Optimal maintenance planning and scheduling in semiconductor back-end manufacturing Optimal maintenance planning and scheduling in semiconductor back-end manufacturing 

24 January 2024 | Online | 14:00 | Michael Geurtsen (Eindhoven University of Technology) 

Abstract

Discovering the optimal maintenance planning strategy can have a substantial impact on production efficiency, yet this aspect is often overlooked in favor of production planning. This is a missed opportunity as maintenance and production activities are deeply intertwined. Our study sheds light on the significance of maintenance planning, particularly in the dynamic setting of assembly lines. We study novel problems such as (1) the integrated production, maintenance and resource scheduling on unrelated parallel machines and (2) real-time dispatching of maintenance using deep reinforcement learning techniques considering flexible planning windows, buffer contents, and machine production states. Using real-world data, we demonstrate the immense potential of solving these problems on improving factory throughput. 

Bio
Michael Geurtsen is a PhD student at  Eindhoven University of Technology -  Department of Operations, Planning, Accounting and Control. He received his BSc and MSc Cum Laude in Mechanical Engineering at the Eindhoven University of Technology. Currently, he conducts research on predictive maintenance and the integration of maintenance, production and resource planning in a semiconductor shop-floor, under the supervision of Prof. Dr. Ivo Adan and Dr. Zumbul Atan. His current research interests are in the area of production, maintenance and resource scheduling/planning, where he aims to apply discrete event simulation, digital twins, machine learning and deep reinforcement learning tools.