
• enterprise integration

Security challenges

Key risks:

• insufficient segmentation
• uncontrolled APIs
• credential misuse
• ransomware
• cloud exposure

Architectures are therefore designed according to:

• IEC 62443
• Defense in Depth
• Zero Trust

⚖️ Relevant standards

PI System is often used within architectures based on:

📈 Role in IT/OT convergence

OSIsoft PI System plays an important role in modern data-driven industrial architectures.

Key trends:

• real-time analytics
• edge computing
• predictive maintenance
• AI integration
• digital twins
• cloud-native OT

Benefits:

• high scalability
• strong integrations
• real-time visibility
• extensive historisation
• contextual analysis

Challenges:

• cybersecurity
• management complexity
• data governance
• lifecycle management
• cloud integration

PI System is thus one of the most important industrial data infrastructure platforms within modern IT/OT-converged environments.