Adaptive virtual metrology method based on Just-in-time reference and particle filter for semiconductor manufacturing. (15th January 2021)

• Record Type: Journal Article
• Authors: Cai, Haoshu; Feng, Jianshe; Zhu, Feng; Yang, Qibo; Li, Xiang; Lee, Jay
• Is Part Of: Measurement. Volume 168(2021)
• Publication Date: 2021-01-15
• Subjects: Just-in-time Model -- Virtual Metrology -- Particle Filter -- Materials Removal Rate -- Chemical Mechanical Planarization -- Semiconductor Manufacturing; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2020.108338
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlight: A dynamic model is proposed to predict material removal rate. The model employs a just-in-time reference strategy. The model uses particle filter to track the process change. The result is validated on a public semiconductor dataset. Abstract: The prediction of the average Material Removal Rate (MRR) for Chemical Mechanical Planarization (CMP) process has been recognized to be a critical factor of Virtual Metrology (VM) modeling as well as wafer-to-wafer process control. In this paper, a novel method is proposed to dynamically predict MRR in CMP process by using a Just-in-time (JIT) model-based strategy. First, the proposed method applies a reference search procedure to query similar data samples from the historical dataset. Second, a Support Vector Regression (SVR) model is used to fuse the similar data samples with the past MRR. Finally, Particle Filter (PF) is employed to estimate and update the fusion result. The application of PF ensures that the method can track the CMP process change and continuously learn from the current dataset. Compared with static models, the proposed method, named JIT-PF model is an online dynamic method with enhanced accuracy. Compared with pure JIT model, JIT-PF model keeps the historical knowledge in the process. A public dataset is used to validate JIT-PF model with the benchmarks from the recent literature. The model outperforms other peer dynamic models and some static models from the literature.