Today, the semiconductor industry must be able to produce Integrated Circuit (IC) withreduced cycle time, improved yield and enhanced equipment effectiveness. Besides thesechallenges IC manufacturers are required to address the products scrap and equipment driftsin a complex and uncertain environment which otherwise shall severely hamper the maximumproduction capacity planned. The objective of this thesis is to propose a generic methodologyto develop a model to predict the Equipment Health Factor (EHF) which will define decisionsupport strategies on maintenance tasks to increase the semiconductor industry performance.So, we are interested here to the problem of equipment failures and drift. We propose apredictive approach based on Bayesian technique allowing intervene early to maintain, forexample, the equipment before its drift. The study presented in this thesis is supported by theIMPROVE European project
