國內紡織業發展60年，整個布料的供應鏈的交期瓶頸常會在染整廠發生，本研究欲探討染整廠排程的瓶頸，並評估應用先進排程系統（Advanced Planning & Scheduling System，簡稱APS）能解決其交期遲滯問題的可行性。 本研究以個案訪談的方式，依照Leavitt (1964) 鑽石模型的架構，深入了解個案公司組織結構、製造流程及作業流程、個案的排程邏輯、過去導入經驗，以及現有的系統設備是否足以支應APS運作需要，再探討應用 APS 是否能解決目前交期遲滯的問題，做為個案公司評估開發 APS 的參考。 研究發現，個案交期遲滯的主因來自於產能不足及重修造成部分染缸超過負載量，形成惡性循環。而個案公司的資訊基礎建設尚不足以開發APS，應著手增加資訊基礎建設，並分析相關製程因素，以優化製程及降低重修率，減少干擾預計交期的變數。未來若資訊基礎建設更加完備時，使排程決策更有數據化依據。 ;The textile industry in Taiwan has been progressing in the past 60 years; however, the fabric manufacturer has challenged dyeing factories for delaying the manufacturing process most of the time. This study attempts to identify the challenges that a dyeing factory has on manufacturing scheduling. Advanced Planning & Scheduling System (APS) is a scheduling system that can compute multiple factors to improve manufacturing efficiency. This study will evaluate the feasibility of adopting an APS to overcome the challenges. The research applies Leavitt’s Diamond Model and conducts a case study to identify the case company’s organizational structure, people’s responsibility, the manufacturing and the operating processes, the system-implementation history, the current scheduling logic that the company applies. The study also evaluates whether or not the current technology infrastructure can support the demand. Furthermore, the study further investigates if the APS is able to resolve the delay problem that the dyeing factory currently has, for helping the company make a solid decision on APS adoption. The results lead to several major findings. First, the delay problem the company is suffering has been caused by the insufficient capacity, and the remake has further over-loaded the capacity, leading to a vicious circle of delaying. The infrastructure and the database of the company are insufficient for automatically judge the color order depending on the depth, so it is not quite feasible for the company to adopt an APS currently. Instead, the company should build up more data basis, and further analyze the relationship between remake and various manufacturing factors that might interfere the scheduling. In the future, only when the infrastructure construction of information flow has been setup, can the scheduling be computed by simulation techniques, and the manager can do a relative more solid decision than before.
