數位化是近年來製造效率和品質提升的主要推動力，結合具有能連接實體和運算能力的虛實整合系統(Cyber Physical System, CPS)，使製造業能面對智慧製造全球化的趨勢。為使製造業者在全球製造環境中取得競爭力，先進規劃與排程(Advance Planning and Scheduling, APS)系統利用先進資訊科技及規劃技術，來改變生產規劃、排程、預測、產能負荷以評估和顧客及供應商間之交流。但臺灣大多數製造業(如金屬製造、汽車零部件組裝等)仍然高度依賴基於excel的生產規劃和ERP排程系統，由於各製程的生產特性及設備的數量差異很大，導致無法確保生產排程規劃結果能有效利用機台產能且合乎現場實際生產狀況。雖然使用APS系統可以滿足特定生產特性和規劃條件，但企業需要付出較高的實施成本，而現今也沒有一個完善的建模方法，使得中小型製造企業(SMEs)使用APS系統時遇到困難的挑戰。本研究將結合雲端運算的特性以及先進規劃排程系統的資訊整合能力，提出一考量製造業現況及需求之雲端先進規劃排程系統架構與功能模組，並依據現有生產規劃流程發展出適用的建模方法。最後，我們亦將本研究所發展的方法應用在國內某車用電子零件業，應用結果顯示此方法能大幅降低系統導入及模型建立的時間並減少人力資源的成本，還能透過雲端先進規劃排程系統快速模擬出符合實際現況的規劃結果。 Digitization has been the main driving force of manufacturing efficiency and quality improvement in recent years. Combined with the Cyber-Physical System (CPS) which connect physical object and computational capabilities, the manufacturing industry facing the global trend of intelligent manufacturing. To enable manufacturers to gain competitive advantage in the global manufacturing environment, the Advanced Planning and Scheduling (APS) system utilizes advanced information technology and planning techniques to change production planning, scheduling, forecasting, capacity load assessment, and communication between customers and suppliers. Most of manufacturing still heavily rely on the excel-based production planning and scheduling of ERP systems are widely used today in Taiwan, because the production characteristics and the number of machines of each process are very different. This way causes the system can't ensure the planning result is effective use of the machine capacity and accord with the actual production status. Although use APS can satisfy the specific production characteristics and planning constraints, the company need to pay the high implementation cost and lack a great modeling approach. The small and medium-sized manufacturing enterprises (SMEs) import advanced planning and scheduling systems have difficulty challenge.This paper combines the characteristics of cloud computing and the information integration ability of APS, to proposes a cloud APS architecture and system functional modules suitable for the current requirements of industry. Then develops an appropriate modeling method based on the existing production planning process. Finally, applied in a well-known automotive parts manufacturing company in Taiwan, to verify that this method can solve the problems in imported the APS in SMEs. So the industry can achieve more accurate scheduling according to market dynamic change, greatly reduce the time of system import and the cost of human resources.
