Glossary · IIoT, data and AI
Data lineage
Also known as: Data provenance
German: Datenherkunft
In data management, data lineage is the documented path of data from its origin through all transformations, systems and uses to its current form. It shows where a value came from, how it was calculated and what depends on it.
- IIoT
In one sentence
Data lineage documents where data comes from, how it was transformed and which reports, models or documents depend on it.
Example
Lineage shows that the OEE figure in the monthly report is calculated from PLC state signals, filtered by the MES shift calendar and aggregated in the warehouse.
How it applies
- Engineering: Lineage helps assess the impact of changes: if a tag is renamed or a calculation changes, lineage shows which dashboards, models and interfaces are affected (Impact analysis).
- AI: For machine learning, lineage links a model version to the exact Training data and preprocessing used, which is necessary for reproducibility and investigation of errors.
- Documentation: Where published documents contain generated data, such as technical data tables, lineage shows which source system and version the values came from.
Data lineage vs. audit trail
An Audit trail records who changed what and when, typically for a record in a single system. Data lineage records how data flows and is transformed across systems. Both support traceability; they answer different questions.