President Future Foundation North America, Virginia, United States
With the dawn of cloud storage and computing in the last decade, almost every industry has seen an explosion in the volume and quality of inline factory data. This achievement coupled with computing capabilities has fueled the growth of smart task automation, big-data analytics, and machine learning (ML) algorithms, all working in tandem to contribute to the recent growth in artificial intelligence (AI).
This article explores three applications of Artificial intelligence in semiconductor process control that have the potential to improve process-capability beyond existing nodes, increase the overall equipment efficiency (OEE) and reduce the overall line cycle-time.
It will discuss how clustering analysis can aid in developing a comprehensive and complete control path, regression analysis can realize a reduction in prediction errors by auto-tuning model parameters, and finally the use of classification to develop a cascading extrapolation model to improve the accuracy of process-setting prediction in a high part mix, low volume manufacturing environment.
