對製造業而言,要在全球的市場中保持競爭力,生產排程的安排是關鍵因素之一,也就是說,製造業者必須利用有限的資源進行生產活動的安排,以求得最好的生產效率和最大的資源使用率。面板產業是我國最重要的產業之一,卻也面臨競爭激烈的壓力,因此本研究乃以某生產彩色濾光片面板廠為研究個案,希望藉由生產排程績效的改善,提升其市場競爭力。由於個案面板廠中的生產作業有70%以上是屬於黃光部門的,因此本研究就以該黃光部門中流程式的生產型態為研究對象,探討如何進行人力分配與工件生產排序的工作,以達到將總完工時間最小化的目的。 人力分配排程問題與傳統排程問題的差異在於:工件的處理時間不再是固定不變的常數,而是與處理此工件的人工數目有關,亦即工件的處理時間為人工數的某種函數。於本研究中,人力分配排程問題的數學模式將會先被建構出來,並顯示其屬於NP-complete的範疇。對於NP-complete的問題,實務上,通常會以啟發法於短時間內,有效的求得最佳解或近似最佳解的解答,因此本研究乃以基因演算法為求解的方法。 為了評估基因演算法的績效,基因演算法、修正NEH法與工廠現行所使用的方法分別被應用於九種不同型態問題的解決上,以探討它們在總完工時間方面的表現。經過比較,發現基因演算法能於最短的時間內得到最好的結果,也就是說,本研究所提出的基因演算模式對此個案工廠的黃光部門來說,確實能提供一個最有效率、效果也最好的生產排程方式。
For the manufacturing industries, in order to stay competitive in the global market, scheduling is one of the key factors that needs to be dealt with. In other words, manufacturing industries must utilize the limited resources to arrange production activities to obtain the best production efficiency and resources utilization. Color filter industry is one of the most important industries in our country and it is under highly competitive pressure all the time. Therefore, this study take a color filter plant as the studying subject and expect to increase its competitiveness through the improvement of scheduling performance. Furthermore, since more than 70% of the production activities are belonged to the Yellow Light department of the color filter plant, the production mode of this specific department is investigated for its worker assignment scheduling problem to minimize the performance measure of makespan. The difference between the worker assignment scheduling problem and the classic scheduling problem is that the job processing times is no longer a constant but a function of the number of workers assigned to work for the job. In this study, the mathematic model of this specific problem will first be constructed and shown to be an NP-complete problem. For the NP-complete problem, practically, heuristics will be used to get the optimal or near optimal solution in a shorter time effectively. In this sense, this study uses genetic algorithm as the heuristic for this specific problem. For the evaluation of the performance of genetic algorithm heuristic, genetic algorithm, modified NEH, and method used by the Yellow Light department are all applied to nine different types of problems to explore their performance regard to the makespan. After the comparison, it is shown that the genetic algorithm presented in this study has the best performance both in the time consuming and makespan minimization categories. In conclusion, for the Yellow Light department of the color filter plant, the genetic algorithm suggested in this study provides an efficient and effective way for solving the worker assignment scheduling problem of that department.
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。