考量彈性流線型機台劣化模式之排程與維護策略之研究

Abstract

近年來製造業的快速更迭，欲達最佳生產效率及產能最大化，企業開始重視設備維護的精準操作安排。由於設備成本的問題及更換的不容易，生產系統除了最小化生產排程完工時間和最大化訂單產量外，關於維護排程問題的分配中如何在瞬息萬變的生產條件下進行作業與機台的合理安排也成為其中一個關鍵。另外透過精確的維護排程可以有效降低機台的故障與損壞率及生產成本，從而提升機台效率，並縮短交貨時間提高客戶滿意度以及市場競爭力。除此之外面對生產環境的不停變化，針對設備劣化的計算也相對複雜，因此提出較符合設備退化的劣化模式，方能反應出最真實的生產狀況。 這樣的背景下，生產排程不僅只是分配作業與機台的順序，還需要考慮機台的維護策略，生產流程需要具備足夠的彈性來應對機台分配與作業排序的雙重決策，因此本研究將在彈性流線型生產排程 (Flexible Flowshop Scheduling) 的基礎下，著重於兩種類型的派工方式：作業和機台加工順序的組合，甚至提出少見的最小機台使用率 (Least Utilized Machine) 作為其中一個派工方法，期望能優化整體的生產績效。機台維護策略同時發揮了重要的作用，雖然機台維護策略會耗用生產時間，但忽視維護工作可能導致機台故障率提高。優秀的維護策略不僅能增加機台的運行時間，也能提升機台的壽命。在維護策略上，將探討三種評判標準：維護種類、工件可否被中斷及維護水準，以定義最佳的維護策略。而為了因應生產環境的不確定性，在機台劣化率的部分將考慮模擬隨機劣化模式，以更貼近真實生產背景。最終透過數值分析與國內某航太企業之實務案例驗證所提出之方法，結果表明最佳維護方法集中於特定維護策略，且所提出之最小機台使用率的組合在某些問題規模也有較好的求解結果。

In recent years, the manufacturing industries have evolved rapidly in improving the efficiency of production and capacity. Companies begin to invest more resources in accurate operation arrangements for equipment maintenance. Due to the high cost of equipment and difficulties of replacements, it has become a major challenge on not only how to arrange jobs and machines, but also on minimizing production scheduling completion time and maximizing order output. The other challenges presented to the industries are failure and damage rates of machinery, delivery times and the calculation of equipment degradation. These can be delt with accurate maintenance scheduling and precise degradation modeling programs so as to meet customer requirements and to increase market competitiveness. Given the above background, production scheduling is not only assigning jobs and machine sequences, but also considering equipment maintenance strategies. The production process needs to be flexible enough to handle the dual decisions of machine sequencing and job sequencing which is one of the goals of this paper to correlate by proposing several dispatching rules, such as EDD, FCFS, SPT and LUM (Least Utilized Machine). Effective maintenance strategies can increase machine operating time as well as machine life. In terms of maintenance strategies, three evaluation criteria were explored in order to define the optimal maintenance strategy: types of maintenance, job interruption, and level of maintenance. Also, to cope with the complexity of the production environment, simulating random degradation models were considered in terms of machine degradation rates. So as to be more aligned with the real production background. Ultimately, through numerical analysis and practical case studies, the results show that the optimal maintenance methods stand on specific maintenance strategies. And the proposed Least Utilized Machine in dispatching rule also achieves better results in certain problem scales.