Exploit the value of production data to discover opportunities for saving power consumption by production tools

This study has demonstrated how to apply datamining technics, Neural Networks (NNs), to estimate the power consumption (kilowatt hour per move, kwh/move) of individual process tool sets in a semiconductor factory, and to analyze the relationships between kwh/move and 19 individual input factors, whi...

Full description

Saved in:
Bibliographic Details
Published in2016 International Symposium on Semiconductor Manufacturing (ISSM) pp. 1 - 4
Main Authors Chih-Min Yu, Chung-Jen Kuo, Ching-Thiam Chung
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2016
Subjects
Artificial neural networks
big data
Data mining
Power demand
power saving
Production
semiconductor
Online AccessGet full text

Cover

More Information
Summary:This study has demonstrated how to apply datamining technics, Neural Networks (NNs), to estimate the power consumption (kilowatt hour per move, kwh/move) of individual process tool sets in a semiconductor factory, and to analyze the relationships between kwh/move and 19 individual input factors, which included "lot size", "process time", "uptime", "usable machine", "Q-time constrain", "sampling rate" and etc.. An empirical study was conducted by using the equipment data of a real fab. The results showed that the proposed approaches can discover rich opportunities, 17.37%, for saving power consumption by production tools.
DOI:10.1109/ISSM.2016.7934536