-
摘要: 根据中国港口的样本特征, 选取了11个主要功能指标, 建立了因子分析模型, 并进行了适用性检验。根据主成分分析法, 提取了3个待定因子, 经过因子旋转和归一化处理, 采用系统凝聚法对中国沿海24个港口的主要功能进行聚类分析。以MATLAB为技术基础, 对烟台港功能类型进行判别。分析结果表明: 根据经济功能因子、城市功能因子和物流功能因子的分类, 中国沿海24个港口可以聚成7类, 第1类为上海港, 第2类为深圳港, 第3类为广州港, 第4类为宁波舟山港, 第5类为青岛港、天津港、大连港, 第6类为厦门港、丹东港、威海港、汕头港、北海港、防城港、海口港、连云港港、营口港、秦皇岛港、日照港, 第7类为唐山港、温州港、台州港、福州港、泉州港、湛江港; 运用人工神经网络方法, 烟台港被判别为第7类。可见, 方法有效。Abstract: According to the sample characteristics of ports in China, 11 major function indexes were selected, factor analysis model was established, and suitability test was carried out. On the basis of principal component analysis, three factors were extracted. By using factor rotation and normalization, the major functions of 24 seaports in China were clustered based on system coagulation method. The functional category of Yantai Port was discriminated by using MATLAB. Analysis result shows that based on the classifications of economic function factor, urban function factor and logistics function factor, 24 seaports in China can be clustered into 7 categories. The first class port is Shanghai Port, the second class port is Shenzhen Port, the third class port is Guangzhou Port, and the forth class port is Ningbo-Zhoushan Port. The fifth class ports are Qingdao Port, Tianjin Port and Dalian Port. The sixth class ports are Xiamen Port, Dandong Port, Weihai Port, Shantou Port, Beihai Port, Fangcheng Port, Haikou Port, Lianyungang Port, Yingkou Port, Qinhuangdao Port and Rizhao Port. The seventh class ports are Tangshan Port, Wenzhou Port, Taizhou Port, Fuzhou Port, Quanzhou Port and Zhanjiang Port. Then Yantai Port is identified into the seventh class by using artificial neural network method. So, the method is effective.
-
表 1 24个港口主要指标数据
Table 1. Major index data of 24 ports
港口 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 上海 50 808 27 377 2 801 1 391 377 13 698 3 221 4 537 25 121 84 347 42 729 深圳 21 125 16 200 2 142 228 200 7 807 3 001 2 252 15 860 14 894 10 604 广州 34 700 7 942 1 100 784 226 8 216 820 3 140 11 628 48 360 33 114 宁波舟山 52 047 23 057 1 093 665 128 4 454 739 1 396 9 210 35 925 15 956 青岛 30 029 20 665 1 002 762 126 4 436 537 1 465 8 115 42 484 34 172 天津 35 593 18 245 850 969 201 6 354 805 2 000 12 129 54 260 27 000 厦门 9 702 5 394 504 174 78 1 560 454 419 2 970 5 853 2 625 大连 24 588 8 585 453 583 95 3 858 470 1 183 5 082 33 956 23 404 连云港 10 060 5 508 300 488 33 750 44 310 974 11 493 8 514 营口 15 085 2 685 204 234 17 704 24 178 1 235 10 130 7 861 唐山 10 853 5 477 24 729 78 3 561 92 810 5 796 21 862 18 340 秦皇岛 25 231 2 788 40 286 28 809 50 240 1 004 7 573 4 864 丹东 3 259 423 22 243 19 564 19 198 563 4 549 3 901 温州 4 263 138 38 772 103 2 424 140 1 083 3 508 18 309 15 175 台州 3 898 445 6 574 58 1 965 138 710 3 061 14 400 8 282 福州 6 703 1 562 118 636 96 2 284 203 1 134 3 518 14 896 8 160 泉州 7 224 671 121 678 128 2 705 85 906 4 184 10 725 6 443 威海 1 620 1 002 35 252 38 1 780 118 484 4 485 8 456 7 353 日照 15 102 9 980 71 285 19 773 93 212 1 520 8 645 3 956 汕头 2 806 330 72 507 32 975 63 569 1 321 2 316 1 818 湛江 6 682 4 095 28 754 39 1 049 33 468 1 141 9 663 5 669 北海 620 238 5 158 10 314 7 80 253 4 545 3 700 防城 3 701 2 968 23 85 8 212 22 38 277 5 096 2 193 海口 2 614 74 35 156 29 443 36 235 305 7 612 2 368 烟台 11 189 3 792 153 652 83 3 434 350 1 023 7 988 22 675 18 468 
下载: 导出CSV
表 2 检验结果
Table 2. Test results
KMO检验统计量 0.852 33 Bartlett球度检验 卡方 524.67 自由度 55 概率 7.87×10-78
下载: 导出CSV
表 3 方差贡献率
Table 3. Variance contributions
指标 特征值 方差贡献率/% 累积贡献率/% X1 9.089 9 82.635 9 82.635 9 X2 0.915 3 8.320 6 90.956 5 X3 0.588 0 5.345 3 96.301 8 X4 0.175 6 1.596 6 97.898 4 X5 0.120 9 1.098 9 98.997 2 X6 0.037 2 0.338 2 99.335 4 X7 0.025 9 0.235 3 99.570 7 X8 0.025 4 0.230 7 99.801 4 X9 0.012 9 0.117 6 99.919 0 X10 0.006 7 0.061 2 99.980 1 X11 0.002 2 0.019 9 100.000 0
下载: 导出CSV
表 4 因子载荷
Table 4. Factor loadings
指标 待定因子1 待定因子2 待定因子3 X1 0.850 1 0.074 8 0.464 9 X2 0.865 8 -0.084 1 0.440 9 X3 0.924 5 -0.358 3 0.032 0 X4 0.762 6 0.544 9 -0.194 6 X5 0.966 3 -0.019 5 -0.202 4 X6 0.981 2 -0.066 9 -0.151 5 X7 0.850 5 -0.493 9 -0.124 4 X8 0.966 3 0.003 6 -0.210 8 X9 0.974 9 -0.157 5 -0.097 0 X10 0.941 2 0.283 6 0.065 8 X11 0.889 8 0.351 1 0.031 7
下载: 导出CSV
表 5 因子旋转结果
Table 5. Factor rotation result
特征值 方差贡献率/% 累积贡献率/% 4.408 8 40.080 4 40.080 4 3.630 9 33.008 1 73.088 5 2.553 5 23.213 3 96.301 8
下载: 导出CSV
表 6 旋转后的因子载荷
Table 6. Factor loadings after rotation
指标 旋转后的待定因子1 旋转后的待定因子2 旋转后的待定因子3 X1 0.344 0 0.398 6 0.816 8 X2 0.468 3 0.296 6 0.802 3 X3 0.834 5 0.258 0 0.470 3 X4 0.208 9 0.912 7 0.199 5 X5 0.720 3 0.611 4 0.287 0 X6 0.743 5 0.568 1 0.338 5 X7 0.931 5 0.164 3 0.296 7 X8 0.708 0 0.631 4 0.279 8 X9 0.780 2 0.479 0 0.382 7 X10 0.407 1 0.736 6 0.512 0 X11 0.340 7 0.768 3 0.457 9
下载: 导出CSV
表 7 功能因子系数
Table 7. Function factor coefficients
指标 旋转后的待定因子1 旋转后的待定因子2 旋转后的待定因子3 X1 -0.274 0 -0.140 4 0.738 7 X2 -0.143 1 -0.256 0 0.702 7 X3 0.307 4 -0.251 3 0.094 2 X4 -0.222 5 0.600 8 -0.246 0 X5 0.206 5 0.157 1 -0.250 9 X6 0.211 2 0.091 3 -0.174 5 X7 0.495 2 -0.280 7 -0.144 0 X8 0.194 8 0.180 6 -0.263 3 X9 0.243 5 -0.013 1 -0.094 1 X10 -0.177 5 0.255 5 0.149 9 X11 -0.209 6 0.326 1 0.096 8
下载: 导出CSV
表 8 归一化结果
Table 8. Normalized results
港口 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 上海 0.952 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 深圳 -0.203 0.181 0.529 -0.781 0.043 0.126 0.863 -0.016 0.255 -0.693 -0.571 广州 0.325 -0.424 -0.217 0.071 0.179 0.187 -0.494 0.379 -0.085 0.123 0.530 宁波舟山 1.000 0.684 -0.221 -0.112 -0.351 -0.371 -0.545 -0.396 -0.280 -0.181 -0.309 青岛 0.144 0.508 -0.286 0.036 -0.359 -0.374 -0.671 -0.366 -0.368 -0.021 0.582 天津 0.360 0.331 -0.395 0.354 0.044 -0.089 -0.503 -0.128 -0.045 0.266 0.231 厦门 -0.647 -0.610 -0.643 -0.864 -0.621 -0.800 -0.722 -0.831 -0.782 -0.914 -0.961 大连 -0.068 -0.377 -0.680 -0.237 -0.528 -0.459 -0.712 -0.491 -0.612 -0.229 0.055 连云港 -0.633 -0.602 -0.789 -0.382 -0.866 -0.920 -0.977 -0.879 -0.942 -0.776 -0.673 营口 -0.438 -0.809 -0.858 -0.772 -0.948 -0.927 -0.990 -0.938 -0.921 -0.810 -0.705 唐山 -0.602 -0.604 -0.986 -0.013 -0.621 -0.503 -0.947 -0.657 -0.554 -0.524 -0.192 秦皇岛 -0.043 -0.801 -0.975 -0.692 -0.889 -0.912 -0.973 -0.910 -0.940 -0.872 -0.851 丹东 -0.897 -0.974 -0.988 -0.758 -0.941 -0.948 -0.993 -0.929 -0.975 -0.946 -0.898 温州 -0.858 -0.995 -0.976 0.052 -0.484 -0.672 -0.917 -0.536 -0.738 -0.610 -0.347 台州 -0.873 -0.973 -0.999 -0.251 -0.727 -0.740 -0.919 -0.702 -0.774 -0.705 -0.684 福州 -0.763 -0.891 -0.919 -0.156 -0.524 -0.693 -0.878 -0.513 -0.737 -0.693 -0.690 泉州 -0.743 -0.956 -0.917 -0.092 -0.348 -0.630 -0.952 -0.614 -0.684 -0.795 -0.774 威海 -0.961 -0.932 -0.979 -0.744 -0.836 -0.767 -0.931 -0.802 -0.660 -0.850 -0.729 日照 -0.437 -0.274 -0.953 -0.694 -0.938 -0.917 -0.947 -0.923 -0.898 -0.846 -0.896 汕头 -0.915 -0.981 -0.952 -0.354 -0.871 -0.887 -0.965 -0.764 -0.914 -1.000 -1.000 湛江 -0.764 -0.706 -0.983 0.025 -0.829 -0.876 -0.984 -0.809 -0.929 -0.821 -0.812 北海 -1.000 -0.988 -1.000 -0.888 -0.987 -0.985 -1.000 -0.981 -1.000 -0.946 -0.908 防城 -0.880 -0.788 -0.987 -1.000 -1.000 -1.000 -0.991 -1.000 -0.998 -0.932 -0.982 海口 -0.923 -1.000 -0.979 -0.891 -0.886 -0.966 -0.982 -0.913 -0.996 -0.871 -0.973 烟台 -0.589 -0.728 -0.894 -0.132 -0.594 -0.522 -0.787 -0.562 -0.378 -0.504 -0.186
下载: 导出CSV
表 9 24个港口功能因子系数
Table 9. Function factor cofficients of 24 ports
序号 港口 旋转后的待定因子1 旋转后的待定因子2 旋转后的待定因子3 1 上海 2.685 7 2.010 5 0.783 9 2 深圳 3.430 8 -1.898 0 -0.023 9 3 广州 0.611 2 1.639 7 -0.086 0 4 宁波舟山 -0.398 2 -0.400 5 2.869 4 5 青岛 -0.609 0 0.765 3 1.753 0 6 天津 -0.136 3 1.362 3 1.106 8 7 厦门 0.372 5 -1.166 8 -0.049 2 8 大连 -0.460 6 0.545 5 0.609 3 9 连云港 -0.595 1 -0.253 2 0.097 9 10 营口 -0.537 9 -0.738 8 0.343 6 11 唐山 -0.562 3 0.919 0 -0.415 4 12 秦皇岛 -0.701 3 -0.772 3 0.684 6 13 丹东 -0.281 1 -0.657 1 -0.517 7 14 温州 -0.315 5 1.137 0 -1.328 5 15 台州 -0.251 8 0.368 9 -1.005 7 16 福州 -0.078 1 0.514 4 -1.050 9 17 泉州 -0.011 5 0.582 8 -1.229 4 18 威海 0.032 2 -0.423 9 -0.763 3 19 日照 -0.608 0 -0.967 1 0.839 6 20 汕头 -0.197 4 -0.173 9 -0.948 8 21 湛江 -0.679 1 0.366 3 -0.540 7 22 北海 -0.251 2 -0.824 8 -0.526 3 23 防城 -0.283 7 -1.137 9 -0.071 8 24 海口 -0.174 4 -0.797 5 -0.530 3
下载: 导出CSV
-
[1] 林国平, 杜从原, 曹宝根. 从港口的代际功能看港口功能的发展[J]. 港口科技动态, 2006, 18 (10): 12-13. doi: 10.3969/j.issn.1673-6826.2006.10.003LIN Guo-pin, DU Cong-yuan, CAO Bao-gen. Development of port functions from the point view of intergenerational function of port[J]. Science and Technology of Ports, 2006, 18 (10): 12-13. (in Chinese) doi: 10.3969/j.issn.1673-6826.2006.10.003 [2] UNCTAD. The fourth generation port[R]. New York: UNCTAD, 1999. [3] ROBINSON R. Ports as elements in value-driven chain systems: the new paradigm[J]. Maritime Policy and Management, 2002, 29 (3): 241-255. doi: 10.1080/03088830210132623 [4] CARBONE V, MARTINO M D. The changing role of ports in supply-chain management: an empirical analysis[J]. Mari-time Policy and Management, 2003, 30 (4): 305-320. doi: 10.1080/0308883032000145618 [5] NOTTEBOOM T E, RODRIGUE J P. Port regionalization: towards a new phase in port development[J]. Maritime Policy and Management, 2005, 32 (3): 297-313. doi: 10.1080/03088830500139885 [6] 曾渊, 杨晗熠, 熊丛博. 港口功能评价方法研究及应用[J]. 海岸工程, 2006, 25 (2): 100-106. doi: 10.3969/j.issn.1002-3682.2006.02.014ZENG Yuan, YANG Han-yi, XIONG Cong-bo. Study and application of harbor function evaluation method[J]. Coastal Engineering, 2006, 25 (2): 100-106. (in Chinese) doi: 10.3969/j.issn.1002-3682.2006.02.014 [7] 曾渊. DEA和MDEA模型在港口功能评价方法中应用的研究[D]. 青岛: 中国海洋大学, 2007.ZENG Yuan. The study on the port assessment based on DEA and MDEA model[D]. Qingdao: Ocean University of China, 2007. (in Chinese) [8] 梁钦. 港口物流功能综合评价研究[D]. 大连: 大连海事大学, 2008.LI ANG Qin. Study on synthetie evaluation of port logistics function[D]. Dalian: Dalian Maritime University, 2008. (in Chinese) [9] 陈航, 栾维新, 王跃伟. 我国港口功能与城市功能关系的定量分析[J]. 地理研究, 2009, 28 (2): 475-483. doi: 10.3321/j.issn:1000-0585.2009.02.021CHEN Hang, LUAN Wei-xin, WANG Yue-wei. Quantitative study on relationship between port function and urban functionin China[J]. Geographical Research, 2009, 28 (2): 475-483. (in Chinese) doi: 10.3321/j.issn:1000-0585.2009.02.021 [10] 孙柯洋, 赵一飞. 基于因子分析和SE-DEA的港口上市公司经营效率评价[J]. 上海海事大学学报, 2010, 31 (1): 67-71. doi: 10.3969/j.issn.1672-9498.2010.01.014SUN Ke-yang, ZHAO Yi-fei. Operating efficiency evaluation of listed port companies based on factor analysis and SE-DEA model[J]. Journal of Shanghai Maritime University, 2010, 31 (1): 67-71. (in Chinese) doi: 10.3969/j.issn.1672-9498.2010.01.014 [11] 翁钢民, 贾苗苗, 徐晓娜. 河北省滨海旅游发展动力系统的因子分析[J]. 地域研究与开发, 2010, 29 (1): 80-89. doi: 10.3969/j.issn.1003-2363.2010.01.016WENG Gang-min, JIA Miao-miao, XU Xiao-na. A study of impetus system about coastal tourism development in Hebei Province[J]. Areal Research and Development, 2010, 29 (1): 80-89. (in Chinese) doi: 10.3969/j.issn.1003-2363.2010.01.016 [12] 卢宁, 李国平, 刘光岭. 中国自主创新与区域经济增长——基于1998-2007年省际面板数据的实证研究[J]. 数量经济技术经济研究, 2010, 27 (1): 3-18. https://www.cnki.com.cn/Article/CJFDTOTAL-SLJY201001002.htmLU Ning, LI Guo-ping, LIU Guang-ling. Self-innovation and regional economy growth: based on China provincial 1998-2007 panel data[J]. The Journal of Quantitative and Technical Economics, 2010, 27 (1): 3-18. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-SLJY201001002.htm [13] 张旋, 王启山, 于淼, 等. 基于聚类分析和水质标识指数的水质评价方法[J]. 环境工程学报, 2010, 4 (2): 476-480. https://www.cnki.com.cn/Article/CJFDTOTAL-HJJZ201002048.htmZHANG Xuan, WANG Qi-shan, YU Miao, et al. An approach for water quality assessment based on hierarchical cluster analysis and comprehensive water quality identification index[J]. Chinese Journal of Environmental Engineering, 2010, 4 (2): 476-480. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HJJZ201002048.htm [14] 唐万梅. 基于MTALAB的神经网络判别模型[J]. 数学的实践与认识, 2006, 36 (2): 46-50. doi: 10.3969/j.issn.1000-0984.2006.02.009TANG Wan-mei. The judging model of nueral network based on MATLAB[J]. Mathematics in Practice and Theory, 2006, 36 (2): 46-50. (in Chinese) doi: 10.3969/j.issn.1000-0984.2006.02.009 [15] 吴贤国, 李惠强, 朱道智, 等. 基于人工神经网络的建筑工程工料估算[J]. 系统工程理论与实践, 2000, 20 (8): 128-131. https://www.cnki.com.cn/Article/CJFDTOTAL-XTLL200008025.htmWU Xian-guo, LI Hui-qiang, ZHUDao-zhi, et al. Estimating of housing building's major material and labour consumption based on neural network[J]. Systems Engineering—Theory and Practice, 2000, 20 (8): 128-131. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-XTLL200008025.htm -
点击查看大图
图(3) / 表(9)
计量
- 文章访问数: 670
- HTML全文浏览量: 108
- PDF下载量: 567
- 被引次数: 0
