Estimating methods of passenger throughput for hub airport based on reverse gravity model
-
摘要: 为了利用航空OD客流数据精确估计机场旅客吞吐量, 描述了全连通线性航线网络和枢纽辐射航线网络, 根据枢纽机场之间的OD客流矩阵, 建立了逆向重力模型标准代数算法和基于哈密顿图原理的简化代数算法, 分析了在不同距离修正指数下, 算法的拟合与估计精度。分析结果表明: 简化代数算法与模型标准代数算法最优吞吐量回归直线的可决系数都为0.84, 斜率分别为0.91和0.87, 比较接近。可见算法的拟合效果优良, 估计精度高, 且简化代数算法需要的数据少, 尤其适合大规模OD矩阵分析。Abstract: To accurately estimate airport passenger throughput based on airline passenger flow OD data, both full-connected linear network and hub-spoke network were discussed, according to passenger flow OD matrix between hub airports, a standard algebraic algorithm of reverse gravity model was developed, and a simplified algebraic algorithm was put forward with the Hamiltonian graphs, and the fitting precisions of two algorithms were studied under defferent ranging exponents.Analysis result indicates that two algorithms' deterministic coefficients of the best throughput regression lines are 0.84, their slope values are 0.91 and 0.87 respectively.So two algorithms have good fitting effects and high estimating precisions, and simplified algebraic algorithm needs only a very small data, which can be used for large-scale OD matrix analysis.
-
Key words:
- traffic planning /
- hub airport /
- passenger throughput /
- reverse gravity model
-
表 1 2004年15个主要机场之间航空OD客流及航线里程
Table 1. Airline OD passenger flows and distances for 15 main airports in 2004
城市 北京 上海 广州 深圳 成都 昆明 海口 西安 杭州 厦门 重庆 青岛 大连 南京 武汉 北京 Q1 387 213 163 161 111 73 105 108 61 59 82 100 92 56 上海 1 178 Q2 171 222 88 53 72 63 1 91 56 94 66 1 67 广州 1 967 1 308 Q3 1 94 69 119 49 107 41 58 17 25 51 47 深圳 2 077 1 343 140 Q4 81 50 144 45 71 44 59 32 14 42 5 成都 1 697 1 782 1 390 1 446 Q5 82 12 39 26 5 20 8 5 26 26 昆明 2 266 2 042 1 357 1 245 711 Q6 27 30 11 13 75 5 5 10 10 海口 2 493 1 762 548 603 1 757 1 046 Q7 21 22 18 27 5 5 24 23 西安 1 034 1 351 1 528 1 635 647 1 228 1 860 Q8 19 5 31 11 5 12 12 杭州 1 200 176 1 099 1 179 1 699 2 089 1 606 1 215 Q9 39 8 26 21 0 18 厦门 1 774 878 567 549 1 911 1 680 1 100 1 932 717 Q10 5 5 5 20 12 重庆 1 640 1 537 1 188 1 290 313 649 1 253 603 1 500 1 516 Q11 5 5 21 12 青岛 646 693 1 867 2 049 1 705 2 373 2 435 1 215 792 1 571 1 577 Q12 28 13 14 大连 579 1 051 2 285 2 368 1 989 2 597 2 920 1 385 1 171 1 929 2 008 358 Q13 16 5 南京 981 273 1 255 1 526 1 618 1 870 1 990 1 104 240 929 1 305 552 910 Q14 9 武汉 1 133 761 873 938 1 047 1 364 1 379 735 656 910 801 1 017 1 460 504 Q15 
下载: 导出CSV
表 2 机场旅客吞吐量估计
Table 2. Estimations of airport passenger throughput
万人 机场吞吐量 实际吞吐量 简化代数算法估计值 标准代数算法估计值 γ=-0.2 γ=-0.3 γ=-0.4 γ=-0.2 γ=-0.3 γ=-0.4 Q1 1 771 3 535 2 406 1 638 3 630 2 508 1 732 Q2 1 430 1 663 1 174 828 1 708 1 223 876 Q3 1 061 1 558 1 090 762 1 599 1 136 806 Q4 972 1 216 846 588 1 249 882 622 Q5 673 866 594 407 890 619 430 Q6 551 713 481 324 732 501 343 Q7 592 768 518 350 788 540 370 Q8 447 670 463 320 688 482 338 Q9 476 548 389 276 562 405 292 Q10 364 464 320 222 476 334 234 Q11 441 635 442 307 653 460 325 Q12 345 437 302 209 535 315 228 Q13 305 334 226 153 343 236 162 Q14 336 427 304 216 439 317 228 Q15 316 503 356 253 516 371 267
下载: 导出CSV
表 3 2种算法比较
Table 3. Comparison of two algorithms
类别 2种算法的机场旅客吞吐量估计值之间的比较 2种算法的机场旅客吞吐量估计值与实际值的比较 简化代数算法 标准代数算法 γ取值 -0.2 -0.3 -0.4 -0.2 -0.3 -0.4 -0.2 -0.3 -0.4 可决系数 1.00 1.00 1.00 0.83 0.84 0.86 0.83 0.84 0.86 斜率 1.03 1.04 1.06 0.63 0.91 1.33 0.61 0.87 1.26
下载: 导出CSV
-
[1] 陈宏民, 郭甘雨. 2000年上海航空客运需求量预测[J]. 系统工程理论方法应用, 1996, 5(2): 18-22.Chen Hong-min, Guo Gan-yu. Forecasting demand of passenger volume of air transportation in Shanghai in 2000[J]. Systems Engineering-Theory Methodology Applications, 1996, 5(2): 18-22. (in Chinese) [2] 葛折贵, 葛折圣. 基于遗传算法和神经网络理论的机场航空业务量的预测模型[J]. 现代交通技术, 2005, 2(4): 65-67.Ge Zhe-gui, Ge Zhe-sheng. Aprediction model of airport aviation portfolio based on genetic arithmetic and artificial neural network theory[J]. Modern Transportation Technology, 2005, 2(4): 65-67. (in Chinese) [3] 焦朋朋. 机场旅客吞吐量的影响机理与预测方法研究[J]. 交通运输系统工程与信息, 2005, 5(1): 107-110.Jiao Peng-peng. Forecasting method andits mechanismof impacts on airport passenger throughput[J]. Journal of Transportation Systems Engineering and Information Technology, 2005, 5(1): 107-110. (in Chinese) [4] Basar G, Bhat C R. Aparameterized consideration set model for airport choice: an application to the San Francisco Bay Area[J]. Transportation Research Part B, 2004, 38(10): 889-904. [5] Hess S, Polak J W. Mixedlogit modeling of airport choice in multiairport regions[J]. Journal of Air Transport Management, 2005, 11(2): 59-68. [6] Shen Guo-qiang. Reverse-fitting the gravity model tointer-city airline passenger flows by an algebraic simplification[J]. Journal of Transport Geography, 2004, 12(3): 219-234. [7] 张远, 李俊, 姜建平. 城市常规公共交通线网规划方法[J]. 长安大学学报: 自然科学版, 2006, 26(6): 86-89.Zhang Yuan, Li Jun, Jiang Jian-ping. Planning methods for conventional transit networkin cities[J]. Journal of Chang'an University: Natural Science Edition, 2006, 26(6): 86-89. (in Chinese) [8] 彭辉, 魏金丽, 陈宽民. 运输通道公路旅客中长距离OD模型构造及分段客运量预测[J]. 中国公路学报, 2006, 19(2): 101-105.Peng Hui, Wei Jin-li, Chen Kuan-min. Construction of OD model of middle and long distance highway trips and prediction of segment passenger volume of transport corridor[J]. China Journal of Highway and Transport, 2006, 19(2): 101-105. (in Chinese) [9] 高强, 朱金福, 陈可嘉. 航空收益管理中多航段舱位控制模型[J]. 交通运输工程学报, 2005, 5(4): 82-85.Gao Qiang, Zhu Jin-fu, Chen Ke-jia. Multi-leg seat inventory control model for airline revenue management[J]. Journal of Traffic and Transportation Engineering, 2005, 5(4): 82-85. (in Chinese) [10] Wojahn O W. Airline network structure and the gravity model[J]. Transportation Research Part E, 2001, 37(4): 267-279. [11] 陈尚云, 杜文, 高世廉. 我国特大城市出行分布模型及其参数的研究[J]. 系统工程, 2002, 20(4): 63-66.Chen Shang-yun, Du Wen, Gao Shi-lian. The study on the model and parameter of urban-trip distribution in metropolises of China[J]. Systems Engineering, 2002, 20(4): 63-66. (in Chinese) [12] Chen G, Shreve WE, Wei B. Hamiltonian graphs involving neighborhood unions[J]. Journal of Graph Theory, 2006, 53(2): 83-100. -
点击查看大图
图(5) / 表(3)
计量
- 文章访问数: 176
- HTML全文浏览量: 62
- PDF下载量: 574
- 被引次数: 0
