How Machine Learning Can Transform Semiconductor Manufacturing Through Equipment & Process Co-Optimization
Right now, increasing throughput is top of mind for semiconductor FABs, as they attempt to overcome the challenges of the global chip shortage.
Looking beyond this, there are significant opportunities for long-term cost savings that can come from optimizing, simplifying or removing processing steps in semiconductor manufacturing.
To achieve these improvements, there’s a need for real-time, accurate and actionable data.
This emerging practice is known as Equipment and Process Co-Optimization, or EPCO for short. It involves the use of real-time data to jointly optimize processes and equipment, fully in sync. EPCO applies machine learning to the engineering best practices of CopyExactly! to help bring optimization of manufacturing processes and equipment into the digital age.
For example, statistical process controls can look at the real effects of chamber to chamber, machine and run to run differences that you see, even on the exact same equipment with the same recipe. Process control has become a lot more complicated as critical dimensions have shrunk, along with the margin for error. This means that individual chamber management is becoming fundamental to ensuring high line yield, with tight statistical process control.
This is what EPCO is all about: ensuring the equipment, chambers and the process are optimized together, often using advanced machine learning techniques.
Atonarp has spent a lot of time understanding the FAB and equipment manufacturers’ problems and challenges. The result of those efforts is Aston – our robust molecular sensor.
Aston provides the accurate, actionable, real-time data that’s critical for effective EPCO. This data enables suitable ML models to be built and tested.
In fact, Aston was designed from the ground up to meet the needs of in-situ molecular analysis to enable EPCO.
Learn more by downloading our Application Brief detailing how Aston enables EPCO for semiconductor manufacturing.