查詢結果分析
相關文獻
- 以遺傳演算法為基礎之二部機器以上排序法之最佳化
- 遺傳演算法於Job Shop排程問題上的研究
- 線性軸幅路網接駁系統最適整合區位、路線與排班模式之研究
- Genetic Algorithm Approach for Designing Fir Hilbert Transformers and Differentiators
- 以類神經網路與遺傳演算法解決系統可用度分派問題
- 自我調適的動態排程系統--限制排程、模糊理論和遺傳演算法的應用
- 用無母數的網路學習於臺股認購權證的定價
- 應用遺傳演算法優選大埔水庫排砂操作規線
- 手寫數字辨識模式之建立--結合遺傳演算法與類神經網路
- A Genetic Algorithm for Tuning Reliable Fuzzy System
頁籤選單縮合
題 名 | 以遺傳演算法為基礎之二部機器以上排序法之最佳化=Using a GA Based Approach for Solving the n-Machine Flow Shop Scheduling Problems |
---|---|
作 者 | 蘇慶隆; 詹榮忠; 楊家和; | 書刊名 | 長榮學報 |
卷 期 | 3:2 2000.01[民89.01] |
頁 次 | 頁69-85 |
分類號 | 494.542 |
關鍵詞 | 遺傳演算法; 模擬退火演算法; 排程問題; 啟發式方法; Genetic algorithm; Simulated annealing algorithm; Scheduling problem; Heuristic methods; |
語 文 | 中文(Chinese) |
中文摘要 | 排程問題 (Scheduling Problem) 於製造及服務業在決策上扮演著重要的角色, 在現今競爭的環境中,有效的排程已經成為市場上生存的必需條件。公司及工廠必須於交貨 日前完成產品,準時交給顧客,假使延誤,除了賠償外,將會造成信譽等其他重要的損失, 因此公司及工廠必須有效安排其所有相關的生產等活動,以使資源做最有效之運用。一個良 好的排程( Scheduling )不但可以減少資源的閒置時間,增加資源的利用率;更可以使工 作能如期完成,以滿足顧客的需求。排程是一種決策,其決策關係著如何將有限的資源為工 作做有效的分配,而決策的過程通常都會有一個或多個目標。Panwalkar(1991) 提出了二部 機器 Flowshop 排程問題加上機器問的運輸時間( Travel Time Between Machines )之最 佳化方法,但三部機器排程問題至今未有文獻提出最佳化的方法,因其複雜特性使得數學規 劃方法求解變得困難,例如其只能求得小型問題之最佳解,但對大型問題卻無法有效成功的 求解。在現實當中此問題實多非小型,因此發展啟發式方法已為較可行的方法而替代了數學 規劃方法求解。 本研究以遺傳演算法( Genetic Algorithm; GA )結合模擬退火演算法 ( Simulated Annealing Algorithm; SA )( The GA-SA Hybrid Algorithm ) ( Mahfound and Goldberg, 1995 )為基礎,將 Panwalkar 之二部機器排程問題推展至三 部機器排程問題之求解最佳化, 根據實驗數據顯示本研究所提出之 GA-SA 可被用於二部機 器排程問題的求解,並可輕易被推展至三部機器以上之排程求解問題, 此為 GA-SA 演算法 優於目前文獻上他法之處。 |
英文摘要 | This study aims to investigate how to apply GA-SA for finding the best solution upon two and more than two machines flow shop scheduling problems. Scheduling problem plays the role of importance in the industry. The industrial environments become very competitive so that the effective scheduling methods are needed in the market. An effective scheduling method is to minimize the cost of resource so as to make the maximum margins. To make a strategic decision that includes scheduling usually has above two goals. But mathematic programming methods are very limited to find two-machine flow shop scheduling problem (Panwalkar, 1991). But the current methodologies are not able to solve more than threemachine flow shop scheduling problems. Because of the complex characteristic of the problems, that makes the current mathematic programming methods are failed to solve the problems. Due to the above reasons, heuristic methods become more important and possible methodologies for finding the optimal or near optimal solution(s). The paper is planning to integrate Genetic Algorithm (GA) and Simulated Annealing Algorithm (SA) for solving the two and more than two machines flow shop scheduling problem. Experimental results are presented to exhibit the superior performance of the proposed methodology for the economic order quantity with quantity discount problem. |
本系統中英文摘要資訊取自各篇刊載內容。