A hybrid and intelligent system for predicting lot output time in a semiconductor fabrication factory

Authors:
Toly Chen;Yu-Cheng Lin
Affiliations:
Department of Industrial Engineering and Systems Management, Feng Chia University, Seatwen, Taichung City, Taiwan;Department of Industrial Engineering and Management, Overseas Chinese Institute of Technology, Taichung City, Taiwan
Venue:
RSCTC'06 Proceedings of the 5th international conference on Rough Sets and Current Trends in Computing
Year:
2006

Citing 2
Cited 4

Genetic Algorithms in Search, Optimization and Machine Learning

Genetic Algorithms in Search, Optimization and Machine Learning
A Case-Based Reasoning Approach for Due-Date Assignment in a Wafer Fabrication Factory

ICCBR '01 Proceedings of the 4th International Conference on Case-Based Reasoning: Case-Based Reasoning Research and Development

Incorporating fuzzy c-means and a back-propagation network ensemble to job completion time prediction in a semiconductor fabrication factory

Fuzzy Sets and Systems
Fuzzy-neural approaches with example post-classification for estimating job cycle time in a wafer fab

Applied Soft Computing
A prediction interval-based approach to determine optimal structures of neural network metamodels

Expert Systems with Applications: An International Journal
Constructing prediction intervals for neural network metamodels of complex systems

IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Predicting the output time of every lot in a semiconductor fabrication factory (wafer fab) is a critical task to the wafer fab. To further enhance the effectiveness of wafer lot output time prediction, a hybrid and intelligent system is constructed in this study. The system is composed of two major parts (a k-means classifier and a back-propagation-network regression) and has three intelligent features: incorporating the future release plan of the fab (look-ahead), example classification, and artificial neural networking. Production simulation is also applied in this study to generate test examples. According to experimental results, the prediction accuracy of the hybrid and intelligent system was significantly better than those of four existing approaches: BPN, case-based reasoning (CBR), FBPN, kM-BPN, by achieving a 9%~44% (and an average of 25%) reduction in the root-mean-squared-error (RMSE) over the comparison basis – BPN.