在自動化物料搬運系統(automated material handling system; AMHS)中,晶舟車的數量顯著影響半導體廠的設備成本和生產效率。然而,製造環境的複雜度和不確定性使得決定晶舟車數量是項挑戰。本研究提出新的模型以找出 AMHS 在購置成本最小且同時滿足運輸需求時的最佳車數分配,並提出以模擬最佳化為基礎稱作「模擬序列超模型(simulation sequential metamodeling;SSM)」的方法來求解模型。數值分析證明了SSM優於傳統的方法,尤其是在限制式數較多或模型變異較大的問題上。最後以一實證研究驗證本方法在實務上的可行性。
The number of vehicles (vehicle fleet size) in automated material handling system (AMHS) significantly influences cost of equipments and efficiency of production. However, it is not an easy task to decide the number of vehicles required in the system due to the complexity and uncertainty of manufacturing system. In this research, we propose a new formulation to determine the optimal number of vehicles of the AMHS so as to achieve the minimal vehicle cost, while satisfying transportation requirement. Furthermore, we develop a simulation-based methodology, called simulation sequential metamodeling, to solve the problem. Numerical experiments show that the proposed method outperforms the traditional approaches, especially in the problems with larger number of constraints or variance level. An empirical study is conducted to verify the viability of the proposed method.