考量人力與機台限制下之雙彈性流線型生產排程

Abstract

隨著勞工的薪資逐年提升，企業爲了要維持自身競爭力必須將其納入成本環節中做考量，因此結合人力資源與生產排程就顯得至關重要，只有兩者相結合才能夠排出一個最高效的排程計劃。雙彈性流線型生產排程是目前許多製造業工廠内常見到的生產模式，例如模具加工廠、電子製造產業、半導體的晶片代工廠等。在流線型生產排程中，所有工件的加工順序都是相同的，而第一個彈性指的就是生產環境中至少存在一台機台具有加工不同種類工件和不同加工階段的彈性。由於在生產製造上目前大部分還是仰賴人機協同進行作業，所以在工件準備進行加工前，需要具有相應技能水平的作業員到其準備投入的機台上進行前置作業。故第二個彈性係指作業員在處理不同加工階段中，機台前置作業的技能彈性。

本研究中所欲探討的研究議題包含許多特性，其中在整個典型的流線型生產排程中具備生產多種產品並且考量途程規劃的特性。在機台上也考量了獨立作業時間、機台加工能力限制、機台效率、機台加工能力和具備多功能機台等特性。在工作人員的部分也考量了獨立前置作業時間、工作人員加工能力限制、工作人員效率和工作人員工作時間表。考量以上特性方能建構有效的生產排程方法，以滿足實務製造現場中的需求。

本研究議題中涵蓋了多種要素是過往學者們沒探討過的全新問題，在求解問題上採用了基因演算法和整數規劃模型爲基礎，開發出一個混合式分段啓發式演算法來求解該問題，該演算法在針對中小型排程問題中能夠更有效率的求解，並於有限時間内取得最合適的排程計劃。最後則是會將混合式分段啓發式演算法的結果與基因演算法進行比較，以證明本研究開發之方法在面對該問題時無論在求解效率或求解結果上皆具備實用性與有效性，並且最後也會使用實務案例進行驗證。

With the annual increase in labor wages, businesses must consider incorporating them into their cost considerations in order to maintain their competitiveness. Therefore, the integration of human resources and production scheduling becomes crucial, as only the combination of the two can generate the most efficient scheduling plan. Dual-flexible flow production scheduling is a common production model found in many manufacturing plants, such as mold processing factories, electronic manufacturing industries, and semiconductor chip foundries. In a flow production scheduling, the processing sequence of all workpieces is the same, and the first flexibility refers to the presence of at least one machine in the production environment that can process different types of workpieces and different processing stages. Since most manufacturing processes still rely on human-machine collaboration, operators with the corresponding skill levels need to perform preparatory tasks on the machine before processing the workpiece. Hence, the second flexibility refers to the skill flexibility of operators in performing pre-processing tasks at different processing stages.

The research topic to be explored in this study includes various characteristics, including the production of multiple products in a typical flow production scheduling while considering route planning. The characteristics of machine consideration include independent operation time, machine processing capacity constraints, machine efficiency, machine capacity, and multi-functional machines. The workforce considerations include independent setup time, workforce processing capacity constraints, workforce efficiency, and workforce schedules. Considering these characteristics is essential to construct an effective production scheduling method that meets the needs of practical manufacturing sites.

This research topic encompasses multiple elements that have not been explored by previous scholars, presenting a completely new set of problems. To solve these problems, a combination of genetic algorithms and integer programming models has been employed as the foundation. A Hybrid Segmented Heuristic Algorithm has been developed to address the problem, allowing for more efficient resolution of medium and small-scale scheduling problems and achieving the most optimal scheduling plan within a limited timeframe. Finally, a comparison will be made between the results obtained from the Hybrid Segmented Heuristic Algorithm and those from the genetic algorithm. This comparison aims to demonstrate the practicality and effectiveness of the method developed in this study, both in terms of solution efficiency and results. Additionally, real-world case studies will be utilized for further validation.