- Physical Review Journals
All JournalsPhysics Magazine
Access Provided by Xinjiang University

Physical Review E

Export citation

Export citation

Choose format for download:

Download Citation
  • Access by Xinjiang University

Phase transitions in traffic flow on multilane roads

Boris S. Kerner1,* and Sergey L. Klenov2,†

  • 1GR/PTF, HPC, G021, Daimler AG, 71059 Sindelfingen, Germany
  • 2Department of Physics, Moscow Institute of Physics and Technology, 141700 Dolgoprudny, Moscow Region, Russia

  • *boris.kerner@daimler.com
  • †s-klenov@mtu-net.ru
PDF

Phys. Rev. E 80, 056101 – Published 4 November, 2009Erratum Phys. Rev. E 81, 069901 (2010)

DOI: https://doi.org/10.1103/PhysRevE.80.056101

Abstract

Based on empirical and numerical analyses of vehicular traffic, the physics of spatiotemporal phase transitions in traffic flow on multilane roads is revealed. The complex dynamics of moving jams observed in single vehicle data measured by video cameras on American highways is explained by the nucleation-interruption effect in synchronized flow, i.e., the spontaneous nucleation of a narrow moving jam with the subsequent jam dissolution. We find that (i) lane changing, vehicle merging from on-ramps, and vehicle leaving to off-ramps result in different traffic phases—free flow, synchronized flow, and wide moving jams—occurring and coexisting in different road lanes as well as in diverse phase transitions between the traffic phases; (ii) in synchronized flow, the phase transitions are responsible for a non-regular moving jam dynamics that explains measured single vehicle data: moving jams emerge and dissolve randomly at various road locations in different lanes; (iii) the phase transitions result also in diverse expanded general congested patterns occurring at closely located bottlenecks.

View figure in article

Erratum

Erratum: Phase transitions in traffic flow on multilane roads [Phys. Rev. E 80, 056101 (2009)]

Boris S. Kerner and Sergey L. Klenov
Phys. Rev. E 81, 069901 (2010)

Article Text

References (41)

  1. R. Herman, E. W. Montroll, R. B. Potts, and R. W. Rothery, Oper. Res. 7, 86 (1959).
  2. D. C. Gazis, R. Herman, and R. W. Rothery, Oper. Res. 9, 545 (1961).
  3. E. Kometani and T. Sasaki, J. Oper. Res. Soc. Jpn. 2, 11 (1958); Oper. Res. 7, 704 (1959).
  4. A. D. May, Traffic Flow Fundamentals (Prentice-Hall, Inc., New Jersey, 1990); M. Cremer, Der Verkehrsfluss auf Schnellstrassen (Springer, Berlin, 1979); W. Leutzbach, Introduction to the Theory of Traffic Flow (Springer, Berlin, 1988); Traffic Flow Theory, edited by N. H. Gartner, C. J. Messer, and A. Rathi (Transportation Research Board, Washington, D.C. 2001); D. C. Gazis, Traffic Theory (Springer, Berlin, 2002); D. Chowdhury, L. Santen, A. Schadschneider, Phys. Rep. 329, 199 (2000); D. Helbing, Rev. Mod. Phys. 73, 1067 (2001); T. Nagatani, Rep. Prog. Phys. 65, 1331 (2002); K. Nagel, P. Wagner, and R. Woesler, Oper. Res. 51, 681 (2003); R. Mahnke, J. Kaupužs, and I. Lubashevsky, Phys. Rep. 408, 1 (2005); H. Rakha, P. Pasumarthy, and S. Adjerid, Transp. Lett. 1, 95 (2009).
  5. B. S. Kerner and P. Konhäuser, Phys. Rev. E 50, 54 (1994).
  6. B. S. Kerner, Phys. Rev. Lett. 81, 3797 (1998).
  7. B. S. Kerner, The Physics of Traffic (Springer, Berlin, 2004).
  8. B. S. Kerner and S. L. Klenov, J. Phys. A 39, 1775 (2006).
  9. B. S. Kerner, in Encyclopedia of Complexity and System Science, edited by R. A. Meyers (Springer, Berlin, 2009), pp. 9302–9355.
  10. B. S. Kerner, Introduction to Modern Traffic Flow Theory and Control: The Long Road to Three-Phase Traffic Theory (Springer, Berlin, 2009).
  11. B. S. Kerner and S. L. Klenov, J. Phys. A 35, L31 (2002).
  12. B. S. Kerner, S. L. Klenov, and D. E. Wolf, J. Phys. A 35, 9971 (2002).
  13. B. S. Kerner and S. L. Klenov, Phys. Rev. E 68, 036130 (2003).
  14. L. C. Davis, Phys. Rev. E 69, 016108 (2004).
  15. H. K. Lee, R. Barlović, M. Schreckenberg, and D. Kim, Phys. Rev. Lett. 92, 238702 (2004).
  16. R. Jiang and Q.-S. Wu, J. Phys. A 37, 8197 (2004).
  17. Kun Gao, Rui Jiang, Shou-Xin Hu, Bing-Hong Wang, and Qing-Song Wu, Phys. Rev. E 76, 026105 (2007).
  18. J. A. Laval, in Traffic and Granular Flow’05, edited by A. Schadschneider, T. Pöschel, R. Kühne, M. Schreckenberg, and D. E. Wolf (Springer, Berlin, 2007), pp. 521–526.
  19. S. Hoogendoorn, H. van Lint, and V. L. Knoop, Transp. Res. Rec. 2088, 102 (2008).
  20. L. C. Davis, Physica A 368, 541 (2006); 361, 606 (2006); 379, 274 (2007); R. Jiang, M.-B. Hua, R. Wang, and Q.-S. Wu, Phys. Lett. A 365, 6 (2007); R. Jiang and Q.-S. Wu, Phys. Rev. E 72, 067103 (2005); Physica A 377, 633 (2007); R. Wang, R. Jiang, Q.-S. Wu, and M. Liu, ibid. 378, 475 (2007); A. Pottmeier, C. Thiemann, A. Schadschneider, and M. Schreckenberg, in Traffic and Granular Flow’05, edited by A. Schadschneider, T. Pöschel, R. Kühne, M. Schreckenberg, and D. E. Wolf (Springer, Berlin, 2007), pp. 503–508; X. G. Li, Z. Y. Gao, K. P. Li, and X. M, Zhao, Phys. Rev. E 76, 016110 (2007); J. J. Wu, H. J. Sun, and Z. Y. Gao, ibid. 78, 036103 (2008).
  21. We can expect that the physical theory of phase transitions occurring in vehicular traffic on multilane roads presented in the paper can be a basis for the further development of new traffic flow models and some interesting transportation applications. However, methodological aspects of traffic flow modeling and possible applications that can follow from the physical theory presented here are out of the scope of this paper and will be considered elsewhere.
  22. B. S. Kerner, Transp. Res. Rec. 1678, 160 (1999); Phys. World 12(8), 25 (1999).
  23. See formulas (16.31) and (16.32) of Ref. [7]: at Sn+1=0 speed fluctuations ξn=0.
  24. Because traffic flow at bottlenecks is usually nonhomogeneous and in the continuum model there are model fluctuations in nonhomogeneous traffic flow, the discrete model exhibits qualitatively the same simulation results of phase transitions at bottlenecks on multilane road as those derived with the continuum model.
  25. Therefore, this overacceleration model could be recommended for general use.
  26. Next Generation Simulation Programs (NGSIM), http://ngsim.camsys.com/
  27. S. Ahn and M. J. Cassidy, in Transportation and Traffic Theory, edited by R. E. Allsop, M. G. H. Bell, and B. G. Hydecker (Elsevier, Amsterdam, 2007), pp. 691–710.
  28. B. S. Kerner, Phys. Rev. E 65, 046138 (2002).
  29. The latter has been proven in simulations made that show qualitatively the same physical effects as presented below, when only one on-ramp and one off-ramp on a road section are used.
  30. A more detailed data analysis and simulations of possible effects of a cascade of lane changing between three or more lanes is out of scope of this paper and will be published elsewhere.
  31. The qualitative difference between the MSP and wide moving jam is that an MSP is caught at a bottleneck, when the MSP reaches the bottleneck at which traffic breakdown is possible (the definition [S]) [7,28]. In contrast, the wide moving jam propagates through this bottleneck while maintaining the mean velocity of the downstream jam front (the definition [J]).
  32. B. S. Kerner, S. L. Klenov, and A. Hiller, J. Phys. A 39, 2001 (2006); B. S. Kerner, S. L. Klenov, A. Hiller, and H. Rehborn, Phys. Rev. E 73, 046107 (2006).
  33. Empirical data shown in Figs. 10(b) and 10(c) have been explained in [34,35] by a “boomerang effect,” i.e., “growing perturbations on a homogeneous freeway section without on- and off-ramps” [34] leading to an F→J transition without bottlenecks (the origin of the axis for road locations was chosen differently in [34,35] and in Fig. 10: x=0 km in Fig. 10 is related to x=465.1 km of related figures of [34,35]). The empirical and numerical analyses presented in Sec. VI C show that the conclusion of [34,35] that the boomerang effect is the reason for wide moving jam emergence in Fig. 10 is invalid. In a more detail, criticisms of studies of traffic congestion made in [34,35] can be found in Sec. 10.3 of [10].
  34. M. Schönhof and D. Helbing, Transp. Sc. 41, 135 (2007).
  35. M. Schönhof and D. Helbing, Transp. Res., Part B: Methodol. 43, 784 (2009).
  36. Recall that the microscopic structure of a wide moving jam consists of a spatiotemporal combination of flow interruption intervals that satisfy (15) and moving blanks, i.e., regions without vehicles moving upstream within the jam; the moving blanks are associated with vehicle motion at a very low speed within the jam [32].
  37. B. S. Kerner. J. Phys. A: Math. Theor. 41, 215101 (2008); . 41, 369801 (2008).
  38. The averaging time interval Tav for the average flow rate q(cong) is assumed to be considerably longer than time distances between any moving jams within a congested pattern, for example, Tav≳30 min.
  39. B. S. Kerner, in Encyclopedia of Complexity and System Science, edited by R. A. Meyers (Springer, Berlin, 2009), pp. 9355–9411.
  40. S. Krauss, P. Wagner, and C. Gawron, Phys. Rev. E 55, 5597 (1997); S. Krauss, Ph.D. thesis, University of Cologne, 1998.
  41. P. G. Gipps, Transp. Res., Part B: Methodol. 15, 105 (1981).

Outline

Information

Phys. Rev. E 80, 056101– Published 4 November, 2009
  • Received 27 May 2009

© 2009 American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation