To overcome the global chip shortage, increasing throughput is now a top priority for semiconductor FABs.

‍

Looking ahead, there are significant opportunities for long-term cost savings by optimizing, simplifying, or eliminating process steps in semiconductor manufacturing.

‍

Real-time, accurate, and actionable data is needed to realize these improvements.

‍

This new methodology is called Equipment and Process Co-Optimization, or EPCO for short. It uses real-time data to co-optimize processes and equipment in perfect synchronization.EPCO applies machine learning to engineering best practices. CopyExactly! helps to digitize the optimization of manufacturing processes and equipment.

‍

For example, statistical process control can look at the actual effects of chamber-to-chamber, machine-to-machine, and run-to-run differences, even when using the exact same recipe on the exact same equipment. Process control has become more complex as critical dimensions have shrunk and error tolerances have decreased. This means that chamber-by-chamber control is becoming the basis for thorough statistical process control and ensuring high productivity.

‍

This is exactly what EPCO strives for, optimizing equipment, chambers, and processes together, often using advanced machine learning techniques.

‍

At Atnerup has spent a great deal of time trying to understand the problems and challenges faced by FABs and equipment manufacturers. The result is Aston, a robust molecular sensor.

‍

Aston provides accurate and actionable real-time data that is essential for effective EPCO. This data allows us to build and test appropriate ML models.

‍

In fact, Aston was designed from the ground up to meet the needs of in-situ molecular analysis for EPCO.

‍

For more information, download Aston's application brief on how EPCO can be implemented in semiconductor manufacturing.