Intelligent manufacturing system design and its adaptation to electronic commerce

Abstract

The commercialization and privatization of the Internet and its successive increase in popularity a few years ago are the foundations that began to propel the growth in Electronic Commerce (e-Commerce). As companies migrate toward responsive e-Commerce models, they need an effective manufacturing system that allows them to react rapidly and continuously, innovate ceaselessly, and move fast and quickly adapt to change in order to deal with dynamic change of marketplace and customer needs. This thesis proposes the use of intelligent manufacturing system (IMS) for that objective, and presents an intelligent decision support system (IDSS) based on data mining and agent technologies for an optimal sampling method decision in semiconductor manufacturing process as a pilot study in the IMS design. This thesis also presents its adaptation to the semiconductor e-Commerce model. IDSSs, incorporating intelligent agent and data mining, are designed to aid the decision-making by taking advantages of those technologies. Data mining technology provides an important contribution to the IDSSs by providing techniques which decision support systems (DSSs) are able to utilize in providing a wide range of information available for decision-makers, and intelligent agent technology offers the intelligent, autonomous, active and co-operative nature to the DSSs. This thesis presents an IDSS, called adaptive sampling system (ASS), for the autonomous decision of an optimal sampling method in semiconductor manufacturing process. ASS has an intelligent agent-based novel architecture that takes advantage of the intelligent, autonomous, and active aspects of intelligent agent technology. And also, it has a successful integration of data mining technology for the optimal sampling decision into a DSS framework by means of applying intelligent agent technology. We design the ASS having a methodology how to generate an optimal static sampling method and the associated decision process based on...