Evaluation function, game-theoretic machine learning algorithm, and the optimal solution for regional ports resources sharing
- 61 Downloads
Seaport resource sharing is critical for regional integration of port clusters. Port resources include the resources of a single port and the strategic resources of port cluster. The decision of a port will have a long-term impact on its competitive port decisions, and understanding the long-term value of an action relative to another is the essence of the opportunity cost trade-off of many reinforcement learning algorithms. The authors construct the evaluation function and the game-theoretic machine learning algorithm of a single port for resources sharing in seaports regional integration. The optimal solution with different parameters and the mechanism of sharing decision have been studied. On the basis of mechanism analysis, the variation law of optimal strategy is further expounded. The following conclusions are given: (1) the resource-sharing decision is beneficial for the overall competitiveness of the port cluster; (2) a “boxed pigs” game exists between ports with different resource output levels; (3) besides the decision of sharing resources or not, the optimal sharing level is also important; (4) opponent’s sharing decision will influence the decision critical point; (5) in the process of game, there is a prisoner’s dilemma.
KeywordsResources sharing Port clusters Evaluation function Game-theoretic machine learning Optimal solution Simulation
We wish to thank the anonymous reviewers who helped to improve the quality of the paper. The authors gratefully acknowledge the helpful comments and suggestions of the reviewers, which have improved the presentation. This paper was partly supported by Humanities and Social Science Project of Ministry of Education of China (Grant Nos. 14jyc630171 and 17YJA870005), the Natural Science Foundation of Zhejiang Province, China (Grant No. LY14D010002).
Compliance with ethical standards
Conflict of interest
All the authors of the manuscript declared that there is no conflict of interest.
- 1.Langen PWD (2004) The performance of seaport clusters, a framework to analyze cluster performance and an application to the seaport clusters of Durban, Rotterdam, and the Lower Mississippi, Rotterdam, ERIM PhD seriesGoogle Scholar
- 4.Hasanbeig M, Pavel L (2018) From game-theoretic multi-agent log linear learning to reinforcement learning. paperuri: (80514eb725d38b07fff0fb0c1705aedc)Google Scholar
- 5.He D et al (2014) A game-theoretic machine learning approach for revenue maximization in sponsored search, arXiv:1406.0728
- 6.Heinrich J, Silver D (2016) Deep reinforcement learning from self-play in imperfect-information games, arXiv:1603.01121
- 11.Mitchell T (1997) Machine learning. McGraw Hill, New York, p 2. ISBN 0-07-042807-7Google Scholar
- 12.Notteboom TE, De Langen PW (2015) Container port competition in Europe. Handb Ocean Contain Transp Logist 220:75–95Google Scholar
- 14.Notteboom TE, César D, Peter W (2009) Ports in proximity: competition and coordination among adjacent seaports. Ashgate Publishing, Farnham, Surrey, United KingdomGoogle Scholar
- 19.Zhou FB (2010) Research on the mode and content of the dislocation development of the Yangtze River Delta ports. Zhejiang daily (theoretical edition)Google Scholar