About This Course
This 3-day, hands-on course is designed to help participants turn fab/process data into higher wafer yields by leveraging leakage-safe machine learning (ML), explainability techniques (such as SHAP), and Run-to-Run (R2R) control/Advanced Process Control (APC). Participants will build and evaluate predictors, prototype virtual metrology, and leave with a practical, data-backed plan to improve wafer yield.
Aim
The course will empower participants to convert fab data into measurable yield gains using cutting-edge ML, explainability methods, and R2R/APC techniques.
Course Objectives
- Learn the fundamentals of IC yield and baseline analytics.
- Apply leakage-safe ML to predict yield effectively.
- Gain hands-on experience in feature engineering and model evaluation.
- Optimize processes using R2R/APC techniques and virtual metrology.
- Interpret yield drivers and develop actionable yield improvement plans.
- Understand MLOps, real-time monitoring, and deployment in fab environments.
Course Structure
📅 Module 1 – Foundations: Yield and Process Data Analytics
- IC Yield Basics: Gain a solid understanding of yield types and their impact on production costs.
- Process Data Overview: Explore key data sources (MES, SPC, wafer maps) and the challenges related to data quality.
- Baseline Analytics for Yield: Use classical defect models and loss analysis tools to analyze yield.
- Hands-On: Engage in dataset creation, basic analysis, and yield estimation exercises.
📅 Module 2 – Machine Learning for Yield Prediction
- Feature Engineering & Data Preparation: Learn how to handle data imbalances and define relevant features for prediction models.
- ML Models for Yield Prediction: Dive into regression and classification techniques tailored for yield prediction.
- Model Evaluation & Interpretability: Explore key metrics and explainability tools like SHAP to interpret model outputs.
- Hands-On: Build your own yield prediction model and learn to interpret the results effectively.
📅 Module 3 – Optimization and Real-time Control
- Run-to-Run (R2R) Control & APC: Use machine learning to optimize production parameters and enhance yield.
- Process Parameter Optimization: Apply Bayesian optimization techniques to improve yield.
- MLOps & Digital Twins: Learn how to deploy models in real-time environments for continuous monitoring.
- Hands-On: Apply R2R control, simulate optimization techniques, and present strategies for yield improvement.
Course Outcomes
- Build and evaluate ML models for yield prediction.
- Apply SPC, Pareto, and classical models for yield analysis.
- Optimize processes using R2R/APC and virtual metrology.
- Implement feature engineering and time-series models.
- Interpret yield drivers using SHAP and permutation importance.
- Develop actionable yield-improvement plans.
- Gain insights into MLOps and best practices for fab data.
Who Should Enrol?
- PhD scholars and postgraduate students in microelectronics, VLSI, materials science, or data science.
- Academicians and researchers involved in semiconductor manufacturing or ML.
- Fab professionals: process/yield/device/test/metrology/equipment engineers, DFM/PDK, MES/IT, quality/OEE engineers.
- Data scientists and ML engineers supporting fab analytics, virtual metrology (VM), or APC/R2R systems.









Reviews
There are no reviews yet.