Intelligent Scheduling Methods for Challenges of Cluster Tools with Concurrent Processing of Multiple Wafer Types*

摘要

With the trends of larger wafer diameters and smaller lot sizes, cluster tools are forced to work during long transient processes. In order to improve productivity, there has growing concern about the scheduling problem of cluster tools with concurrent processing of multiple wafer types. Since there is no intermediate buffer between process modules, the scheduling of cluster tools is more complicated than the traditional mixed assembly line. This paper reviews the modeling and scheduling of cluster tools with wafer residency time constraints, activity time variation, process module failure and multi-cluster tools. Then, in order to find the off-line and dynamic method to solve the scheduling problem of concurrent processing of multiple wafer types, intelligent search algorithm and reinforcement learning algorithm is the future research directions, respectively.