Neurofuzzy modeling of chemical vapor deposition processes

The modeling of semiconductor manufacturing processes has been the subject of intensive research efforts for years. Physical-based (first-principle) m odels have been shown to be difficult to develop for processes such as plas ma etching and plasma deposition, which exhibit highly nonlinear and comple x multidimensional relationships between input and output process variables , As a result, many researchers have turned to empirical techniques to mode l many semiconductor processes. This paper presents a neurofuzzy approach a s a general tool for modeling chemical vapor deposition (CVD) processes. A live-layer feedforward neural network is proposed to model the input-output relationships of a plasma-enhanced CVD deposition of a SEN film. The propo sed five-layer network is constructed from a set of input-output training d ata using unsupervised and supervised neural learning techniques. Product s pace data clustering is used to perform the partitioning of the input and o utput spaces, Fuzzy logic rules that describe the input-output relationship s are then determined using competitive learning algorithms. Finally, the f uzzy membership functions of the input and output variables are optimally a djusted using the backpropagation learning algorithm, A salient feature of the proposed neurofuzzy network is that after the training process, the int ernal units are transparent to the user, and the input-output relationship of the CVD process can be described linguistically in terms of IF-THEN fuzz y rules. Computer simulations are conducted to verify the validity and the performance of the proposed neurofuzzy network for modeling CVD processes.