Purpose The purpose of this study is to propose an innovative methodology in solving the lean production design from semiconductor crystal-ingot pulling manufacturing which is an important industry. Due to the complexity of the system, it is computationally prohibited by an analytical approach; thus, simulation optimization is adopted for this study. Design/methodology/approach Four control factors that affect the system’s performance, including the pulling strategy, machine limitations, dispatching rules and batch-size control, are identified to generate the future-state value stream mapping. Taguchi two-step procedure and simulation optimization are used to determine the optimal parameter values for a robust system. Findings The proposed methodology improved the system performances by 6.42 and 12.02 per cent for service level and throughput, respectively. Research limitations/implications This study does not investigate operations management issues such as setup reduction, demand forecasting and layout design. Practical implications A real-world crystal-ingot pulling manufacturing factory was used for the case study. The results are promising and are readily applied to other industrial applications. Social implications The improved performances, service level and throughout rate, can result in an improved customer satisfaction level and a reduced resources consumption, respectively. Originality/value The proposed methodology innovatively solved a practical application and the results are promising.