Fukuoka Institute of Technology REPOSITORY SYSTEM

HOME    repository info    mypage        Japanese    library    university    Feedback

Fukuoka Institute of Technology Repository >
02.University Bulletins & Report >
01.RESEARCH BULLETIN OF FUKUOKA INSTITUTE OF TECHNOLOGY >
Vol.39.No.1(57) >

 
Title :自己組織化マップ法による巡回セールスマン問題の解法
Authors :加藤 友彦
小林 徹
Issue Date :30-Sep-2006
Abstract :The traveling salesman problem is one of the most difficult ones in optimization problems. In this study we examine the method o[ B. Angeniol et al. based on the self-organizing maps (SOM) by T. Kohonen, in comparison with the Monte-Carlo (MC) methods that we employed so far. It turns out in the 561- and 1000- city problems that the SOM method gives extremely better results than the MC method in shorter calculation time. In order to improve the SOM further, we propose two methods. One is MC_SOM method in which the MC method is used as a preliminary treatment of the SOM. The other is a kind of simulated annealing method. Both methods are examined for 3795- and 5925- city problem and it is found that they give certain improvements if some proper trealrnents are made.
Type Local :紀要論文
ISSN :02876620
Publisher :福岡工業大学
URI :http://hdl.handle.net/11478/841
citation :福岡工業大学研究論集
AN10036974
39
1
61
67
Citation :福岡工業大学研究論集 Vol.39 no.1 p.61 -67
Appears in Collections:Vol.39.No.1(57)

Files in This Item:

File Description SizeFormat
11478-841_61p加藤友彦.pdf1072KbAdobe PDFView/Open