Industrial Time-Series ML for Semiconductor Manufacturing
Build practical time-series machine learning models for predictive maintenance and yield insight.
Programme overview
This manufacturing programme equips participants to design, build and validate time-series machine learning models for predictive maintenance and yield insight in semiconductor assembly, test and packaging environments.
Participants work with real manufacturing-style sensor data — vibration, temperature, cycle time and yield traces — to design anomaly detection systems, control false alarms and build MVP-ready predictive maintenance pipelines.
What participants learn
- Understand the differences between static machine learning and time-series machine learning
- Design rolling window and lag-based feature engineering pipelines
- Build anomaly detection models for rare event detection
- Calibrate precision-recall trade-offs for industrial environments
- Design early-warning systems with lead-time validation
- Evaluate false positive cost against failure cost
- Design MVP-ready predictive maintenance pipelines
Programme details
- Category: Manufacturing & ESG
- Course type: Semiconductor Training
- Duration: Two (2) full-day workshop
- Who should attend: Engineers, data and equipment specialists applying machine learning to manufacturing sensor data.
Elite Indigo Consulting is a Penang-based capability development partner registered with HRD Corp; eligible programmes are HRD Corp claimable.