Theme: History as evidence, not merely storage
Quick definition
A process historian is a system for collecting, timestamping, storing and retrieving historical process values, alarms and events from control and automation systems. It is optimised for many continuous time series and rapid trend access, but is not automatically a complete information model or legal archive.
In context: from signal to useful history
History begins at the source: a value receives identity, timestamp and quality status before storage. OPC UA Historical Access distinguishes between raw data, processed values and aggregates, among other concepts. A client can therefore request original samples or hourly averages without confusing the two.
A concrete operational case
After an overnight stoppage, the operations team needs to determine whether a pump, valve or sensor initiated the sequence. The historian aligns pressure, command, feedback, alarms and quality flags on one timeline. The analysis works only if systems share time, signals retain stable identities and gaps have not been filled without traceability.
Compression changes the evidence
Historians can reduce storage by retaining changes or representative points. This is valuable for long time series but affects which rapid events can be reconstructed. Collection rate, deadband, compression and retention should follow use case and consequence rather than a generic default.
Historian, data platform or both?
The historian remains close to operations and provides efficient access to process history. A broader data platform can combine it with maintenance, finance, weather and object metadata. Copying everything, however, creates competing truths. Define which system is authoritative for raw values, corrections, aggregates and long-term retention.
Opportunity and limitation
Well-designed history makes troubleshooting, energy review and performance comparison reproducible. Incorrect timestamps, unclear units, tag changes and filtered peaks can also produce precise but misleading charts. More data does not compensate for weak context.
HubMind’s view
Treat the historian as a measurement system with preserved provenance, not simply as the facility’s memory. Connect every series to stable object identity, unit, quality and change history. This separates what the facility did from what later analysis assumes.
Related concepts
OPC UA, Data Governance, information modelling and Big Data.
Sources and further reading
Scope and freshness
This page describes generic process history and OPC UA Historical Access 1.05.04 rather than a specific product. Reviewed 31 August 2026; review after a specification change or changed retention and information-classification requirements.