ItalyFood and restaurants2023 – ongoingBarilla Group
Machine-learning predictive maintenance on Databricks
Barilla used machine learning to schedule maintenance before equipment failures
Problem
Barilla's fragmented data infrastructure made manufacturing information difficult to combine. Equipment failures disrupted production and added operating costs.
Approach
Barilla centralized data on Azure Databricks and used MLflow to put machine-learning models into operation. Manufacturing teams used predictions of equipment failure to schedule maintenance proactively. Databricks reported that reducing unplanned downtime also reduced operating costs and equipment-related energy use.
Results
No published figures
Takeaway
Failure predictions supported maintenance scheduling within a broader manufacturing-data program. Barilla did not isolate numerical savings attributable to predictive maintenance.
Categories
- Industry
- Food and restaurants
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Databricks described Barilla's operational machine-learning applications, including failure prediction and proactive maintenance scheduling. It reported reduced downtime and associated costs without separating predictive-maintenance savings from the wider data program.
Full details are in the original sources.
