工廠生產裝配線的排程、順序及工作下單時機一直是生產管理議題中十分重要的工作,其目的在於適時的配置製造資源,以最有效率的方式來完成產品。而在規模較大的企業裡,製程的改變往往牽動著許多環節與成本,為了減少不必要的錯誤與成本,許多企業紛紛導入了模擬(Simulation)的概念,但因為傳統電腦模擬若要解最佳化問題,必須以窮舉法的方式將問題的所有可行解皆代入模擬系統裡面,非常的費時且不具效率。本研究結合最佳化之演算技術,期望能讓電腦模擬系統可以更快更有效率的搜尋最佳的組合。 本研究利用電腦模擬技術建構一生產裝配線,並分別使用粒子群演算法(Particle swarm optimization, PSO)、突變式粒子群演算法(Particle swarm optimization with mutation based on similarity, PSOMS)及基因演算法(Genetic algorithm)來進行搜尋,希望在固定生產線配置的情境下,藉由模擬最佳化的過程找到每一原物料的最佳進料時間,以求得總完工時間與在製品等待時間最短。最後透過實驗設計與統計檢定法,證實突變式粒子群演算法(PSOMS)明顯優於基因及粒子群演算法。
Assembly line design is an important part of process. Some processes have to change in order to increase the efficiency. Computer simulation has been applied on process design for many decades. Traditionally, simulation has to run all possible alternatives of assembly line. Therefore, simulation is not considered as an optimization technique. Since particle swarm optimization algorithm has been widely used for solving optimization problems in different research areas and gained good performance. This research presents the uses of simulation and improved particle swarm optimization to optimize the management parameters in production. To start with, this study reviews the main parameters to be taken into account for managing an assembly line with bypass workstations. Then, this research creates an assembly line by Flexsim software. And then, using particle swarm optimization (PSO) with mutation based on similarity (PSOMS) and genetic algorithm (GA) to find the optimize solution. The simulation results show that PSOMS is better than other algorithms through experimentation design and statistic test.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。