A General Framework for State Estimation in High-Mix Semiconductor Manufacturing

Abstract

In this work, the characteristics inherent in non- threaded state estimation problems, i.e., state estimation without segregating the process data into different bins, are analyzed for high-mix semiconductor manufacturing processes. A general framework is introduced for the non-threaded state estimation methods. The framework is based on the best linear unbiased estimate (BLUE) of a Gauss-Markov model, and non-threaded state estimation methods based on least squares, the Kalman filter and recursive least squares (RLS) are analyzed using the general framework. The three methods are compared analytically and by using a simulation example. Bayesian-enhanced adaptive versions for the Kalman filter-based and RLS-based methods are proposed and several examples demonstrate the effectiveness of the proposed adaptive methods.