考量批量生產特性之兩階段混合流線型生產排程

Abstract

在實際生產製造環境中，因為需將許多的變數及不確定性同時納入考慮，因此當今企業便透過導入功能更具彈性的機台以因應製造生產環境的發展與變化，藉由納入批量生產的特性至新型機台中以提升製程的生產效率。但新型機台及既有機台並行使用的現象將會因為機台特性的差異而對生產排程有重大的影響，因此若能將具批量生產的新機台及單一生產的舊機台同時考量至生產排程規劃，將可使排程結果更貼近於實際生產現場排程狀況。 本研究以混合流線型生產排程問題為基礎，探討如何將工件建構成批量的形式，並考量各批量間的加工順序，在新舊機台具備的加工容量不一致之條件下，讓多種工件在同一機台可同時進行加工，以減少工件的完工時間，期望能達成交期限制。另外研究中亦納入獨立與相依設置時間、產品種類及製程型態等因素，以更符合實際製造現場，建構出能考量實務中包含的條件限制且在有限資源下使效益最大化地發揮的一套生產排程計劃，於客戶給定的交期期限內提供足夠的產能及產出量，以減少排程計畫與實際生產狀況的偏差。不僅如此，面對製造業在生產技術上的快速發展，需將求解生產排程結果的時間上限也一併列入考慮，以及時求得最適排程計畫，避免因求解時間過長造成訂單交期的延遲。本研究將探討以最小化總延遲時間為目標，求解具有批量生產機台之兩階段混合流線型生產排程問題，並以混合整數線性規劃模型與基因演算法的建構，求解一套完整的排程計畫，以順應製造環境中競爭且多變的特性。

In the actual manufacturing environment, since many variables and uncertainties need to be taken into consideration at the same time, companies introduced machines with more flexible functions to cope with the development of the manufacturing environment nowadays, and incorporating more production characteristics into new machines to improve the production efficiency of the whole manufacturing operation. However, the coexistence of new machines and old machines will have a significant impact on production scheduling due to differences in machine characteristics. Therefore, if new machines with batch processing and old machines with single processing can be considered to the production scheduling at the same time, the results will be closer to the actual production environment. Based on the hybrid flowshop scheduling problem, this research discusses how to construct jobs into batches, and considers batch sequence, so that multiple jobs can be processed simultaneously on the same machine to reduce the completion time of jobs, and finally meet the deadlines of jobs. In addition, other factors such as independent and dependent setup time, product types, and process recipes are also included in this research, so as to be more in line with the actual manufacturing environment, and then to construct a production scheduling plan which not only considers the conditions in the practical operations and maximize profits under limited resources, but also provides sufficient production capacity and satisfying the due date given by the customers and reduce the deviation between the scheduling plan and the actual production environment. Furthermore, in the face of the rapid development of production technology in the manufacturing industry, the computing time must also be taken into consideration, so as to obtain the optimal or close-to-optimal solutions in a reasonable time and avoid tardiness of jobs. This research will propose the solving method of the two-stage hybrid flow shop scheduling problem with batch production machines with the objective of minimizing the total tardiness time, and obtain a complete scheduling plan by constructing a mixed integer linear programming model and genetic algorithm, to accommodate the competitive and changing nature of the manufacturing environment.