基于KFCM双重聚类的铁路客运产品类别划分

详细信息    查看全文 | 推荐本文 |

英文篇名：Classification of Railway Passenger Products Based on Kernel Fuzzy C-means Double Clustering 作者：刘帆洨 ; 彭其渊 ; 鲁工圆 ; 潘金山 英文作者：LIU Fan-xiao;PENG Qi-yuan;LU Gong-yuan;PAN Jin-shan;School of Transportation and Logistics, Southwest Jiaotong University;National United Engineering Laboratory of Integrated and Intelligent Transportation; 关键词：铁路运输 ; 客运产品 ; 有效能力隶属度 ; 核模糊C均值聚类(KFCM) ; 双重聚类 英文关键词：railway transportation;;passenger products;;effective capability membership;;Kernel fuzzy C-means;;double clustering 中文刊名：JTGC 英文刊名：Journal of Transportation Engineering and Information 机构：西南交通大学交通运输与物流学院;综合交通运输智能化国家地方联合工程实验室; 出版日期：2019-06-15 出版单位：交通运输工程与信息学报 年：2019 期：v.17;No.64 基金：中国铁路总公司科技研究开发计划重大课题(Z2017-X002) 语种：中文; 页：JTGC201902003 页数：7 CN：02 ISSN：51-1652/U 分类号：20-26

摘要

客运产品是铁路运输市场的主要供给,开行方案是客运产品设计的核心内容。将客运产品进行类别划分,是不同类型客运产品需求演变趋势分析的重要基础,有利于简化客运产品优化设计问题。本文以不同列车开行方案属性为样本特征变量,考虑列车能力利用对客运产品优化设计的影响,结合平均列车客座率提出了有效能力隶属度,构建了基于KFCM的双重聚类模型对样本进行聚类,利用Xie-beni和分离系数法确定最佳聚类数。最后对京沪高铁进行实例分析,研究结果表明,将该线客运产品分为4类可获得较好的聚类效果,不同类别的客运产品表现出明显的结构特性。
        Passenger transport products are the main supply of railway transport market, and the operation plan is the core for passenger transport products design. It is the significant foundation for analyzing the evolution trend of passenger demand for different types of products, which is helpful to simplify the optimization design of passenger products. Considering the influence of train capacity utilization on passenger product design, the attributes of different train operation plan are taken as sample characteristic variables. Combining with the average train passenger load factor, the effective capacity membership is proposed. A double clustering model based on KFCM is constructed for the samples, and the optimal clustering number is determined by Xie-beni and separation coefficient method. Finally, a case study of Beijing-Shanghai high-speed railway is carried out. The results show that better clustering results can be obtained by dividing passenger products into four categories, and different types of passenger products show obvious structural characteristics.

引文

[1]强丽霞,刘军,李春艳,等.高速铁路客运产品设计方法及策略研究[J].铁道运输与经济,2013,35(9):18-23.
    [2]汤莲花,李春艳.基于旅客满意度评价的铁路客运产品设计调整[J].铁路计算机应用,2012,21(8):25-28.
    [3]汤莲花,徐行方.高速铁路客运产品组合设计策略研究[J].铁道运输与经济,2018,40(11):44-48.
    [4]王培.中国高速铁路客运产品设计[J].铁道经济研究,2010,6:23-26.
    [5]王正彬,马驷.城际客运专线列车开行方案模型与算法[J].交通运输工程与信息学报,2017,1(15):28-33.
    [6]CLAESSENS M T,Dijk N M V,ZWANEVELD P J.Cost optimal allocation of rail passenger lines[J].European Journal of Operational Research,2007,110(3):474-489.
    [7]CHANG Y H,YEH C H,SHEN C C.AMultiobjective model for passenger train services planning:application to Taiwan’s high-speed rail line[J].Transportation Research Part B,2000,34(2):91-106.
    [8]张振利,崔艳萍.铁路客运产品属性研究[J].铁道运输与经济,2013,35(11):27-31.
    [9]李学民.高速铁路客运产品的设计与营销[J].铁道运输与经济,2011,33(10):45-47.
    [10]周凯建.基于比较视野的铁路客运产品设计优化与创新[J].理论学习与探索,2017,1:52-54.
    [11]史峰,周文梁,陈彦,等.基于弹性需求的旅客列车开行方案优化研究[J].铁道学报,2008,30(3):1-6.
    [12]史峰,李彦霖,胡心磊,等.面向服务水平的高速铁路列车开行方案优化[J].中国铁道科学,2018,39(5):127-136.
    [13]刘云,刘富,侯涛,等.优化核参数的模糊C均值聚类算法[J].吉林大学学报:工学版,2016,46(1):246-251.
    [14]王振武,何关瑶.核函数选择方法研究[J].湖南大学学报:自然科学版,2018,45(10):155-160.
    [15]XIE X L,BEINI G.A validity method for fuzzy clustering[C].IEEE Trans on Pattern Analysis and Machine Intelligence,1991,13(8):841-847.
    [16]ZAHID N,LIMOURI M,ESSAID A.A new cluster validity for fuzzy clustering[M].Pattern Recognition,1999,32(7):1089-1097.