AI Techniques in Semiconductor Manufacturing

_{^{AI Techniques in Semiconductor Manufacturing}}

Core Objectives:

This workshop is designed to help participants understand:

• The real-world deployment, impact, and lessons learned from AI techniques

• Strategies for building observable and scalable multi-agent workflows

• The industry transition from restrictive data silos toward autonomous discovery.

• Methods for achieving tangible business results and automation through diverse AI applications 

We’ve curated a technical agenda featuring Topics such as:

• Agentic AI

• Gaussian Process Regression

• Time Series Modeling

• Bayesian Optimization

• And more...

You will walk away with actionable insights on:

• Yield Enhancement through Edge-Driven Defect Detection and Classification
  • Demonstrates how edge AI leverages real-time sensor and image data—through virtual metrology, anomaly detection, and SPC-integrated feedback loops—to enable early defect detection, classification, and yield optimization.
• From Prediction to Action: Causal AI for Real-Time Root-Cause Analysis in Semiconductor Manufacturing

  • How causal AI unifies multi-modal fab data into an intelligence layer to move beyond predictive alerts toward rapid root cause identification, prescriptive actions, and continuous improvement—significantly reducing RCA time and improving yield and operational efficiency.

• From Data Silos to Autonomous Discovery: Agentic AI in Semiconductors

  • Agentic AI in semiconductor workflows as autonomous, decision-making systems (beyond RAG) that orchestrate multi-agent reasoning across complex fab data—requiring robust data foundations, domain-grounded algorithms, SME-driven knowledge (including reasoning traces), and structured evaluation frameworks (agentic harness) to ensure reliable, scalable deployment.

• Making Sense of Equipment Time‑Series Data: From Signals to Insight
  • Introduces how semiconductor equipment time-series data (sensor signals, traces, run-to-run data) can be processed and modeled to uncover equipment behavior over time, enabling engineers to detect drift, changes, and anomalies for improved monitoring and troubleshooting.
• Achieving Business Results from Air-gapped Agentic AI Automation in Semiconductor Manufacturing - Lessons from the Field
  • Highlights real-world deployment of agentic AI in fabs, demonstrating how autonomous systems are driving yield, capacity, and operational gains across engineering workflows (FDC, yield, test, maintenance) while navigating constraints like data privacy and legacy integration—along with practical adoption strategies and hands-on use cases.
• And more...