State monitoring and fault diagnosis method for large-scale semiconductor manufacture process

Abstract

The invention relates to a state monitoring and fault diagnosis method for a large-scale semiconductor manufacture process. The method comprises the following operation steps of: (1) unpacking of multiple batches of process data: unpacking a three-dimensional data set (namely variable number, sampling time and production batch) into a two-dimensional data set; (2) elimination of redundant information in the data set; (3) characteristic extraction of the data set; (4) state modeling of a manufacture process based on a Gaussian mixture model; (5) state monitoring of the manufacture process; and(6) fault diagnosis of the manufacture process. According to the method disclosed by the invention, the state monitoring and fault diagnosis of the large-scale semiconductor manufacture process can be achieved, thereby the operation reliability of a semiconductor manufacture system and the quality of an output wafer are improved, and further the production and operation cost is reduced.

Description

technical field [0001] The present invention is a state monitoring and fault diagnosis method for a large-scale semiconductor manufacturing process, involving multi-sensing signal data conversion, large-scale process data set dimension reduction and important feature extraction, data probability density distribution space description modeling, process state monitoring And fault diagnosis, to realize the status monitoring and fault diagnosis of complex large-scale semiconductor manufacturing process. The invention belongs to the technical field of state monitoring and fault diagnosis of manufacturing process. Background technique [0002] At present, the semiconductor manufacturing system presents a development trend of high automation, high precision, high reliability, and high intelligence, emphasizing the controllability, reliability, and maintainability of the manufacturing process. The semiconductor manufacturing process is quite complicated. The manufacturing process n...

AI Technical Summary

Problems solved by technology

This method has the following deficiencies: [1] There is no effective elimination of redundant information in the process variable data set due to the large-scale process data set caused by the collected batch-manufacturing multi-sensing signals; [2] The feature extraction stage uses the principal element The analysis method can only extract the characteristics of the global variance information in the data set, but cannot effectively extract the local variance information, and may lose important information; [3] The spatial distribution of variable data in the semiconductor manufacturing process is usually nonlinear and multimodal , therefore, the SPE statistical index proposed based on the premise that the given data set should conform to the Gaussian distribution is usually unable to effectively detect nonlinear and multi-modal manufacturing process anomalies; [4] this method only monitors semiconductor manufacturing process anomalies , without identifying the cause of the process abnormality, that is, the failure diagnosis of the manufacturing process cannot be carried out

Images