Defective Wafer Detection Using Sensed Numerical Features

NCU Institutional Repository > 資訊電機學院 > 資訊工程學系碩士在職專班 > 博碩士論文 > Item 987654321/83938

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/83938

题名:	Defective Wafer Detection Using Sensed Numerical Features
作者:	林靖宇;Lin, Ching-Yu
贡献者:	資訊工程學系在職專班
关键词:	GRU神經網絡;XGBoost;深度學習;缺陷晶圓檢測;GRU neural networks;XGBoost;Deep learning;Defective wafer detection
日期:	2020-07-30
上传时间:	2020-09-02 17:43:21 (UTC+8)
出版者:	國立中央大學
摘要:	半導體製造的基本過程之一是切片，這意味著將晶棒切成許多晶片。在切片過程中，可能會產生有缺陷的晶圓。不幸的是，識別缺陷晶片的檢查既浪費時間又檢查困難。為了解決這個問題，我們建立了一個系統，該系統在切片及表面檢查過程中會使用傳感器收集晶圓特性（例如溫度，厚度，晶片表面上的圖案等）以檢測晶片在製造過程中是否有缺陷。我們在此系統中應用兩種不同模型-GRU 神經網絡和XGBoost。此兩種模型經過微調後，根據實際數據分析的實驗結果顯示在晶圓的缺陷檢測方面， GRU 神經網絡的預測準確度和模型訓練時間均優於XGBoost;One of the fundamental processes in semiconductor manufacturing is slicing, which means cutting an ingot into many wafers. During the slicing process, it is possible to produce defective wafers. Unfortunately, the inspection to identify defective wafers is time-consuming and di cult. To solve this problem, we build a system, which uses sensors to collect features (e.g., temperature, thickness, pattern on wafer surface, etc.) during the slicing process to detect if the wafers are defective in the manufacturing process. Two di erent models, the GRU neural network and XGBoost, are implemented in the proposed system. After ne-tuning both models, experimental results based on real dataset indicate that the GRU neural network outperforms XGBoost for wafer defective detection in both the prediction accuracy and model training time. ii
显示于类别:	[資訊工程學系碩士在職專班 ] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	195	检视/开启

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....