Effective monitoring and process control in semi-conductor manufacturing using feature selection

In semi conductor manufacturing the wafer fabrication process is under constant surveillance via the stringent monitoring of measurements and signals collected from metrology steps and machine sensors. However, not all of this data is equally valuable within this process control domain. Engineers typically have a much larger number of signals than are actually required and can be feasibly investigated. Process control data contains a combination of useful information, irrelevant information as well as noise. It is often the case that useful information is buried in the latter two. Current business improvement methodologies are not always appropriate to address this issue. If we consider each type of measurement or signal as a feature, then feature selection may be used to identify the most predictive features. Once these features have been identified causal relevance may then be considered within the scope of larger business improvement projects such as Six Sigma. Process engineers may then apply this new learning to ensure a small scrap rate further downstream in the process, increase the throughput and reduce the per unit production costs. Working in partnership with industry this research aims to address this complex problem as part of their process control engineering in the context of wafer fabrication production and enhance current business improvement techniques with the application of feature selection as an intelligent systems technique