基于复合分布的混合交通时间间隔模型

摘要/Abstract

摘要：

非复合分布模型可用于分析交通流量达1 800 vph的车辆时间间隔，但并不适用于更高交通流量的情况.为解决此类问题，提出了一些基于复合分布的模型.但这类模型的参数标定过程复杂，在一定程度上限制了其应用.针对流量介于1 900 vph到4 100 vph的车辆时间间隔，本文分别采用5种复合分布模型进行分析，即指数-极值分布(EEV)、对数正态-极值分布(LEV)、威布尔-极值分布(WEV)、威布尔-对数正态分布(WLN)和指数-对数正态分布(ELN).然后采用两种方法进行拟合优度检验——基于累计函数分布检验（CDF）和双样本（Cramervon Mises）&K样本（AndersonDarling）检验.结果表明，在分析车辆时间间隔方面，威布尔-极值分布(WEV)是最佳的复合分布模型，在Cramervon Mises检验和K样本AndersonDarling检验中均具有良好的一致性.

关键词: 城市交通, 混合交通, 车辆时间间隔, AD检验, Cramervon Mises检验

Abstract:

The timegap data have been modeled through noncomposite distribution up to a flow level of 1 800 vph. It has been found that these models are not capable of modeling time gap data at higher flow levels. Some composite distributions have been proposed to overcome this problem. But, due to the fact that the calibration of model parameters used in composite distributions is tedious, there use may be relatively limited. In this paper, five mixture models namely Exponential+Extreme-value (EEV), Lognormal+Extreme-value (LEV), Weibull +Extreme-value (WEV), Weibull+Lognormal (WLN) and Exponential+ Lognormal (ELN) have been used to model time gap data for flows ranging from 1 900 vph to 4 100 vph. Two types of goodnessoffit tests namely cumulative distribution function (CDF) based and twosample (Cramervon Mises test) & Ksample (AndersonDarling test) based tests were performed. Among all the five models, Weibull+ Extreme Value was found to be the best mixture model for modeling time gap data as it performed consistently well in Cramervon Mises test and Ksample AndersonDarling test.

Key words: urban traffic, mixed traffic, vehicular time gap, AD test, Cramervon Mises test

中图分类号:

U491

DUBEY Subodh Kant，PONNU Balaji，ARKATKAR Shriniwas S. 基于复合分布的混合交通时间间隔模型[J]. 交通运输系统工程与信息, 2013, 13(3): 91-98.

DUBEY Subodh Kant，PONNU Balaji，ARKATKAR Shriniwas S. Time Gap Modeling Using Mixture Distributions under Mixed Traffic Conditions[J]. Journal of Transportation Systems Engineering and Information Technology, 2013, 13(3): 91-98.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/abstract/abstract18587.shtml

http://www.tseit.org.cn/CN/Y2013/V13/I3/91

参考文献

[1]Kanagaraj V, Asaithambi G, Srinivasan K, et al. Vehicle class wise analysis of time gaps and headways under heterogeneous traffic[C]. Proc. 90th Annual Meeting of Transportation Research Board, Washington D C, 2011. 

[2]Arasan V T, Koshy R Z. Headway distribution of heterogeneous traffic on urban arterials[J]. J. Inst. Eng. (India), Part AG, 2003(84): 210-215. 

[3]Chandra S, Kumar R. Headway modelling under mixed traffic on urban roads[J]. Road & Transport Research, ARRB, Australia, 2001(10): 61-71. 

[4]Cooray K, Ananda M M A. Modeling actuarial data with a composite lognormal-Pareto model[J]. Scandinavian Actuarial Journal, 2005(5): 321-334.

[5]Scollnik D P M. On composite lognormalPareto models[J]. Scandinavian Actuarial Journal, 2007(1): 20-33.

[6]Buckley D J. A semiPoisson model of traffic flow[J]. Transportation Science, 1968(2): 107-133. 

[7]Dawson R F, Chimni L A. The hyperlang probability distribution—A generalized traffic headway model[C]. HRR 230, Highway Research Board, National Research Council, Washington D C, 1968：1-14.

[8]Tolle J E. The lognormal headway distribution model[J]. Traffic Engineering Control, 1971(13): 22-24. 

[9]Cowan R J. Useful headway models[J]. Transportation Research, 1975(9): 371-375.

[10]Wasielewski P. Car following headways on freeways interrupted by the semiPoisson headway distribution model[J]. Transportation Science, 1979(13): 36-55.

[11]Chishaki T, Youichi T. Headway distribution models based on the distinction between leaders and followers[C]. Proceeding of 9th International Symposium of Traffic and Transportation Theory, UNU Science, 1984, The Netherlands：43-63. 

[12]Hagring O. Calibration of headway distributions[C]. Proceedings of the 9th MiniEuro conference, 1997, Montenegro (Yugoslavia). 

[13]Zhang G, Wang Y, Wei H. Examining headway distribution models with urban freeway loop event data[J]. Transportation Research Record 1999, 2007：141-149. 

[14]Dempster A P, Laird N M, Rubin D B. Maximum likelihood from incomplete data via the EM algorithm[J]. Journal of the Royal Statistical Society. Series B (Methodological), 1977, 39(1): 1-38.

[15]Stephens M A. Use of the KolmogorovSmirnov, CramerVon Mises and related statistics without extensive tables[J]. Journal of the Royal Statistical Society Series B (Methodological), 1970(32): 115-122.

[16]Xiao Y, Gordon A. A C++ program for the Cramervon Mises twosample test[J]. Journal of Statistical Software, 2007(17): 1-15. 

[17]Anderson T W, Darling D A. Asymptotic theory of certain goodness of fit criteria based on stochastic processes[J]. Annals of Mathematical Statistics, 1952(23): 193-212. 

[18]Scholz F W, Stephens M A. Ksample AndersonDarling tests of fit, for continuous and discrete cases[R]. Technical Report No. 81, Department of statistics, University of Washington, 1986. 

[19]Dubey S K, Ponnu B, Arkatkar S S. Time gap modeling under mixed traffic conditions: a statistical analysis[J]. Journal of Transportation System Engineering and Information Technology, 2012, 12(6): 21-25.