Deterministic and Stochastic Data-driven Approaches for Semiconductor Scheduling with Critical Queue Time Constraints

• Applicant: Professor Dr. Lars Mönch
• Subject Area: Operations Management and Computer Science for Business Administration ; Data Management, Data-Intensive Systems, Computer Science Methods in Business Informatics
• Term: since 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 529014501

In this collaborative research effort, the applicants from Taiwan and Germany are interested in designing complementary solution methods for large-sized complex job shop scheduling problems with critical queue time constraints. Such time constraints, also known as maximum time-lags in deterministic machine scheduling research, are crucial in industrial domains such as semiconductor manufacturing or steel processing. In semiconductor manufacturing, critical queue time constraints must be respected to prevent native oxidation and contamination effects on the wafer surface. A wafer is a thin disc made from silicon on which integrated circuits are built layer by layer. Oxidation and contamination effects result in scrapped wafers. Taking into account time constraints leads to challenging coordination problems on the shop floor which are not well understood so far. By executing the resulting scheduling decisions in a dynamic and stochastic environment provided by discrete-event simulation models of large-scaled modern wafer fabs, the applicants are interested in identifying operational conditions under which each of the complementary approaches work well. Moreover, they strive for designing and testing hybrid approaches that combine the benefits of deterministic and stochastic approaches. This joint research project combines the complementary expertise of the research teams from the National Taiwan University and the University of Hagen. On the German side, Prof. Lars Mönch will focus on the design of deterministic scheduling algorithms and the construction of a simulation infrastructure that allows for executing schedules in a dynamic and uncertain environment using a rolling horizon approach. On the Taiwanese side, Prof. Cheng-Hung Wu will mainly focus on stochastic optimization algorithms directly taking into account process uncertainty. Moreover, he is interested in designing hybrid algorithms that combine the advantages of the deterministic and stochastic approaches.

• DFG Programme: Research Grants
• International Connection: Taiwan
• Partner Organisation: National Science and Technology Council (NSTC)
• Cooperation Partner: Professor Dr. Cheng-Hung Wu, Ph.D.