Data-driven aggregate modeling of a semiconductor wafer fab to predict WIP levels and cycle time distributions

Abstract In complex manufacturing systems, such as a semiconductor wafer fabrication facility (wafer fab), it is important to accurately predict cycle times and work-in-progress (WIP) levels. These key performance indicators are commonly predicted using detailed simulation models; however, the models computationally expensive have high development maintenance costs. this paper, we propose an aggregate modeling approach, where each work area, i.e., group of functionally similar workstations, in fab aggregated into single-server queueing system. The parameters system can be derived directly from arrival departure data that area. To obtain fab-level predictions, our proposed methodology builds network models, represents entire consisting different areas. viability method practice demonstrated by applying real-world fab. Experiments show model make accurate but also provide insights limitations modeling.