Semiconductor fabrication process multi-performance prediction method

Abstract

The invention relates to a semiconductor fabrication process multi-performance prediction method, comprising the steps of selecting an articles being processed level parameter, an equipment parameter and a workpiece parameter which represent the state of a semiconductor production line as influence factors of performance indexes; collecting relevant data of the production line, preprocessing by using a principal component analysis method, removing redundant information, constructing a multi-performance prediction model by using a Bayes neural network, and controlling the complexity of the network model by introducing a Bayes method; analyzing whether the model performance conforms to a performance evaluation criteria by a model precision proof method, and performing online correction on the network model structures which do not conform to the standard; finally determining the key factors influencing the average workpiece processing period and the equipment utilization rate. According to the semiconductor fabrication process multi-performance prediction method, the defects that the performance prediction model in the semiconductor field is limited by constraint conditions, the generalization performance is very poor and the like are improved, the problem that the single performance prediction model in the semiconductor field is not applicable to multi-performance prediction modeling is solved, and the method is an improvement of the semiconductor fabrication process multi-performance prediction method.

Description

technical field [0001] The invention belongs to the field of advanced manufacturing technology, and relates to a multi-performance prediction method of a semiconductor manufacturing process using a Bayesian neural network. Background technique [0002] As one of the pillar industries of the national economy, the semiconductor manufacturing industry has important strategic significance for my country's economic development. How to improve the performance and production efficiency of the semiconductor manufacturing system is the focus of the semiconductor manufacturing industry. The semiconductor manufacturing system is one of the most complex manufacturing systems at present. It has the characteristics of high uncertainty, multiple entry, and multiple objectives. To improve its performance and increase production efficiency, it is necessary to study how to quickly obtain the optimal dynamic scheduling scheme. This is a major problem in the field of semiconductor optimization...

AI Technical Summary

Problems solved by technology

[0004] In 2011 IEEE Transactions on Semiconductor Manufacturing, Yair Meidan et al. used a comprehensive filter and wrapper method to identify and predict the key influencing factors of the workpiece processing cycle. However, if this method is used to predict and model multiple performance indicators, there is a predictive model structure The problem is too complicated and difficult to control, and the correlation between multiple performances is not considered; the patent application number 201310239501.0 discloses a performance prediction method for dynamic scheduling of semiconductor production lines, using extreme learning machines for predictive modeling, which can be Dynamic real-time scheduling provides the basis, but this method has not yet analyzed the key influencing factors of the predicted performance, so there may be a problem that due to too many adjustable parameters, the optimal scheduling solution cannot be obtained accurately and in time

Images