Multi-criteria Dynamic Scheduling Methodology for Controlling a Semiconductor Wafer Fabrication System

Scheduling semiconductor wafer fabrication system (SWFS) is a very difficult problem. The assignment of wafer lots to machines is a complex problem that is often solved in real-time with simple dispatching rules. This paper proposes a multi-criteria dynamic scheduling methodology based on fuzzy logic and design of experiment (DOE). We adopt a fuzzy multi-criteria decision strategy to simultaneously take into account multiple aspects in every dispatching decision. Since most global dispatching rules only consider the static bottleneck to keep line balance, we also incorporate in the multi-criteria algorithm a specific heuristic rule that takes into account the dynamic bottleneck. Moreover, we also adopt DOE and multiple response optimization method to tune the weights associated to each decision criteria in the decision algorithm. A mini-fab model developed by Intel in collaboration with ASU is utilized to illustrate the proposed approach.