單元製造系統設計-多目標製造單元形成演算法之發展

Abstract

在今日競爭的製造環境中，迅速湧現的技術反映出顧客的喜好是隨時在改 變的。因此，管理者積極的尋找更具彈性與效率的生產方式，群組技術( GT)是其中被廣泛採用的一項哲學。單元製造(CM)是群組技術的一種應用 。單元製造系統設計是單元製造能否成功實行的關鍵。單元形成是建構初 始單元製造系統設計的第一步，也是最困難的，一般稱為機器單元形成( MCF)或機器工件分群(MCG)問題。目前解決機器單元形成／機器工件分群 問題的演算法已經提出很多，然而卻忽略很多重要的考量，因而限制了這 些演算法在實際單元製造系統環境中的實用性。在本論文中，首先基於生 產流程分析(PFA) 建構一個機器單元形成／機器工件分群問題，同時考慮 許多重要因素，例如單元佈置、機器產能、機器複製、機器投資成本、工 件搬運成本、工件加工順序及時間。然後提出一個多目標的數學規劃模式 以同時最小化(1)總成本(包括機器投資成本及工件搬運成本)(2)製造單元 內機器負荷變異及(3)製造單元間機器負荷變異。接著，提出以基因法則 演算法(GA)及模擬退火法(SA)(本論文將之改良為平行式模擬退火法( PSA))為基底的演算法以解決所述的機器單元形成／機器工件分群問題。 實驗分析和比較的結果展現本文所提的兩個演算法的有效性，並顯示這些 演算法可以在很短的電腦執行時間內為實際發生在工廠內的機器單元形成 ／機器工件分群問題提供一個穩健的製造單元規劃。 Rapidly emerging technologies in today's competitive manufacturing environment reflect customers' preferences that are constantly fluctuating. Hence, managers actively seek more flexible and efficient production approaches, among which include the widely adopted group technology (GT). Cellular manufacturing (CM) is a specific application of GT. CMS design is critical to ensure that CM is successfully implemented. Cell formation is the first and most difficult aspect of constructing a preliminary CMS design. This is known as machine-cell formation (MCF) or machine-component grouping (MCG) problem. A lot of algorithms have been developed to resolve the MCG/MCF problem. Many of these algorithms did not address some important issues, thereby limiting the practical nature of their approaches in a real CMS environment. In this dissertation, the MCG/MCF problem is first constructed based on production flow analysis (PFA) under realistic considerations, e.g. cellular layout, machine capacity, machine duplication and investment cost, intercell and intracell part transportation cost, set up time, operation sequence and processing time of a part. A multi-objective mathematical programming model is then formulated, which aims to minimize (1) total cost which includes intercell and intracell part transportation cost and machine investment cost, (2) intracell machine loading unbalance and (3) intercell machine loading unbalance. Two novel methods, genetic algorithm (GA) and simulated annealing (SA) (as modified with the merits of GA, call parallel SA (PSA)), are adopted to resolve the formulated MCG/MCF problem. Illustrative examples, comparisons and experimental analyses demonstrate the effectiveness of the proposed GA-based and PSA-based algorithms. The proposed algorithms can be used to solve real MCG/MCF problems in factories by providing a robust manufacturing cell formation in a short execution time.