Summary: | 碩士 === 國立交通大學 === 工業工程與管理系 === 88 === Multiple-priority orders and fluctuating product mix owing to product diversification and various customer requests have been critical issues in wafer fabrication. However, relevant literature studying about production planning of wafer fabrication largely assumes single-priority orders and constant product mix to simplify the production environment. Therefore, this work presents a novel hierarchical and mathematical production planning model capable of considering these important production characteristics without the assistance of simulation tool.
The model proposed contains four modules. Initially, the production performance estimation module estimates performance indexes such as production cycle time, work-in-process, daily moves, and the threshold value of two-boundry (TB). Secondly, the master production schedule (MPS) module evaluates the capacity load and batch rule, and in sequence schedules the release cycle list, release time list, output time list, and assigns due date of each orders. Thirdly, the detail schedule (DS) module schedules the layer output time list, and designs the release mechanic based on the batch-CONWIP release rule. Furthermore, for the transient stage due to the fluctuating product mix, the transient stage adjustment module clarifies the adjustment of relevant planning, especially on cycle time estimation and due date assignment for cross-stage orders.
Simulation experiment reveals that the proposed model can efficiently estimates the production performance and designs relevant plans during MPS and DS stage. Furthermore, both the production target and due date requirement can be achieved. The planning result can output to the target planning system or the production activity control system, thus providing a valuable reference when making a shop floor decision. Consequently, this model capable of processing multiple-priority orders and fluctuating product mix is effective and efficient, and applicable to a more complex and real environment.
|