Application of Data-Based Parameter Optimizing Approaches in IC Manufacturing System Simulation Model

When building an Integrated circuit (IC) manufacturing system simulation model, the practices normally simplify some complex features to ensure build ability and run ability. However, this brings the consequence of model imprecision; parameters disable to represent actual situation, etc. This paper proposes parameter optimizing approaches in IC manufacturing system simulation model based on the offline weighted average method and Kolmogorov–Smirnov (K–S) test method. They are on the basis of the processing time’s feature, for the parameter reflects the overall character of the sample’s former performance and at the same time, the system has a great amount of historical data. The approaches are to optimize the whole model by optimizing the significant processing time parameter. It turns out that both the weighted average method and the K–S test based method can improve the precision of model.