2007 Fiscal Year Final Research Report Summary
Optimization for transport network of waterbus and bottleneck analysis with multi agent system
Project/Area Number |
17360424
|
Research Category |
Grant-in-Aid for Scientific Research (B)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
Naval and maritime engineering
|
Research Institution | National Maritime Research Institute |
Principal Investigator |
MAJIMA Takahiro National Maritime Research Institute, Center for Logistics Research, National Maritime Research Institute, Chief Researcher (30392690)
|
Co-Investigator(Kenkyū-buntansha) |
WATANABE Daisuke Faculty of Marine Technology, Tokyo University of Marine Science and Technology, Associate Professor (30435771)
TAKADAMA Keiki Faculty of Electro-Communications, The University of Electro-Communications, Assistant Professor (20345367)
|
Project Period (FY) |
2005 – 2007
|
Keywords | Optimized Logistics / Simulation Engineering / Algorithm / Disaster Prevention / Multi-Agent System |
Research Abstract |
It is being recognized that transport system with the combination of rivers and ships under disaster circumstance will work effectively as alternative route for the onshore transport system. In fact, the transport operation with ships worked effectively in the Great Hanshin-Awaji earthquake in 1995. Meanwhile, Tokyo metropolitan has been suffering from chronic congestion and heavy commuter rushes for long time. The transport system with rivers and ships is also expected to mitigate such conditions. This research project focused on the waterbus line network as public transit network and exploited two algorithms and their hybrid algorithm to generate optimized waterbus line network. One is algorithm with knowledge based and the other is that of meta-heuristics. Followings describe research achievements for each algorithm. 1) Algorithm with knowledge based approach Firstly, measures or indices used in the research field regarding to the complex network are applied to grasp the characteristi
… More
c of transportation network .The target networks are already existing subway, railway network and a hypothetical waterbus network in Tokyo. The analysis results clarified the characteristic of the transportation network and the roll of waterbus network against the already existing network. Using the knowledge of the measures and the framework of the network evolution model actively investigated in the research domain of the complex network, we developed algorithms to generate public transit network for waterbus. Furthermore the algorithm is modified to incorporate the meta-heuristic method to avoid local optimized solutions. It is recognized by comparison of results for benchmark problem that the developed algorithm has potency to outperform the algorithms presented by precedent research work. 2) Algorithm with meta-heuristic approach Fundamental investigation regarding to the LCS (Learning Classifier System) was conducted for understanding its mechanism and features appropriate for the problem to generate bus line network. After the investigations, we decided to employ the Pittsburgh Genetics-Based Machine Learning. In case of disaster situation, it can be considered that the OD(Origin and Destination) demand changes frequently. The developed algorithm works considering several patterns of OD demand and succeeds to yield the robust networks. Less
|
Research Products
(70 results)
-
-
-
-
-
-
[Journal Article] Learning Classifier Systems : Workshops, IWLCS 2003-2005, Revised Selected Papers2007
Author(s)
Kovacs, T., Llora, X., Takadama, K., Lanzi, P. L., Stolzmann, W., and Wilson, S. W.(Eds.)
Pages
345
Description
「研究成果報告書概要(欧文)」より
-
-
-
-
-
-
-
-
[Presentation] Hub airport location in air cargo system2008
Author(s)
Watanabe, D., Majima, T., Takadama, K., and Katsuhara, M.
Organizer
International Conference on Instrumentation, Control and Information Technology(SICE'08)
Place of Presentation
Tokyo
Year and Date
20080800
Description
「研究成果報告書概要(欧文)」より
-
-
[Presentation] Hub airport location in air cargo system2008
Author(s)
Watanabe, D., Majima, T., Takadama, K., and Katsuhara, M.
Organizer
International Conference on Instrumentation, Control and Information Technology(SICE2008)
Place of Presentation
The University of Electro-Communications
Year and Date
2008-08-20
Description
「研究成果報告書概要(和文)」より
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
[Book] Learning Classifier Systems: Workshops, IWLCS 2003-2005, Revised Selected Papers.2007
Author(s)
Kovacs, T., Llora, X., Takadama, K., Lanzi, P.L., Stolzmann, W., and Wilson, S.W.(Eds.)
Total Pages
345
Publisher
Springer-Verlag
Description
「研究成果報告書概要(和文)」より