Plant lakehouse
MES, SAP, quality inspection and line-sensor data in one Azure Databricks lakehouse.
- Where
- Mercedes-Benz
- When
- 2025 – now
- Context
- Phase 1 of the plant roadmap
- Type
- Data platform
10+sources unified
1governed source of truth
3 teamsproduction, quality, maintenance
How it works
- 1SourcesMES, SAP, QC, sensors
- 2IngestPySpark, Data Factory
- 3StoreDelta Lake, quality-checked
- 4ModelShared plant tables
- 5ServeDashboards, RAG, ML
The problem
Each team worked from its own extracts, so the same metric could show different numbers in different meetings.
What I built
- Ingested 10+ siloed plant sources with PySpark into Delta Lake on Azure Databricks.
- Modeled shared tables for production, quality and maintenance, with data-quality checks.
- Made the lakehouse the base for every later dashboard, assistant and model.
Stack
Azure DatabricksPySparkDelta LakeData FactoryADLSData modeling