Abstract For the past few years, electronics manufacturing may have been the victim of its own success. Unlike in heavier industries, rationalization is a concept that was unknown in the sector until only a few years ago and even now, many companies are struggling with cost-cutting measures. Production systems in electronics manufacturing need to be highly flexible because of a varying and evolving environment. Therefore real-time process control and, possibly as a result, production optimization are extremely challenging areas. Traditional approaches often do not work due to a lack of robustness or reliability.

For this reason, a new generation of decision support systems is needed in response to some specific problems. The thesis addresses topics such as design of intelligent interfaces for knowledge acquisition and elicitation, use of that knowledge for improved data analysis and diagnostics, real-time feedback control, self-tuning capabilities, and evaluation of optimization methods in discrete processes. Topics covered therefore include the whole scope of a decision support system, from its design through to the evaluation of its performance as well as interaction capabilities as a vehicle for sharing information.

The aim of this research is to streamline the development of a new generation of decision support systems by providing tools and methods for a better integration of knowledge in an evolving environment. The main interest lies not only in improved data analysis, but also in better formalization and use of diagnosis. Case studies presented in this thesis demonstrate the practical feasibility of such an approach.