Ray Use Cases
Semiconductor Fab Production Scheduling Using Deep Reinforcement Learning
September 19, 4:00 PM - 4:30 PMSemiconductor manufacturing is often regarded as the most complex manufacturing process in the world, producing chips in high volume at the nanometer scale. Our work focuses on a tractable method for optimizing scheduling in a dynamic fab. Changing priorities in the fab on a day-to-day basis requires flexible optimization methods to respond to these changes.
With our solution fab operators can augment their current workflow to quickly and efficiently create optimized schedules. These schedules are based on a user-defined set of KPIs, e.g. related to cycle times and on-time delivery. Our solution minds.ai Maestro, built on top of RLLib, uses Deep Reinforcement Learning to efficiently interpret and scale to the complex dynamics in the fab in order to arrive at highly optimized schedules, and subsequently bring these into production.
minds.ai Maestro runs natively on Linux, macOS and Windows, showing that Windows-only enterprise software can leverage Ray.
About Jasper
Jasper van Heugten is the Director of Advanced Technology at minds.ai. He leads the research and development of the minds.ai Maestro product, an optimization suite for Semiconductor Manufacturing leveraging state-of-the-art Deep Learning (AI) methods, including Deep Reinforcement Learning. He has experience in C&SVP-level tech consulting and scaling AI from early idea to production at Fortune 100 companies across multiple industries, such as Semiconductor, Pharma, Automotive, Renewables, and Big Tech. He holds a PhD in Theoretical Physics from Utrecht University, the Netherlands.
Jasper van Heugten
CTO, minds.ai
Ready to Register?
Come connect with the global community of thinkers and disruptors who are building and deploying the next generation of AI and ML applications.
