Synchronized production planning and scheduling in semiconductor fabrication

• We propose an iterative model for synchronized production planning and scheduling.
• The iterative algorithm can be applied to a flexible manufacturing system.
• The model makes good production planning with a minimal gap for scheduling.
• The model converges in the iterative process to a better solution.
• The model outperforms the existing approaches in terms of feasibility and efficiency.

Semiconductor fabrication line runs with several hundreds of steps on several hundreds of equipment in the type of re-entrant flow process. The hierarchical planning decisions are made in a way that the production planning are determined on the integrated models at first, and the scheduling operations are then performed. The most complexity comes from the difficulties of synchronization of the production planning and scheduling. An iterative approach to achieve the synchronization is suggested to coordinate the input and output quantity of the production plan when generating a schedule. In the proposed algorithm, not only the input and output quantity of the production plan but also the production quantity is utilized as a coordination factor in the scheduling. Furthermore, the manufacturing lead time, the number of setup events, and the available work-in-process (WIP) level are updated through an iterative procedure of simulation and optimization for the synchronization of planning and scheduling decision. Computational experiments show that interaction framework yields good performance in terms of feasibility, demand satisfaction, and the manufacturing lead time. It is shown that a production planning model can be generated with practical parameters and can be implemented on the scheduling and dispatching levels.