應用基因演算法與家族式派工於傳輸整合步進機在小批量情境下之排程問題

Abstract

傳輸整合步進機為半導體工廠中的瓶頸機台，在小批量工件的情境下具有產能損失的問題。加上傳輸整合步進機在加工不同的電路佈局時需要更換光罩，若有過多的光罩更換次數也會導致生產效率不佳。為了解決上述議題，本論文發展一個新的排程方法，結合家族式派工與基因演算法（Family-based Genetic Algorithm ,GA-F），來提高傳輸整合步進機的生產力。此排程方法可以使傳輸整合步進機在具有小批量和光罩設置的情境下提升機台利用率。我們並執行大量實驗比較GA-F與其他兩種的派工法則：GA-FT (traditional family-based genetic algorithm)與GA-I (individual-based genetic algorithm)，實驗結果顯示，GA-F均勝過這兩種派工方法。

An in-line stepper, a bottleneck machine in a semiconductor fab, may have capacity loss in a small-lot scenario. In such a machine, a setup or mask change is required in processing jobs with different circuit layouts. Different job sequences may require different number of setups and result in different productivity for an in-line stepper. This paper developed a scheduling method (called family-based genetic algorithm, GA-F) in order to increase the machine utilization of in-line steppers, in a small-lot scenario which includes setup characteristics. Two other sequencing methods, called GA-FT (traditional family-based genetic algorithm) and GA-I (individual-based genetic algorithm), are taken as benchmarks and compared with the GA-F method Numeric experiments indicate that the GA-F method outperforms these two benchmarks.