利用基因演算法應用於動態排程-以傳統鋁門窗業為例

Use Genetic Algorithm apply in dynamic scheduling–for Traditional Windows Factory

陳明偉(Ming-Wei Chen)

指導教授 ： 葉佳炫

元智大學/資訊學院/資訊管理學系/碩士(2007年)

https://doi.org/10.6838/YZU.2007.00003

摘要

近年來人工智慧已廣泛的被應用在生產排程的領域上，包括基因演算法(Genetic Algorithms)、基因規劃(Genetic Programming )、模糊理論(Fuzzy Theory)、類神經網路(Neural Network)等，而鋁門窗生產的規模及結構變動讓此傳統產業的工作排程漸漸趨向NP-hard問題，因此如何在可接收的時間內產生有效的排程，是現階段傳統鋁門窗普遍遇到的問題。
因為現在大多文獻討論的都為改善既有的靜態工作排成，所以本研究目的主要是依據現實工廠流程，模擬動態的生產行為模式，使用啟發式方法(Heuristic)找出初始解後再利用排程的空檔時間以基因演算法找出是否有更好的解，進而提升排程的品質。

關鍵字

動態排程 ； 生產排程 ； 氣密窗 ； 基因演算法

並列摘要

The artificial intelligence has been fully utilized in the field of productions, which includes the Genetic Algorithms, Genetic Programming, Fuzzy Theory and Neural Network etc. However, the scale of tight window production and the changes of infrastructure makes productions of tradition industry having the problems of NP-hard step by step, therefore, how to build up the effective process of production has become a major issue which encountered by the tight window industry.

Owing to the factor that currently most of the documents are discussing how to improve the passive manner of production process, therefore, this research is mainly based on the current production process, building up the active business model with Heuristic to find out the first solution, then make the best use of the interval of production process with Genetic Algorithms to find out if any better solution, so as to uplift quality of production process.

並列關鍵字

Dynamic Scheduling ； Production Scheduling ； Tight Window ； Genetic

參考文獻

[4]Haupt, R.L. (1995), ”An Introduction to Genetic Algorithm for Electromagnetics ”, IEEE Antennas and Propagation Magazine,pp.7-15.

[7]Syswerda, G. (1989) , “Uniform crossover in genetic algorithms”, in Proc. 3rd Int. Conf. Genetic Algorithms, pp.2-9.

[9]Funda Sivrikaya-Serifoglu, Gunduz Ulusoy (1999) ,“Parallel machine scheduling with earliness and tardiness penalties”, Copmputers & Operations Research, 26, pp.773-787.

[13]J.J. Grefenstette (1986), ”Optimization of. Control Parameter for Genetic Algorithms”, IEEE Transactions on Intelligent Systems,. Vol. SMC-16, No.1, pp.122-128.

[1]Dileep R. Sule, (1996), ”Industrial Scheduling,” PWD Publishing Co, Boston, MA.