[1] Sun Lu, Zhang Huimin, Gao Rong, et al. Gaussian mixture models for clustering and classifying trafficflow in real-time for traffic operation and management [J]. Journal of Southeast University (English Edition), 2011, 27 (2): 174-179. [doi:10.3969/j.issn.1003-7985.2011.02.012]

Copy

Gaussian mixture models for clustering and classifying trafficflow in real-time for traffic operation and management()

用于交通运营管理的实时交通流状态分类高斯混合模型

Share：

Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]

Volumn:

27

Issue:

2011 2

Page:

174-179

Research Field:

Traffic and Transportation Engineering

Publishing date:

2011-06-30

Info

Title:

Gaussian mixture models for clustering and classifying trafficflow in real-time for traffic operation and management

用于交通运营管理的实时交通流状态分类高斯混合模型

Author(s):

Sun Lu^{1, 2}, Zhang Huimin³, Gao Rong⁴, Gu Wenjun¹, Xu Bing¹, Chen Liliang¹

¹School of Transportation, Southeast University, Nanjing 210096, China
²Department of Civil Engineering, The Catholic University of America, Washington DC 20064, USA
³Jinzhong Bureau of Highway Administration of Shanxi Province, Jinzhong 030600, China
⁴Xinzhou Bureau of Highway Administration of Shanxi Province, Xinzhou 034000, China

孙璐^{1, 2}, 张惠民³, 高荣⁴, 顾文钧¹, 徐冰¹, 陈鲤梁¹

¹东南大学交通学院, 南京 210096; ²Department of Civil Engineering, Catholic University of America, Washington DC 20064, USA; ³山西省公路局晋中分局, 晋中 030600; ⁴山西省公路局忻州分局, 忻州 034000

Keywords:

traffic flow patterns;  Gaussian mixture model;  level of service;  data mining;  cluster analysis;  classifier

交通流型;  高斯混合模型;  服务水平;  数据挖掘;  聚类分析;  分类

PACS:

U491

DOI:

10.3969/j.issn.1003-7985.2011.02.012

Abstract:

Based on Gaussian mixture models(GMM), speed, flow and occupancy are used together in the cluster analysis of traffic flow data. Compared with other clustering and sorting techniques, as a structural model, the GMM is suitable for various kinds of traffic flow parameters. Gap statistics and domain knowledge of traffic flow are used to determine a proper number of clusters. The expectation-maximization(E-M)algorithm is used to estimate parameters of the GMM model. The clustered traffic flow patterns are then analyzed statistically and utilized for designing maximum likelihood classifiers for grouping real-time traffic flow data when new observations become available.Clustering analysis and pattern recognition can also be used to cluster and classify dynamic traffic flow patterns for freeway on-ramp and off-ramp weaving sections as well as for other facilities or things involving the concept of level of service, such as airports, parking lots, intersections, interrupted-flow pedestrian facilities, etc.

采用高斯混合模型GMM, 同时以交通流量、平均速度和密度3种交通流宏观特征为指标, 对交通流状态进行聚类和分类.和其他聚类分类方法比较, 高斯混合模型是结构化的模型, 适合于各种情形交通流参数.高斯混合模型中子类的个数通过Gap统计量结合交通流的领域知识加以确定, 而模型的其他参数则由E-M算法进行估计.所建立的GMM模型可以作为实时交通流状态的分类器对新的观察值开展有效的分类识别和预报.同时, 聚类分析和模式识别也可以用来对其他含有服务水平概念的设施进行聚类和分类分析, 比如机场、停车场、交叉口等.

References:

[1] Transportation Research Board. Highway capacity manual [M]. Washington DC, USA: National Research Council, 2001.
[2] Hall F L, Wakefield S, Al-Kaisy A. Freeway quality of service: what really matters to drivers and passengers? [J].Transportation Research Board, 2001(1776):17-23.
[3] Hall F L, Hurdle V F, Banks J H. A synthesis of recent work on the nature of speed-flow and flow-occupancy(or density)relationships on freeways [J]. Transportation Research Record, 1992(1365):12-18.
[4] Helbing D, Hennecke A, Treiber M. Phase diagram of traffic states in the presence of inhomogeneities [J]. Physics Review Letters, 1999, 82(21):4360-4363.
[5] Kerner B S, Rehborn H. Experimental properties of complexity in traffic flow [J]. Physics Review E, 1996, 53(5): 4275-4278.
[6] Chasey A D, de la Garza J M, Drew D R. Comprehensive level of service: needed approach for civil infrastructure systems[J].Journal of Infrastructure Systems, 1997, 3(4):143-153.
[7] Hua J, Faghri A. Dynamic traffic pattern classification using artificial neural networks[J].Transportation Research Record, 1993(1399):14-19.
[8] Mead W C, Fisher H N, Jones R D, et al. Application of adaptive and neural network computational techniques to traffic volume and classification monitoring [J]. Transportation Research Record, 1994(1466):116-123.
[9] Lingras P. Classifying highways: hierarchical grouping versus Kohonen neural networks [J]. Journal of Transportation Engineering, 1995, 121(4):364-368.
[10] Saito M, Fan J. Multilayer artificial neural networks for level-of-service analysis of signalized intersections [J]. Transportation Research Record, 1999(1678):216-224.
[11] Yang H, Qiao F. Neural network approach to classification of traffic flow states [J]. Journal of Transportation Engineering, 1998, 124(6):521-525.
[12] Kikuchi S, Chakroborty P. Ways to treat uncertainty in level of service determination[C]//The 83rd Annual Meeting of Transportation Research Board. Washington DC, USA, 2003.
[13] Mitchell T. Machine learning [M]. McGraw Hill, 1997.
[14] Hastie T, Tibshirani R, Friedman J. The elements of statistical learning: data mining, inference, and prediction [M]. Springer-Verlag, 2001.
[15] Shekar S, Lu C T, Chawla S, et al. Data mining and visualization of twin-cities traffic data, TR 01-015 [R]. Twin Cities, MN, USA: Department of CSE, University of Minnesota, 2000.
[16] Oh C, Ritchie S. Real-time inductive-signature-based level of service for signalized intersections [J]. Transportation Research Record, 2002(1802):97-104.
[17] Klodzinski J, Al-Deek H M. New methodology for defining level of service at toll plazas[J].Journal of Transportation Engineering, ASCE, 2002, 128(2):173-181.
[18] Sun L, Yang J, Mahmassani H, et al. Data mining based adaptive regression for developing equilibrium static traffic speed-density relationships [J]. Canadian Journal of Civil Engineering, 2010, 37(3):389-400.
[19] Hartigan J A. Clustering algorithms [M]. New York: Wiley, 1975.
[20] Hartigan J A, Wong M A. A K-means clustering algorithm [J]. Applied Statistics, 1979, 28:100-108.
[21] Duda R O, Hart P E, Stork D G. Pattern classification [M]. 2nd ed. New York: John Wiley & Sons, Inc., 2001.
[22] Gordon A D. Classification [M]. 2nd ed. London: Chapman & Hall/CRC, 1999.
[23] Rencher A C. Methods of multivariate analysis [M]. John Wiley & Sons, 2002.
[24] Sun L, Zhou J. Development of multiregime speed-density relationships by cluster analysis [J]. Transportation Research Record, 2005(1934): 64-71.
[25] Dempster A P, Laird N M, Rubin D B. Maximum likelihood from incomplete data via the EM algorithm(with discussion)[J]. Journal of the Royal Statistical Society, Series B, 1977, 39(1):1-38.
[26] Witten I H, Frank E. Data mining: practical machine learning tools and techniques [M]. 2nd ed. Morgan Kaufmann, 2005.
[27] StatSoft Inc. Electronic textbook: cluster analysis [EB/OL].(2006)[2010-06-20]. www.statsoft.com/textbook/.
[28] TransGuide Program. The advanced traffic management system(ATMS)at San Antonio[EB/OL].(2006)[2010-06-20]. http://www.transguide.dot.state.tx.us/.
[29] Tibshirani R, Walther G, Hastie T. Estimating the number of clusters in a dataset via the gap statistic[J]. Journal of the Royal Statistical Society, Series B, 2001, 63(2): 411-423.

Memo

Memo:

Biography: Sun Lu(1972—), male, doctor, professor, sunl@cua.edu.
Foundation items: The US National Science Foundation(No.CMMI-0408390, CMMI-0644552), the American Chemical Society Petroleum Research Foundation(No.PRF-44468-G9), the Research Fellowship for International Young Scientists(No.51050110143), the Fok Ying-Tong Education Foundation(No.114024), the Natural Science Foundation of Jiangsu Province(No.BK2009015), the Postdoctoral Science Foundation of Jiangsu Province(No.0901005C).
Citation: Sun Lu, Zhang Huimin, Gao Rong, et al. Gaussian mixture models for clustering and classifying traffic flow in real-time for traffic operation and management[J].Journal of Southeast University(English Edition), 2011, 27(2):174-179.[doi:10.3969/j.issn.1003-7985.2011.02.012]

Common functions