SwitzerlandChemicals and materials2022 – ongoingClariant
TrendMiner process analytics with machine learning in chemical plants
Clariant found the root cause of plant anomalies with basic analysis in about 85% of cases and used machine learning for the rest
Problem
Clariant wanted to cut raw material use and shorten batch cycle times. Process engineers who were not statisticians needed to analyze time-series data from plants.
Approach
Clariant adopted TrendMiner, a self-service analytics tool, first at its German plants and later at sites worldwide. It built a cloud data lake, trained each site and named a core team member per site. Python notebooks added machine learning for harder cases, such as setting an acceptable temperature range and tracing a batch anomaly.
Results
Takeaway
Most problems yielded to simple statistics. Machine learning was needed only for the harder remainder.
Categories
- Size
- Large enterprise
- Industry
- Chemicals and materials
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Company's own claim
Sources
Source summaries
A Clariant author describes the rollout of TrendMiner from German plants to sites worldwide and the use of Python notebooks for machine learning. Statistical analysis found the root cause of anomalies about 85% of the time, and machine learning handled the other 15%. The article gives no savings figures.
Full details are in the original sources.
