Virtual metrology on Chemical Mechanical Planarization process based on Just-In-Time Learning

This paper deals with missing data in semiconductor manufacturing derived from measurement sampling strategies. The idea is to construct a virtual metrology module to estimate non measured variables using a new modified Just-In-Time Learning approach (JITL). The classical method has been proposed and applied to chemical process. This latter presents some drawbacks and our main contribution is to improve the existing algorithm version of the JITL approach. In collaboration with our industrial partner STMicroelectronics in Rousset, the efficiency of the proposed method is illustrated by using industrial data-sets derived from Chemical Mechanical Planarization (CMP) process that enable to compare results obtained with the classical and the modified version.