Data mining for yield enhancement in semiconductor manufacturing and an empirical study
Expert Systems with Applications: An International Journal
Proceedings of the 40th Conference on Winter Simulation
Student Solutions Manual for Ramsey/Schafer's The Statistical Sleuth: A Course in Methods of Data Analysis, 3rd
Hi-index | 0.00 |
A Bayesian Ranking Scheme is proposed for the reliable diagnosis of various yield-loss factors induced in semiconductor manufacturing. The aim is to cope with three problems: (FICV) false identification due to confounding variables, (FISV) false identification due to suppressor variables, and (MISC) miss identification due to severe multicollinearity. The proposed scheme reuses both the results from knowledge-based and data-driven inference tools as input data, where the former resembles expert's knowledge on diagnosing pre-known yield-loss factors while the latter serves for exploring new yield-loss factors. Two successive stages with specific designs for yield diagnosis services are addressed: Bayesian Variable Selection for reliable model construction and Relative Importance Assessment for facilitating interpretations on model parameters. A simulation example is designed to demonstrate the usefulness of Bayesian Ranking Scheme on solving FICV, FISV and MISC problems.