What is a Historian?
A Historian (or Process Historian) is a specialised database that stores, manages and analyses large volumes of industrial process data.
It is used to store time-series data from Sensors, PLCs, SCADA systems and other OT devices quickly, efficiently and over long periods.
🧠 What does a Historian do?
A Historian:
- Collects data from industrial systems (such as SCADA, DCS, PLCs)
- Stores data as time series, including timestamp, value and status
- Optimises storage, even with millions of data points per day
- Makes data accessible for reporting, analysis, maintenance and troubleshooting
🧱 Typical characteristics of a Historian
| Characteristic | Description |
|---|---|
| High-performance storage | Optimised for time-series data, not for relational queries |
| Compression & aggregation | Reduces data volume without losing relevance |
| Real-time & historical | Supports both live Monitoring and long-term analysis |
| Integration with SCADA/MES | Receives data from existing automation systems |
| Data export | Supports linkage with BI tools, ERP, Cloud or dashboards |
📦 Examples of Historian software
- OSIsoft PI System
- AVEVA Historian (formerly Wonderware)
- GE Proficy Historian
- Siemens SIMATIC Process Historian
- Ignition Tag Historian (Inductive Automation)
🔄 Historian vs. SQL database
| Aspect | Historian | SQL database |
|---|---|---|
| Purpose | Time-series data from processes | Relational data |
| Performance with large data volumes | Optimised for speed & scale | Less efficient with millions of data points |
| Timestamps | Essential and automatic | Manual or not central |
| Compression | Advanced, for industrial data flows | Limited or absent |
🏭 Application examples
- Analysis of temperature, pressure or flow data in a factory
- Monitoring of energy consumption per installation or machine
- Supporting Predictive Maintenance
- Reporting on batch processes and quality data
- Archiving data for Compliance or audits (e.g. in pharmaceuticals)
🔐 Historian and cybersecurity
Because historians often collect data from OT systems and forward it to IT or Cloud, they need to be properly secured:
- Use of DMZs or data brokers
- Read-only access towards IT or Cloud environments
- Authentication and encryption for data transport
- Logging of access and modifications
📥 How are tags stored in a Historian?
Storing data in a Historian is not random, but follows clever storage strategies. This is needed in order to:
- Optimise database performance
- Avoid unnecessary storage of redundant data
- Capture only relevant changes or measurement points
There are several ways in which Tags (such as pressure, temperature or status values) can be stored in a Historian.
🕑 1. Cyclic storage (time-based)
In cyclic storage, the value of a Tag is periodically written, regardless of whether the value has changed.
| Property | Description |
|---|---|
| Interval | Configurable, e.g. every 1s, 5s or 1 min |
| Advantage | Consistent dataset, useful for Trending |
| Disadvantage | Potentially much redundant data |
For example:
Tank_Levelis stored every second, even if the value does not change.
⚙️ 2. Event-driven storage (Value Change)
In event-driven storage, the value is only stored on a change in the value (status or analogue).
| Property | Description |
|---|---|
| Storage on change | Yes, from a minimum deviation |
| Advantage | Less storage space, more relevant for analysis |
| Disadvantage | No regular timestamps in the dataset |
For example:
Valve_Open(0/1) is only logged when opened or closed.
📉 3. Hysteresis-based storage (deadband)
In hysteresis-based storage (also known as deadband Logging), an analogue value is only stored when the change is greater than a configured threshold.
| Property | Description |
|---|---|
| Threshold (±) | E.g. 0.5 °C difference since the last value |
| Advantage | Combines precision and efficiency |
| Disadvantage | May ignore small fluctuations |
For example:
Oven_Tempis only stored once the temperature has changed by more than 0.5°C since the previous measurement.
🔄 Combining methods
Many Historian systems combine methods, such as:
- Cyclic logging every 10 minutes
- Event-driven storage for status signals
- Hysteresis of 1% on analogue values
This produces a smart and efficient log of the most important process data.
📌 In summary
A Historian is a powerful industrial database for collecting, storing and analysing time-series data from production processes. It forms the basis for process optimisation, maintenance, quality control and Industry 4.0 analytics.