Pith. sign in

REVIEW 3 major objections 5 minor 63 references

A car-following framework for traffic instability and lane changes

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read There is a vehicle-density-dependent critical reaction time in Newell's car-following model: perturbations decay below it, grow above it, and the threshold approaches Newell's classical stability boundary in the dense limit.

desk verdict The finite-N stability result for Newell's model is real and clean; the lane-changing model needs calibration and a fixed arithmetic error before its behavioral claims can be trusted. read the letter →

arxiv 2501.01988 v1 pith:SQ426HOW submitted 2024-12-30 math.OC

Nicholas Mankowski , Hassan Mushtaq , Hanliang Guo This is my paper · ORCID
classification math.OC MSC 90B2034K20
keywords car-followingmodeltrafficstabilitycriticalreactiontimelanechangebehaviorfrustrationleveldelaydifferentialequationsphantomjamloadbalancing
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper establishes that Newell's first-order car-following model on a periodic ring has a vehicle-density-dependent critical reaction time: perturbations to the uniform equilibrium decay when the reaction time $\Delta$ is below the threshold and grow when it is above the threshold, with high reaction times eventually producing a collision. The threshold decreases as vehicle density rises, and in the dense limit it converges to Newell's classical continuum stability boundary $\lambda\Delta=1/2$, so the result is a finite-$N$ correction to the 1961 criterion. The paper also builds a stochastic lane-change model in which a driver's frustration level $\varphi$, updated by four rules, maps to a lane-change-attempt probability $P(\varphi)=\frac{2}{\pi}\arctan(\varphi)$; this reproduces load balancing between lanes and shows that an aggressive driver changes lanes about four times as often as a control driver for less than a 2 percent velocity gain. A careful reader would care because the stability half gives a quantitative prediction for when phantom jams should form, and the lane-change half offers a mechanism connecting driver psychology to observable traffic patterns, even though that mechanism is not empirically calibrated.

What carries the argument

The load-bearing object is the Jacobian $J=c(I+A+B)$ of the linearized delay-differential equation $\dot{y}(t)=J y(t-\Delta)$, where $c=-\lambda \exp(-\frac{\lambda}{V}(\frac{L}{N}-d))$ and the matrices $A$ and $B$ encode the nearest-neighbor interaction on a periodic ring. The paper proves that $J$ is diagonalizable by showing that the normalized matrix has $N-1$ distinct eigenvalues $1-e^{2\pi i k/N}$, so the characteristic equation $\det(\tilde{\lambda}I - J e^{-\tilde{\lambda}\Delta})=0$ factors into a product of scalar equations. That factorization makes the stability boundary computable for arbitrary $N$ from the largest real part of the characteristic roots, and it is the mechanism through which density enters the critical reaction time. The lane-changing mechanism is carried by the frustration level $\varphi$, which rises when the adjacent lane has a larger headway, falls when it has a smaller headway, jumps by $p$ when the driver is passed, and resets to zero after a lane change; the probability of attempting a lane change in the next second is $P(\varphi)=\frac{2}{\pi}\arctan(\varphi)$, rescaled per simulation time step.

What would settle it

Compute the largest real part of the characteristic roots of the factored equation from Section 3.1 for several $N$ and $\Delta$: the central claim predicts the zero crossing decreases monotonically with $N$ and converges to $1/(2\lambda)=0.5$ s as $N\to L/d$, so a non-monotone threshold or a different dense limit would refute it. A ring-road experiment with controlled density and driver reaction times could test the same boundary by checking whether phantom jams appear only above the predicted density-dependent cutoff.

Watch

Extended reading notes

Core claim

The central claim is that single-lane traffic has a well-defined stability boundary in reaction time: for $N$ identical vehicles with headway $L/N$ on a ring of length $L$, the equilibrium is stable for $\Delta < \tau(N)$ and unstable for $\Delta > \tau(N)$, where $\tau(N)$ decreases with $N$ and approaches $1/(2\lambda)$ as $N\to L/d$. The paper proves this analytically by diagonalizing the Jacobian of the linearized delay-differential system and computing the characteristic roots, and confirms it numerically by measuring the cyclic growth rate of velocity oscillations; for $N=50$, $L=1000$ m and $\lambda=1$/s, both methods give $\tau\approx 0.7$ s, while the dense limit gives $0.5$ s. In the unstable regime the initial small perturbation grows into backward-propagating congestion waves and eventually a collision. For the multi-lane extension, the paper claims that the frustration-driven stochastic lane-change rule $P(\varphi)=\frac{2}{\pi}\arctan(\varphi)$ makes an initially imbalanced two-lane road load-balance automatically, and that an aggressive driver (doubled $\lambda$) executes about four times as many lane changes as a control driver while gaining less than 2 percent in driving distance.

Load-bearing premise

The lane-changing results stand on the assumption that real drivers' frustration follows the paper's four ad hoc rules and the arctangent probability mapping, with the rates $r$ and $p$ chosen by hand; the paper offers no empirical calibration for these rules, so if real lane-change decisions are not governed by them, the load-balancing, 2-percent-benefit, and fourfold-cost conclusions would not necessarily hold, even though the single-lane stability threshold would survive.

Editorial extensions

If this is right

  • For a fixed number of vehicles $N$, there is a sharp reaction-time cutoff; small disturbances decay below it, and above it they amplify into backward-moving congestion waves that can end in a collision.
  • The cutoff falls as density rises, so denser traffic requires faster reactions; in the continuum limit $N\to L/d$ it recovers Newell's $\lambda\Delta=1/2$ condition exactly.
  • The frustration-based lane-change rule makes an initially empty lane fill automatically: the occupancy difference drops from 50 cars to about 5 and total flow rises toward the two-lane optimum.
  • An 'aggressive' driver, modeled by doubling $\lambda$, changes lanes about four times as often as a control driver yet gains less than 2 percent in distance, so frequent discretionary lane changes provide minimal velocity benefit in homogeneous dense traffic.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper leaves implicit that the single-lane threshold gives a direct, quantitative target for ring-road experiments on phantom jams: if measured jam onset does not track the predicted density-dependent reaction-time cutoff, the first-order Newell model itself would need revision.
  • The frustration model is not calibrated, but it makes a testable prediction of its own: the per-second lane-change attempt rate should be a saturating function of the time spent in worse driving conditions and should reset after each successful change, which simulator or survey data could confirm or reject.
  • Because the stability and lane-change modules are independent, the 2-percent benefit and fourfold lane-change cost should be read as properties of the proposed frustration dynamics rather than of Newell's car-following model; a different psychological rule could reverse those conclusions.
  • The paper's speculation that aggressive drivers may gain more when traffic forms isolated slow 'packs' can be tested directly by introducing heterogeneous maximal velocities into the same simulation; that would show whether the near-zero-sum result is general or a uniform-density artifact.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper studies a deterministic Newell-type car-following model on a ring and a stochastic frustration-based lane-changing extension. For single-lane traffic, it linearizes the delay differential equation, proves that the Jacobian is diagonalizable, reduces the characteristic equation to a product over eigenvalues, and numerically identifies a density-dependent critical reaction time. The paper reports that this critical time decreases with vehicle density and approaches Newell's lambda*Delta = 1/2 criterion in the dense limit, with nonlinear simulations agreeing on tau about 0.7 s for N = 50. For multi-lane traffic, the paper introduces four frustration-update rules and an arctangent map from frustration to lane-change attempt probability, and uses simulations to claim load-balancing between lanes, a fourfold increase in lane changes for an 'aggressive' driver, and a less than 2% velocity benefit of aggressive driving.

Significance. If fully supported, the single-lane stability result is a useful finite-N extension of Newell's continuum criterion, with a clean analytic derivation and a consistent numerical confirmation; the diagonalization of the Jacobian is a nice technical contribution that makes the delay-equation root-finding tractable. The lane-changing mechanism is conceptually novel in adding a psychological state variable, but its current value is illustrative rather than predictive because the frustration dynamics and its parameters are not calibrated. The paper would be strengthened by explicit sensitivity analysis and by correcting the flow-rate arithmetic in the multi-lane section.

major comments (3)
  1. [Section 4.2] The quoted equilibrium flow at 25 veh/km is arithmetically incorrect: substituting rho = 25 veh/km into Eq (1) with Table 1 gives v_infinity = 40 - 40 exp(-(1/40)(40 - 7.5)) m/s, approximately 22.25 m/s, and hence q = rho * v_infinity, approximately 2002 veh/h, not 1668 veh/h. In addition, the simulation configuration has N = 50 vehicles on a 1000 m ring, so the per-lane density after balancing is 25 veh/km only if the load split is exactly equal; the measured per-lane flow of about 1800 veh/h is below both this theoretical value and the model's maximum q* = 2065 veh/h at rho* = 34 veh/km. The sentence claiming that the mechanism 'maximizes the flow rate' is therefore not supported by the model's own fundamental diagram.
  2. [Section 2 and Section 5] The four frustration rules and the mapping P(phi) = (2/pi) arctan(phi) are introduced without empirical calibration, and the Discussion explicitly concedes that r and p 'can be better calibrated' with future experiments. As a result, the quantitative multi-lane claims - the roughly 30 s load-balancing timescale, the fourfold lane-change cost, and the less than 2% velocity benefit - are functions of the chosen parameters and functional forms rather than tested predictions. The paper should either provide a sensitivity analysis over r, p, and the choice of P, or explicitly re-label these results as demonstrations of model behavior rather than faithful reproductions.
  3. [Section 2] The lane-change success criterion requires no vehicle in [x_j - d, x_j + d] in the adjacent lane, a fixed spatial gap that is independent of the vehicles' speeds and of the reaction time Delta. Since velocities in this model range from 0 to V = 40 m/s, the same 7.5 m gap is treated as safe at both low and high speeds, which can distort lane-change frequencies and the aggressive-driver comparison. Please justify this criterion or test whether the main conclusions survive with a speed-dependent safe-gap rule.
minor comments (5)
  1. [Throughout] The text contains frequent spacing artifacts in words such as 'tra ffic' and 'di fferent' (for example, in the Abstract and throughout Section 1).
  2. [Section 4.2] The phrase 'the upper-right panel of Figure 6' refers to a configuration not shown; Figure 6 is a two-panel line plot without an upper-right panel.
  3. [General] No code or data availability statement is included; providing the simulation code would strengthen reproducibility of the multi-lane results.
  4. [Section 3.1] The symbol lambda is used both for the model parameter in Eq (1) and for the characteristic roots in Eq (6); although lambda-tilde is introduced for the latter, the notation could be clarified for readers.
  5. [Section 4.1] The paper reports growth rates k for Delta = 0 and 0.5 s as -1.073 and -0.790, but it does not state how many oscillation periods were used for the exponential fit; please specify the fitting procedure.

Circularity Check

1 steps flagged · score 4.0 of 10

Secondary load-balancing claim reduces to the frustration-rule definition; the central stability derivation is independent and sound.

  1. self definitional [Section 2 (frustration rules 1–2 and Eq. (2)); Section 4.2 (load-balancing experiment)]
    "1. dϕ/dt = r if the driver’s headway in the current lane is smaller than the perceived headway in an adjacent lane. 2. dϕ/dt =−r if the driver’s headway in the current lane is larger than the perceived headway in an adjacent lane. ... Evidently, ∆N gradually reduces from 50 to about 5 within about 30 seconds and keeps oscillating around 5. The gradual decrease of ∆N recovers what we observe in the real-world, as an empty lane will soon be occupied by as many vehicles as in the adjacent lane to make both lanes “symmetric”."

    Frustration is defined, by rule 1/2, to increase exactly when the current lane has less headway than the adjacent lane and to decrease when it has more headway; the mapping P(ϕ) is then chosen to be monotonically increasing in ϕ. Therefore the model encodes at the individual level: “move from the lower-headway lane toward the higher-headway lane.” The load-balancing observation—an initially empty lane fills until lane occupancies roughly equalize—is the aggregate restatement of that same encoded rule, not an emergent independent prediction.

full rationale

The paper’s primary derivation—the density-dependent critical reaction time τ(N)—is self-contained and not circular. It linearizes Newell’s first-order car-following equation (1), diagonalizes the Jacobian, solves the characteristic equation (8), and compares the resulting stability boundary with direct simulation; the limiting agreement with Newell’s λΔ=1/2 criterion is checked against an external 1961/1962 result, not imported as an assumption. No parameter is fitted to the stability data, and no self-citation is load-bearing. The multi-lane section is different: the “frustration” rules define frustration to rise when the current lane’s headway is smaller and fall when it is larger, and P(ϕ) is chosen monotonically increasing, so the load-balancing demonstration in Section 4.2 reduces to the model’s own definition: drivers are, by construction, more likely to attempt moves into the lane with more headway, and the empty lane fills. This is a designed-in corollary rather than an independently reproduced phenomenon. The remaining lane-change claims (2% speed benefit, fourfold lane-change cost) are outputs of the arbitrary, uncalibrated r and p choices and are better assessed as modeling limitations than as circular steps; the paper itself flags that r and p “can be better calibrated.” A separate flow-rate arithmetic issue (1668 veh/h vs. Eq. (1) at 25 veh/km) is a correctness concern, not a circularity. Overall, only the secondary load-balancing “reproduction” is circular, so the score is moderate, not high.

Assumptions & free parameters 3 free parameters · 6 assumptions · 1 invented entities

The stability result (critical reaction time) is derived from the model with no fitted parameters beyond the physical constants in Table 1, so its circularity burden is low. The lane-changing results rest on several ad hoc choices (r, p, the arctan mapping, the gap rule) that are not empirically grounded, so the behavioral predictions carry a substantial free-parameter load.

free parameters (3)
  • r (frustration increase/decrease rate) = 0.1 or 0.2
    Chosen by hand in Section 4.2; sets how quickly frustration builds or decays and thus the rate of lane-change attempts. Not calibrated to any driver data.
  • p (frustration increment when passed) = 0.1 or 0.2
    Ad hoc discrete jump in frustration when another vehicle overtakes; directly changes lane-change probability. No empirical basis.
  • Aggressive driver lambda = 2.0 /s vs 1.0 /s
    Used in Section 4.2 to define an aggressive driver; arbitrary choice, not tied to measurements.
assumptions (6)
  • domain assumption Newell's exponential speed-headway relation (Eq. 1) governs vehicle motion in all simulations and analyses.
    The framework is built entirely on this first-order car-following law from Ref. [13]; no independent validation is provided for real traffic.
  • domain assumption Periodic ring boundary with identical vehicles and drivers for the single-lane stability analysis.
    The linear stability and simulations use a ring (primary cell plus image cells); real highways have open boundaries and heterogeneous drivers, which may change the critical reaction time.
  • ad hoc to paper The four frustration update rules (rules 1-4 in Section 2).
    Proposed without empirical calibration; the paper states no best mapping exists without data.
  • ad hoc to paper P(phi) = (2/pi) arctan(phi) maps frustration to lane-change attempt probability.
    Chosen only to satisfy monotonicity, boundedness, and saturation; no empirical support.
  • ad hoc to paper Lane-change success requires no vehicle in [x_j - d, x_j + d] in the adjacent lane.
    Simple gap-acceptance rule; no safety margin beyond the minimum headway d, and no validation.
  • standard math Standard characteristic-equation theory for delay differential equations (Smith [57]).
    Used to derive the stability determinant in Section 3.1; standard and accepted.
invented entities (1)
  • Frustration level phi(t)
    purpose: Latent psychological variable that grows in the worse lane, shrinks in the better lane, resets on lane change, and drives stochastic lane-change attempts via P(phi).
    No empirical measurement, calibration, or falsifiable prediction is provided; the paper suggests future fMRI/survey work but presents no data. It is an ad hoc construct, so announcement of this entity is not backed by independent evidence.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A car-following framework for traffic instability and lane changes." pith.science (2026). https://pith.science/paper/SQ426HOW

@misc{pith2026250101988,
  author       = {Pith},
  title        = {Pith review of: A car-following framework for traffic instability and lane changes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SQ426HOW}},
  note         = {Machine review of arXiv:2501.01988}
}
read the original abstract

This paper develops a computational framework based on a car-following model to study traffic instability and lane changes. Building upon Newell's classical first-order car-following model, we show that, both analytically and numerically, there exists a vehicle-density-dependent critical reaction time that determines the stability of single-lane traffic. Specifically, perturbations to the equilibrium system decay with time for low reaction time and grow for high reaction time. This critical reaction time converges to Newell's original result in the continuum limit. Additionally, we propose a psychology-based lane-changing mechanism that builds a quantitative connection between the driver's psychological factor (frustration level) and the driving condition. We show that our stochastic lane-changing model can faithfully reproduce interesting phenomena like load-balancing of different lanes. Our model supports the result that more frequent lane changes only marginally benefit the driver's overall velocity.

Figures

Figures reproduced from arXiv: 2501.01988 by the authors.

Figure 1
Figure 1. (a) The driving velocity ˙x j as a function of headway hj ≡ x j+1 − x j given by (1). V is the maximal velocity, d is the minimal headway for a non-zero velocity, and λ is the rate of change of the velocity at hj = d. In the equilibrium state where all vehicles have the same headway h∞, the equilibrium velocity of each vehicle v∞ can be found directly via (1). (b) The equilibrium flow rate q as a function of vehicle… view at source ↗
Figure 2
Figure 2. Schematic figure of our car-following model. Each rectangle denotes a vehicle whose position [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Flowchart of the numerical method. The steps are grouped into Lane-changing, Collision-detecting, and Forward-moving stages, denoted [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Dynamics of 50 vehicles in a single lane with various reaction times [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Stability of a single-lane traffic system. (a) The cyclic growth rate of velocity difference k of the non-linear simulations and the maximal real part of the characteristic roots of the linearized system as functions of reaction time ∆. (b) The critical reaction time τ…
Figure 6
Figure 6. Figure 6: The absolute difference between the numbers of vehicles in each lane ∆N and the flow rate Q(t, L/2) as functions of time. The curves are the mean value of 10 Monte Carlo simulations. The averaging time window for flow rate calculation is δ = 5s. All vehicles are initia…
Figure 6
Figure 6. Figure 6: Evidently, ∆N gradually reduces from 50 to about 5 within about 30 seconds and keeps oscillating around 5. The gradual decrease of ∆N recovers what we observe in the real-world, as an empty lane will soon be occupied by as many vehicles as in the adjacent lane to make …
Figure 7
Figure 7. Figure 7: (a) Number of lane-changes ∆l of the aggressive driver (λ = 2) and a control driver (λ = 1) over time. Curve and the error bars shows the mean and the standard deviation of 20 Monte Carlo simulations. (b) Driving distances of the aggressive driver and a control driver …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

63 extracted references · 62 canonical work pages

  1. [1]

    B. D. Greenshields, J. R. Bibbins, W. Channing, H. H. Miller, A study of traffic capacity, in: Highway research board proceedings, volume 14, Washington, DC, 1935, pp. 448–477

    work page 1935

  2. [2]

    M. J. Lighthill, G. B. Whitham, On kinematic waves ii. a theory of tra ffic flow on long crowded roads, Proceedings of the royal society of london. series a. mathematical and physical sciences 229 (1955) 317–345

    work page 1955

  3. [3]

    P. I. Richards, Shock waves on the highway, Operations research 4 (1956) 42–51

    1956

  4. [4]

    R. E. Chandler, R. Herman, E. W. Montroll, Tra ffic dynamics: studies in car following, Operations research 6 (1958) 165–184

    work page 1958

  5. [5]

    Helbing, Tra ffic and related self-driven many-particle systems, Reviews of modern physics 73 (2001) 1067

    D. Helbing, Tra ffic and related self-driven many-particle systems, Reviews of modern physics 73 (2001) 1067

    work page 2001

  6. [6]

    Treiber, A

    M. Treiber, A. Kesting, Tra ffic flow dynamics, Traffic Flow Dynamics: Data, Models and Simulation, Springer-Verlag Berlin Heidelberg 227 (2013) 228

    work page 2013

  7. [7]

    D. M. Levinson, K. J. Krizek, Planning for place and plexus: Metropolitan land use and transport, Routledge, 2007

    work page 2007

  8. [8]

    Nagel, M

    K. Nagel, M. Schreckenberg, A cellular automaton model for freeway tra ffic, Journal de physique I 2 (1992) 2221–2229

    work page 1992

Show all 63 references
  1. [9]

    Toledo, Driving behaviour: models and challenges, Transport Reviews 27 (2007) 65–84

    T. Toledo, Driving behaviour: models and challenges, Transport Reviews 27 (2007) 65–84

    2007

  2. [10]

    Saberi, H

    M. Saberi, H. Hamedmoghadam, M. Ashfaq, S. A. Hosseini, Z. Gu, S. Shafiei, D. J. Nair, V . Dixit, L. Gardner, S. T. Waller, et al., A simple contagion process describes spreading of traffic jams in urban networks, Nature communications 11 (2020) 1616

    2020

  3. [11]

    Amb ¨uhl, M

    L. Amb ¨uhl, M. Menendez, M. C. Gonz ´alez, Understanding congestion propagation by combining percolation theory with the macroscopic fundamental diagram, Communications Physics 6 (2023) 26

    2023

  4. [12]

    Kometani, Dynamic behavior of tra ffic with a nonlinear spacing-speed relationship, Theory of Traffic Flow (Proc

    E. Kometani, Dynamic behavior of tra ffic with a nonlinear spacing-speed relationship, Theory of Traffic Flow (Proc. of Sym. on TTF (GM)) (1959) 105–119

    1959

  5. [13]

    G. F. Newell, Nonlinear e ffects in the dynamics of car following, Operations research 9 (1961) 209–229

    1961

  6. [14]

    Treiterer, J

    J. Treiterer, J. Myers, The hysteresis phenomenon in tra ffic flow, Transportation and traffic theory 6 (1974) 13–38

    1974

  7. [15]

    G. A. Bekey, G. O. Burnham, J. Seo, Control theoretic models of human drivers in car following, Human Factors 19 (1977) 399–413

    1977

  8. [16]

    Aron, Car following in an urban network: simulation and experiments, Planning and Transport Research and Computation (1988)

    M. Aron, Car following in an urban network: simulation and experiments, Planning and Transport Research and Computation (1988)

    1988

  9. [17]

    P. G. Gipps, A behavioural car-following model for computer simulation, Transportation research part B: methodological 15 (1981) 105–111

    1981

  10. [18]

    P. G. Gipps, A model for the structure of lane-changing decisions, Transportation Research Part B: Methodological 20 (1986) 403–414

    1986

  11. [19]

    Ozaki, Reaction and anticipation in the car following behavior, in: Proceedings of the 13-th International Symposium on Tra ffic and Transportation Theory, 1993, 1993

    H. Ozaki, Reaction and anticipation in the car following behavior, in: Proceedings of the 13-th International Symposium on Tra ffic and Transportation Theory, 1993, 1993

    1993

  12. [20]

    Q. I. Yang, H. N. Koutsopoulos, A microscopic tra ffic simulator for evaluation of dynamic tra ffic management systems, Transportation Research Part C: Emerging Technologies 4 (1996) 113–129

    1996

  13. [21]

    D. Chen, J. Laval, Z. Zheng, S. Ahn, A behavioral car-following model that captures tra ffic oscillations, Transportation research part B: methodological 46 (2012) 744–761

    2012

  14. [22]

    B. S. Kerner, H. Rehborn, Experimental properties of complexity in tra ffic flow, Physical Review E 53 (1996) R4275

    1996

  15. [23]

    B. S. Kerner, Experimental features of self-organization in tra ffic flow, Physical review letters 81 (1998) 3797

    1998

  16. [24]

    Bando, K

    M. Bando, K. Hasebe, A. Nakayama, A. Shibata, Y . Sugiyama, Structure stability of congestion in tra ffic dynamics, Japan Journal of Industrial and Applied Mathematics 11 (1994) 203–223

    1994

  17. [25]

    Bando, K

    M. Bando, K. Hasebe, A. Nakayama, A. Shibata, Y . Sugiyama, Dynamical model of tra ffic congestion and numerical simulation, Physical review E 51 (1995) 1035

    1995

  18. [26]

    Holland, A generalised stability criterion for motorway tra ffic, Transportation Research Part B: Methodological 32 (1998) 141–154

    E. Holland, A generalised stability criterion for motorway tra ffic, Transportation Research Part B: Methodological 32 (1998) 141–154

    1998

  19. [27]

    B. S. Kerner, Introduction to modern tra ffic flow theory and control: the long road to three-phase traffic theory, Springer Science & Business Media, 2009

    2009

  20. [28]

    R. E. Wilson, Mechanisms for spatio-temporal pattern formation in highway tra ffic models, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 366 (2008) 2017–2032

    2008

  21. [29]

    Seibold, M

    B. Seibold, M. R. Flynn, A. R. Kasimov, R. R. Rosales, Constructing set-valued fundamental diagrams from jamiton solutions in second order traffic models, arXiv preprint arXiv:1204.5510 (2012)

    2012 arXiv

  22. [30]

    G. F. Newell, Theories of instability in dense highway tra ffic, J. Operations Research Society of Japan 5 (1962) 9–54. 13

    1962

  23. [31]

    Kesting, M

    A. Kesting, M. Treiber, How reaction time, update time, and adaptation time influence the stability of tra ffic flow, Computer-Aided Civil and Infrastructure Engineering 23 (2008) 125–137

    2008

  24. [32]

    J. A. Laval, C. F. Daganzo, Lane-changing in tra ffic streams, Transportation Research Part B: Methodological 40 (2006) 251–264

    2006

  25. [33]

    J. A. Laval, C. S. Toth, Y . Zhou, A parsimonious model for the formation of oscillations in car-following models, Transportation Research Part B: Methodological 70 (2014) 228–238

    2014

  26. [34]

    Shang, F

    M. Shang, F. Hauer, R. Stern, Do cut-ins matter: Assessing the impact of lane changing and string stability on tra ffic flow, in: 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), IEEE, 2020, pp. 1–6

    2020

  27. [35]

    Chauhan, V

    P. Chauhan, V . Kanagaraj, G. Asaithambi, Understanding the mechanism of lane changing process and dynamics using microscopic tra ffic data, Physica A: Statistical Mechanics and its Applications 593 (2022) 126981

    2022

  28. [36]

    Y . Gao, D. Levinson, Lane changing and congestion are mutually reinforcing?, Communications in Transportation Research 3 (2023) 100101

    2023

  29. [37]

    Zhang, J

    K. Zhang, J. Rong, Y . Gao, Y . Chen, E ffects of lane imbalance on capacity drop and emission in expressway merging areas: A simulation analysis, Sustainability 16 (2024) 10388

    2024

  30. [38]

    Hidas, Modelling lane changing and merging in microscopic tra ffic simulation, Transportation Research Part C: Emerging Technologies 10 (2002) 351–371

    P. Hidas, Modelling lane changing and merging in microscopic tra ffic simulation, Transportation Research Part C: Emerging Technologies 10 (2002) 351–371

    2002

  31. [39]

    Hidas, Modelling vehicle interactions in microscopic simulation of merging and weaving, Transportation Research Part C: Emerging Technologies 13 (2005) 37–62

    P. Hidas, Modelling vehicle interactions in microscopic simulation of merging and weaving, Transportation Research Part C: Emerging Technologies 13 (2005) 37–62

    2005

  32. [40]

    Kesting, M

    A. Kesting, M. Treiber, D. Helbing, General lane-changing model mobil for car-following models, Transportation Research Record 1999 (2007) 86–94

    2007

  33. [41]

    Ahmed, M

    K. Ahmed, M. Ben-Akiva, H. Koutsopoulos, R. Mishalani, Models of freeway lane changing and gap acceptance behavior, Transportation and traffic theory 13 (1996) 501–515

    1996

  34. [42]

    K. I. Ahmed, Modeling drivers’ acceleration and lane changing behavior, Ph.D. thesis, Massachusetts Institute of Technology, 1999

    1999

  35. [43]

    Toledo, H

    T. Toledo, H. N. Koutsopoulos, M. E. Ben-Akiva, Modeling integrated lane-changing behavior, Transportation Research Record 1857 (2003) 30–38

    2003

  36. [44]

    D. J. Sun, L. Elefteriadou, Lane-changing behavior on urban streets: A focus group-based study, Applied ergonomics 42 (2011) 682–691

    2011

  37. [45]

    in-vehicle

    D. Sun, L. Elefteriadou, Lane-changing behavior on urban streets: An “in-vehicle” field experiment-based study, Computer-Aided Civil and Infrastructure Engineering 27 (2012) 525–542

    2012

  38. [46]

    Nagatani, Self-organization and phase transition in tra ffic-flow model of a two-lane roadway, Journal of Physics A: Mathematical and General 26 (1993) L781

    T. Nagatani, Self-organization and phase transition in tra ffic-flow model of a two-lane roadway, Journal of Physics A: Mathematical and General 26 (1993) L781

    1993

  39. [47]

    Rickert, K

    M. Rickert, K. Nagel, M. Schreckenberg, A. Latour, Two lane tra ffic simulations using cellular automata, Physica A: Statistical Mechanics and its Applications 231 (1996) 534–550

    1996

  40. [48]

    Wagner, K

    P. Wagner, K. Nagel, D. E. Wolf, Realistic multi-lane traffic rules for cellular automata, Physica A: Statistical Mechanics and its Applications 234 (1997) 687–698

    1997

  41. [49]

    Nagel, D

    K. Nagel, D. E. Wolf, P. Wagner, P. Simon, Two-lane traffic rules for cellular automata: A systematic approach, Physical Review E 58 (1998) 1425

    1998

  42. [50]

    Maerivoet, B

    S. Maerivoet, B. De Moor, Cellular automata models of road tra ffic, Physics reports 419 (2005) 1–64

    2005

  43. [51]

    Worrall, A

    R. Worrall, A. Bullen, Y . Gur, An elementary stochastic model of lane-changing on a multilane highway, Highway Research Record (1970)

    1970

  44. [52]

    Pentland, A

    A. Pentland, A. Liu, Modeling and prediction of human behavior, Neural computation 11 (1999) 229–242

    1999

  45. [53]

    J.-B. Sheu, S. G. Ritchie, Stochastic modeling and real-time prediction of vehicular lane-changing behavior, Transportation Research Part B: Methodological 35 (2001) 695–716

    2001

  46. [54]

    Toledo, R

    T. Toledo, R. Katz, State dependence in lane-changing models, Transportation research record 2124 (2009) 81–88

    2009

  47. [55]

    Singh, B

    K. Singh, B. Li, Estimation of tra ffic densities for multilane roadways using a markov model approach, IEEE Transactions on industrial electronics 59 (2011) 4369–4376

    2011

  48. [56]

    Zheng, Recent developments and research needs in modeling lane changing, Transportation research part B: methodological 60 (2014) 16–32

    Z. Zheng, Recent developments and research needs in modeling lane changing, Transportation research part B: methodological 60 (2014) 16–32

    2014

  49. [57]

    H. L. Smith, An introduction to delay di fferential equations with applications to the life sciences, volume 57, springer New York, 2011

    2011

  50. [58]

    B. S. Kerner, The physics of tra ffic, Physics World 12 (1999) 25

    1999

  51. [59]

    D. Chen, S. Ahn, J. Laval, Z. Zheng, On the periodicity of tra ffic oscillations and capacity drop: The role of driver characteristics, Trans- portation research part B: methodological 59 (2014) 117–136

    2014

  52. [60]

    James, Road rage and aggressive driving: Steering clear of highway warfare, Prometheus Books, 2009

    L. James, Road rage and aggressive driving: Steering clear of highway warfare, Prometheus Books, 2009

    2009

  53. [61]

    Zheng, S

    Z. Zheng, S. Ahn, C. M. Monsere, Impact of tra ffic oscillations on freeway crash occurrences, Accident Analysis & Prevention 42 (2010) 626–636

    2010

  54. [62]

    Zheng, S

    Z. Zheng, S. Ahn, D. Chen, J. Laval, The e ffects of lane-changing on the immediate follower: Anticipation, relaxation, and change in driver characteristics, Transportation research part C: emerging technologies 26 (2013) 367–379

    2013

  55. [63]

    J. O. Linke, S. P. Haller, E. P. Xu, L. T. Nguyen, A. E. Chue, C. Botz-Zapp, O. Revzina, S. Perlstein, A. J. Ross, W.-L. Tseng, et al., Persistent frustration-induced reconfigurations of brain networks predict individual differences in irritability, Journal of the American Aca...

    2023

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.