供應鏈大量客製化產品生產排程問題之研究

A Study on Scheduling Problem of Mass Customized Products of the Supply Chain

現今世界的消費者需求已較過去不同，消費者不再喜歡大量生產的制式產品，個人化及差異化商品的趨勢造成企業的管理架構必須被改變，除了成本低、品質高等兩項基本需求之外，產品的差異化、生產製造的彈性、準確的交貨能力等都是目前企業應該要專注的項目。因此，如何全面管理生產供應鏈是工廠生產客製化產品的重要課題。本論文使用傳統基因演算法，探討供應鏈大量客製化產品之生產排程問題，由於基因演算法主要是透過適應函數方式來評估可行的排程，但針對複雜的排程問題則缺乏有效且高正確率的解題效果。本研究相互比較傳統交配方法（順序交配、位置基礎交配、順序基準交配），並根據上述交配方法發展出一個生產排程系統，可用在實驗模擬中的各種方案比較。由實驗結果顯示，位置基礎交配、順序基準交配方法之基因演算法在供應鏈大量客製化生產排程問題上較順序交配的基因演算法有更少的總時程。

In the global competitive environment, the market’s demand changes fast and dramatically. Therefore, enterprises not only have to satisfy the diversified demands of customers, but enterprises also have to respond to the market in a low-cost and fast way. Those are the reasons why we need “mass customization,” which combines the advantages of “mass production” and “customization.” “Mass customization” offers the customized products/services in a mass production’s technology; in other words, “mass customization” reaches the goal of customization at the same cost and within the same time of “mass production.” The purpose of this research is to propose a genetic algorithm The genetic algorithm compares different crossover. The crossover including Order crossover, position-based crossover, order-based crossover. Finally, several examples were offered to verify the feasibility of the genetic algorithm. In examples, position-based crossover and order-based crossover is better than Order crossover. Because position-based crossover and order-based crossover can find the min makespan.