Abstract:The virtual metrology model using tree-based ensemble models in semiconductor wafer physical vapor deposition (PVD) is verified with high predict accuracy and good performance.The virtual metrology model using tree-based ensemble model in semiconductor PVD process is significantly better than LASSO regression, partial least squares regression, support vector machine, Gaussian process regression and artificial neural network regression.The performance of virtual metrology model based on SMBO with surrogate model can find better hyperparameters faster and obtain higher virtual metrology accuracy comparing to the commonly used random search hyperparameter optimization method.In order to improve the accuracy of semiconductor wafer virtual metrology, and overcome the physical metrology delay of wafer acceptance test, a virtual physical vapor deposition metrology method based on combination of tree-based ensemble models is proposed to conduct online virtual metrology on semiconductor wafer electrical parameters, and use hyperparameter optimization technique to perform model optimization and to achieve real-time alarm on process deviation. This combination of tree-based ensemble model combines Bagging, Boosting, and Stacking techniques. First, based on 4 types of base learner, Random Forest, Extra-Trees, XGBoost, and lightGBM, preliminary virtual metrology is performed on wafer PVD process, and then transforms the predict results of the 4 base learners into meta feature vector as the input of meta learner lightGBM to perform further virtual metrology. The Sequential model-based optimization algorithm is used to improve the accuracy of virtual metrology. First, the initial hyperparameter of the sequential model-based optimization is initialized by using random sampling, then the combination model is approximated by the surrogate model of tree-structured Parzen estimator, and the recommended hyperparameters is obtained by using EI (Expected Improvement), and then the optimized combination model is obtained. Finally, the superiority of the method proposed in this paper is verified by studying the results comparing to the common virtual metrology methods on the PVD process. The experiment shows the result of resistivity metrology using the combination of tree-based ensemble models in the PVD process is significantly better than LASSO regression, partial least squares regression(PLSR), support vector machine(SVR), Gaussian process regression(GPR) and artificial neural network regression(ANN).

Machine Learning based CVD Virtual Metrology in Mass Produced Semiconductor Process

Development of Convolutional Neural Network Based Gaussian Process Regression to Construct a Novel Probabilistic Virtual Metrology in Multi-Stage Semiconductor Processes

Just-In-Time Modeling with Variable Shrinkage Based on Gaussian Processes for Semiconductor Manufacturing

Convolutional Neural Networks for Multi-Stage Semiconductor Processes

Recurrent feature-incorporated convolutional neural network for virtual metrology of the chemical mechanical planarization process

Application of Gaussian Processes with Variable Shrinkage Method and Just-in-time Modeling in the Semiconductor Industry

Virtual Metrology for Semiconductor Chemical Mechanical Planarization Process Using Wide & Deep Learning.

Virtual Metrology Based on Graph Convolutional Neural Network for Semiconductor PVD Process

Spatial Batch Optimal Design Based on Self-Learning Gaussian Process Models for LPCVD Processes

Phase Partition Based Virtual Metrology for Material Removal Rate Prediction in Chemical Mechanical Planarization Process

Introducing machine learning-based application for writer main pole CD metrology by dual beam FIB/SEM

Virtual metrology of semiconductor PVD process based on combination of tree-based ensemble model

Virtual metrology in semiconductor manufacturing: Current status and future prospects

Deep Learning Regression of VLSI Plasma Etch Metrology

Virtual metrology modeling of reactive ion etching based on statistics-based and dynamics-inspired spectral features

Reference-based Virtual Metrology method with uncertainty evaluation for Material Removal Rate prediction based on Gaussian Process Regression

Convolutional Neural Networks for Automatic Virtual Metrology

Quantifying the CVD-grown Two-Dimensional Materials Via Image Clustering.

Statistical Feature Extraction and Hybrid Feature Selection for Material Removal Rate Prediction in Chemical Mechanical Planarization Process

Applying Machine Learning Models on Metrology Data for Predicting Device Electrical Performance

Machine Learning Applied to Electron Beam Lithography to Accelerate Process Optimization of a Contact Hole Layer