Accuracy and real-time considerations for implementing various virtual metrology algorithms

Yu Chuan Su, Tung Ho Lin, Fan Tien Cheng, Wei Ming Wu

Research output: Contribution to journalArticlepeer-review

50 Citations (Scopus)


In the semiconductor industry, run-to-run (R2R) control is an important technique to improve process capability and further enhance the production yield. As the dimension of electronic devices shrink increasingly, wafer-to-wafer (W2W) advanced process control (APC) becomes essential for the critical stages of production processes. W2W APC requires the metrology values of each wafer; however, it will be highly time and cost consuming to obtain actual metrology values from each wafer by physical measurement. Recently, an efficient and cost-effective approach denoted "virtual metrology (VM)" was proposed to substitute the actual metrology. To implement VM in W2W APC, both conjecture-accuracy and real-time requirements need to be considered. In this paper, various VM algorithms, including back-propagation neural networks (BPNN), simple recurrent neural networks (SRNN), and multiple regression (MR), are evaluated to see whether they can meet the accuracy and real-time requirements of W2W APC or not. The fifth generation TFT-LCD chemical-vapor deposition process is used to test and verify the requirements. Test results show that both one-hidden-layered BPNN and SRNN VM algorithms achieve acceptable conjecture accuracy and meet the real-time requirements of semiconductor and TFT-LCD W2W APC applications.

Original languageEnglish
Article number4589025
Pages (from-to)426-434
Number of pages9
JournalIEEE Transactions on Semiconductor Manufacturing
Issue number3
Publication statusPublished - 2008 Aug

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Industrial and Manufacturing Engineering
  • Electrical and Electronic Engineering


Dive into the research topics of 'Accuracy and real-time considerations for implementing various virtual metrology algorithms'. Together they form a unique fingerprint.

Cite this