Using data mining technology to design an intelligent quality analysis control system for semiconductor packaging industry

  • Authors:
  • Ruey-Shun Chen;Ruey-Chyi Wu

  • Affiliations:
  • Institute of Information Management, National Chiao Tung University, Hsinchu City, Taiwan;Institute of Information Management, National Chiao Tung University, Hsinchu City, Taiwan

  • Venue:
  • ICCOMP'06 Proceedings of the 10th WSEAS international conference on Computers
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

The aims of this paper is to depict an intelligent quality analysis control system involved in using data warehouse, data mining, decision tree, and Bayesian classification analysis to discover the main inconsistency reasons in the manufacturing process of semiconductor packaging plants and compare the correctness of classification analysis of the two methods, so as to set up an intelligent quality analysis control system providing an efficiency tool for analyzing problems, with a view to identifying the causes of problems, making decision immediately, and eventually reducing the cycle time taken to solve quality-related problems. The contributions of this research are illustrated as follows. Predictions made by the target group by means of decision tree analysis are more accurate than those made by Bayesian classification, indicating that decision tree analysis is an effective mean of classification analysis in semiconductor packaging quality problems, whereas evaluation of feasible methods by data warehouse, and data mining followed by establishment of the basis for a quality analysis system environment, that is characteristic of knowledge sharing may be applied to analysis of to the quality problems in all corporation.