pith. sign in

arxiv: 2606.26868 · v1 · pith:KATJN2DJnew · submitted 2026-06-25 · ⚛️ physics.soc-ph · nlin.AO

Human adaptive variability stabilises collective traffic dynamics

Shirui Zhou , Ching Jin , Junfang Tian , Shiteng Zheng , Rui Jiang , Shiquan Zhong , Shoufeng Ma , Vittorio Loreto This is my paper

Pith reviewed 2026-06-26 02:49 UTC · model grok-4.3

classification ⚛️ physics.soc-ph nlin.AO
keywords traffic dynamicscar-followingadaptive variabilitytime headwaynonlinear dampingadaptive cruise controlstop-and-go wavescollective stability
0
0 comments X

The pith

Human drivers' speed-dependent headway adjustments generate nonlinear damping that stabilizes traffic flow unlike rigid automated control.

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

The paper demonstrates that human car-following behavior involves continuously reshaping time-headway distributions in a non-monotonic way as speeds change. This variability produces a damping effect that suppresses the spread of local perturbations into larger traffic waves. Large-scale experiments show human platoons dissipate disturbances while commercial adaptive cruise control systems amplify them, leading to higher fuel consumption and emissions. The central mechanism is the shift from efficiency-oriented to risk-sensitive regulation at different speeds. If correct, this means human variability serves as a stabilizing feature in collective dynamics rather than a flaw to eliminate.

Core claim

Human car-following does not follow a fixed proportional spacing rule. Drivers continuously reshape their time-headway distributions across speed regimes, exhibiting a non-monotonic shift from efficiency-oriented to risk-sensitive regulation. This speed-dependent variability generates nonlinear damping that suppresses the synchronisation and propagation of local errors, in contrast to rigid algorithmic controllers that amplify small perturbations into severe stop-and-go waves.

What carries the argument

Speed-dependent non-monotonic reshaping of time-headway distributions that generates nonlinear damping of traffic disturbances.

If this is right

  • Commercial rule-based ACC controllers amplify local perturbations into stop-and-go waves.
  • Human-driven platoons progressively dissipate disturbances and maintain smoother flow.
  • Production ACC systems increase fuel consumption and carbon emissions by 2.7- to 5.0-fold compared to human driving.
  • Robust interactive systems benefit from embedding adaptive behavioral flexibility instead of rigid uniformity.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar adaptive variability could stabilize other large-scale AI systems involving human-AI interaction.
  • Traffic models that assume fixed rules may underestimate stability from behavioral shifts.
  • Future experiments could test mixed fleets of human and automated vehicles to observe interaction dynamics.

Load-bearing premise

The controlled experiments and calibrated simulations accurately attribute the dissipation of disturbances specifically to the non-monotonic time-headway variability rather than other unmeasured factors.

What would settle it

An experiment where drivers are forced to maintain fixed time-headways across speeds and then checking whether their platoons show amplified waves similar to those in ACC simulations.

Figures

Figures reproduced from arXiv: 2606.26868 by Ching Jin, Junfang Tian, Rui Jiang, Shiquan Zhong, Shirui Zhou, Shiteng Zheng, Shoufeng Ma, Vittorio Loreto.

Figure 1
Figure 1. Figure 1: Conceptual illustration of two contrasting modes of perturbation propagation in car-following. (a) In one mode, a small disturbance introduced by an upstream vehicle propagates backwards through the traffic stream, opposite to the direction of vehicle motion. As successive followers respond, the disturbance becomes increasingly organised: some vehicles slow sharply or stop, while others accelerate to close… view at source ↗
Figure 2
Figure 2. Figure 2: Human platoons dissipate perturbations; an ACC-model platoon amplifies them into stop-and-go waves. (a) Empirical space–time trajectories from the 25-vehicle human-driven platoon (Hefei experiment, Run 14, vlead,mean ≈ 39 km h−1 ; see Supplementary Section S1 for run-selection criteria). Individual fluctuations remain incoherent and diffuse, allowing the collective system to attenuate perturbations and mai… view at source ↗
Figure 3
Figure 3. Figure 3: Microscopic turning in the speed–spacing relation is a robust feature of human driving across independent datasets. (a) Empirical lower and upper bounds of the speed–spacing relation for human-driven vehicles and commercially implemented ACC. Under human driving, the upper boundary exhibits a clear turning point: spacing increases gradually at low speeds but rises more sharply beyond a critical speed. Unde… view at source ↗
Figure 4
Figure 4. Figure 4: Ex-Gaussian time-headway statistics explain the segmented speed–spacing boundary in human driving. (a) Empirical time-headway distributions for three representative speed bins from D2 (9–16, 16–23, and 23–34 km h−1 ). Hollow circles are empirical bin counts; solid curves are the corresponding maximum-likelihood ex-Gaussian fits37, 44, 45. Across all three speed regimes the ex-Gaussian form closely captures… view at source ↗
read the original abstract

Automated systems are often designed on the assumption that replacing human behavioural variability with precise, uniform algorithmic control improves collective performance. In automotive traffic, this principle underlies commercial adaptive cruise control (ACC). Using two large-scale human-driving experiments comprising 2.95 million car-following observations, a 25-vehicle platoon experiment and a controlled 11-driver protocol, cross-validated with 0.77 million observations from the NGSIM dataset and data from 22 production ACC systems, together with empirically calibrated ACC simulations, we show the opposite: rigid algorithmic uniformity creates systemic fragility. Commercial rule-based controllers amplify small local perturbations into severe stop-and-go waves, increasing fuel consumption and carbon emissions by approximately 2.7- to 5.0-fold across scenarios. Human-driven platoons, by contrast, progressively dissipate disturbances and maintain smoother flow. We identify the behavioural mechanism behind this advantage: human car-following does not follow a fixed proportional spacing rule. Drivers continuously reshape their time-headway distributions across speed regimes, exhibiting a non-monotonic shift from efficiency-oriented to risk-sensitive regulation. This speed-dependent variability generates nonlinear damping that suppresses the synchronisation and propagation of local errors. Our findings challenge the view that human variability is merely suboptimal noise to be eliminated. More broadly, they suggest that robust large-scale interactive AI systems should embed adaptive, human-inspired behavioural flexibility rather than rely on rigid uniformity.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper claims that human drivers adaptively reshape time-headway distributions in a non-monotonic, speed-dependent manner, generating nonlinear damping that dissipates local perturbations and stabilizes collective traffic flow. In contrast, commercial ACC systems with rigid control amplify disturbances into stop-and-go waves. This is supported by 2.95 million car-following observations from two human-driving experiments (including a 25-vehicle platoon and 11-driver protocol), cross-validated against 0.77 million NGSIM observations and 22 production ACC systems, plus empirically calibrated ACC simulations.

Significance. If the central causal attribution holds, the result challenges the design premise of many automated driving systems and has implications for modeling interactive AI systems more broadly. Strengths include the large observation counts, NGSIM cross-validation, and use of production ACC data for comparison. The work provides empirical evidence against the assumption that eliminating behavioral variability improves collective performance.

major comments (2)
  1. [Abstract and experimental design (25-vehicle platoon, 11-driver protocol)] The central claim attributes disturbance dissipation specifically to the non-monotonic shift in time-headway distributions. However, the 25-vehicle platoon and controlled 11-driver protocols (described in the abstract and experimental sections) do not include an explicit counterfactual comparison holding other factors fixed while varying only the time-headway distribution; without this, the isolation from confounders such as attention, anticipation, or vehicle dynamics remains untested.
  2. [ACC simulation calibration and comparison sections] The empirically calibrated ACC simulations are used to contrast with human behavior, yet the calibration parameters (listed among free parameters) are not shown to avoid inadvertently reproducing the very adaptive variability under test; this affects the strength of the claim that rigid uniformity creates fragility.
minor comments (2)
  1. [Abstract] The abstract provides no details on statistical controls, data exclusion criteria, or potential selection effects in the platoon experiments, which would strengthen the presentation of the large observation counts.
  2. [Behavioral mechanism description] Notation for time-headway distributions and the precise definition of the non-monotonic shift should be clarified with an equation or figure reference to aid reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major comment below with clarifications on our design and planned revisions.

read point-by-point responses

  1. Referee: [Abstract and experimental design (25-vehicle platoon, 11-driver protocol)] The central claim attributes disturbance dissipation specifically to the non-monotonic shift in time-headway distributions. However, the 25-vehicle platoon and controlled 11-driver protocols (described in the abstract and experimental sections) do not include an explicit counterfactual comparison holding other factors fixed while varying only the time-headway distribution; without this, the isolation from confounders such as attention, anticipation, or vehicle dynamics remains untested.

    Authors: We acknowledge that the experiments do not provide an explicit counterfactual isolating only the time-headway distribution. The 25-vehicle platoon and 11-driver protocol instead compare collective outcomes under human driving (with observed adaptive variability) against production ACC systems and calibrated simulations under matched conditions. Cross-validation with 0.77 million NGSIM observations further supports consistency of the non-monotonic pattern. We will add an explicit limitations paragraph discussing potential remaining confounders such as anticipation and attention in the revised manuscript. revision: partial

  2. Referee: [ACC simulation calibration and comparison sections] The empirically calibrated ACC simulations are used to contrast with human behavior, yet the calibration parameters (listed among free parameters) are not shown to avoid inadvertently reproducing the very adaptive variability under test; this affects the strength of the claim that rigid uniformity creates fragility.

    Authors: We will include the complete calibration parameter values, objective function, and fitting procedure in the supplementary information of the revised manuscript. These parameters are fixed and non-adaptive by construction, drawn from production ACC data without embedding speed-dependent time-headway reshaping. revision: yes

Circularity Check

0 steps flagged

No circularity; claims rest on direct experiments and simulations

full rationale

The paper advances its claims through two large-scale human-driving experiments (2.95 million observations), a 25-vehicle platoon test, an 11-driver controlled protocol, NGSIM cross-validation (0.77 million observations), data from 22 production ACC systems, and empirically calibrated ACC simulations. No derivation chain, equation, or self-citation reduces the reported stabilization effect or the non-monotonic time-headway mechanism to a fitted input or prior result by construction. The analysis is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim is empirical and rests primarily on the validity of the experimental data and the attribution of stabilization to the identified behavioral mechanism; calibration of ACC simulations introduces fitted elements but no new postulated entities.

free parameters (1)
  • ACC simulation calibration parameters
    The abstract states that simulations are empirically calibrated to 22 production ACC systems to enable comparison.
axioms (1)
  • domain assumption The observed differences between human and ACC behavior are attributable to time-headway variability rather than confounding factors in the experimental setups.
    This assumption underpins the identification of the behavioral mechanism as the source of nonlinear damping.

pith-pipeline@v0.9.1-grok · 5794 in / 1292 out tokens · 103885 ms · 2026-06-26T02:49:15.589781+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

50 extracted references · 36 canonical work pages

  1. [1]

    Traffic and related self-driven many-particle systems.Rev

    Helbing, D. Traffic and related self-driven many-particle systems.Rev. Mod. Phys.73, 1067–1141, DOI: 10.1103/ RevModPhys.73.1067 (2001)

    2001

  2. [2]

    Shladover, S. E. Connected and automated vehicle sys- tems: Introduction and overview.J. Intell. Transp. Syst. 22, 190–200, DOI: 10.1080/15472450.2017.1336053 (2018)

    work page doi:10.1080/15472450.2017.1336053 2017

  3. [3]

    K., Chien, C

    Swaroop, D., Hedrick, J. K., Chien, C. C. & Ioannou, P. A comparison of spacing and headway control laws for automatically controlled vehicles.Veh. Syst. Dyn.23, 597–625 (1994)

    1994

  4. [4]

    & Rajagopal, K

    Darbha, S. & Rajagopal, K. R. Intelligent cruise con- trol systems and traffic flow stability.Transp. Res. Part C: Emerg. Technol.7, 329–352, DOI: 10.1016/ S0968-090X(99)00024-8 (1999)

    1999

  5. [5]

    Wilson, R. E. & Ward, J. A. Car-following models: fifty years of linear stability analysis - a mathemati- cal perspective.Transp. Plan. Technol.34, 3–18, DOI: 10.1080/03081060.2011.530826 (2011)

    work page doi:10.1080/03081060.2011.530826 2011

  6. [6]

    Phys.10, 033001, DOI: 10.1088/1367-2630/10/3/033001 (2008)

    Sugiyama, Y .et al.Traffic jams without bottle- necks—experimental evidence for the physical mecha- nism of the formation of a jam.New J. Phys.10, 033001, DOI: 10.1088/1367-2630/10/3/033001 (2008)

    work page doi:10.1088/1367-2630/10/3/033001 2008

  7. [7]

    Simon, H. A. The architecture of complexity.Proc. Am. Philos. Soc.106, 467–482 (1962)

    1962

  8. [8]

    E.Diversity and Complexity(Princeton Univer- sity Press, Princeton, NJ, 2010)

    Page, S. E.Diversity and Complexity(Princeton Univer- sity Press, Princeton, NJ, 2010)

    2010

  9. [9]

    Anderson, P. W. More is different.Science177, 393–396, DOI: 10.1126/science.177.4047.393 (1972)

    work page doi:10.1126/science.177.4047.393 1972

  10. [10]

    L., Kiss, I

    Zhang, Y ., Ocampo-Espindola, J. L., Kiss, I. Z. & Mot- ter, A. E. Random heterogeneity outperforms design in network synchronization.Proc. Natl. Acad. Sci.118, e2024299118, DOI: 10.1073/pnas.2024299118 (2021)

    work page doi:10.1073/pnas.2024299118 2021

  11. [11]

    & Page, S

    Hong, L. & Page, S. E. Groups of diverse problem solvers can outperform groups of high-ability problem solvers. Proc. Natl. Acad. Sci.101, 16385–16389, DOI: 10.1073/ pnas.0403723101 (2004)

    2004

  12. [12]

    A.et al.The importance of cognitive diversity for sustaining the commons.Nat

    Baggio, J. A.et al.The importance of cognitive diversity for sustaining the commons.Nat. Commun.10, 875, DOI: 10.1038/s41467-019-08549-8 (2019)

    work page doi:10.1038/s41467-019-08549-8 2019

  13. [13]

    P., Liu, Y .-Y

    Meng, Y ., Cornelius, S. P., Liu, Y .-Y . & Li, A. Dy- namics of collective cooperation under personalised strat- egy updates.Nat. Commun.15, 3125, DOI: 10.1038/ s41467-024-47380-8 (2024)

    2024

  14. [14]

    Gunter, G.et al.Are commercially implemented adaptive cruise control systems string stable?IEEE Transactions on Intell. Transp. Syst.22, 6992–7003, DOI: 10.1109/ TITS.2020.3000682 (2021)

    arXiv 2020

  15. [15]

    Makridis, M.et al.Empirical study on the properties of adaptive cruise control systems and their impact on traffic flow and string stability.Transp. Res. Rec.2674, 471–484, DOI: 10.1177/0361198120911047 (2020)

    work page doi:10.1177/0361198120911047 2020

  16. [16]

    & Sun, J

    Ye, Y ., Sun, J. & Sun, J. Car-following characteristics of commercially available adaptive cruise control systems and comparison with human drivers.Transp. Res. Rec. 2677, 1401–1414, DOI: 10.1177/03611981221113313 (2023). 12/14

    work page doi:10.1177/03611981221113313 2023

  17. [17]

    & Ciuffo, B

    Makridis, M., Mattas, K., Anesiadou, A. & Ciuffo, B. Openacc: An open database of car-following experiments to study the properties of commercial acc systems.Transp. Res. Part C: Emerg. Technol.125, 103047, DOI: 10.1016/ j.trc.2021.103047 (2021)

    arXiv 2021

  18. [18]

    Ciuffo, B.et al.Requiem on the positive effects of com- mercial adaptive cruise control on motorway traffic and recommendations for future automated driving systems. Transp. Res. Part C: Emerg. Technol.130, 103305, DOI: 10.1016/j.trc.2021.103305 (2021)

    work page doi:10.1016/j.trc.2021.103305 2021

  19. [19]

    E.et al.Dissipation of stop-and-go waves via control of autonomous vehicles: Field experiments

    Stern, R. E.et al.Dissipation of stop-and-go waves via control of autonomous vehicles: Field experiments. Transp. Res. Part C: Emerg. Technol.89, 205–221, DOI: 10.1016/j.trc.2018.02.005 (2018)

    work page doi:10.1016/j.trc.2018.02.005 2018

  20. [20]

    Shi, X. & Li, X. Empirical study on car-following charac- teristics of commercial automated vehicles with different headway settings.Transp. Res. Part C: Emerg. Technol. 128, 103134, DOI: 10.1016/j.trc.2021.103134 (2021)

    work page doi:10.1016/j.trc.2021.103134 2021

  21. [21]

    & Zhang, Y

    Tang, S., Zou, Y ., Wu, S., Xie, Y . & Zhang, Y . Comparing car-following behavior patterns of human-driven vehicles and autonomous vehicles in a mixed traffic environment. IEEE Transactions on Intell. Transp. Syst.26, 6814–6830, DOI: 10.1109/TITS.2025.3539757 (2025)

    work page doi:10.1109/tits.2025.3539757 2025

  22. [22]

    & Imprialou, M

    Papadoulis, A., Quddus, M. & Imprialou, M. Evaluating the safety impact of connected and autonomous vehicles on motorways.Accid. Analysis & Prev.124, 12–22, DOI: 10.1016/j.aap.2018.12.019 (2019)

    work page doi:10.1016/j.aap.2018.12.019 2018

  23. [23]

    & Ding, S

    Abdel-Aty, M. & Ding, S. A matched case-control analysis of autonomous vs human-driven vehicle ac- cidents.Nat. Commun.15, 4931, DOI: 10.1038/ s41467-024-48526-4 (2024)

    2024

  24. [24]

    Jiang, R.et al.Traffic experiment reveals the nature of car-following.PLOS ONE9, e94351, DOI: 10.1371/ journal.pone.0094351 (2014)

    2014

  25. [25]

    Jiang, R.et al.On some experimental features of car- following behavior and how to model them.Transp. Res. Part B: Methodol.80, 338–354, DOI: 10.1016/j.trb.2015. 08.003 (2015)

    work page doi:10.1016/j.trb.2015 2015

  26. [26]

    & Watts, D

    Mason, W. & Watts, D. J. Collaborative learning in networks.Proc. Natl. Acad. Sci.109, 764–769, DOI: 10.1073/pnas.1110069108 (2012)

    work page doi:10.1073/pnas.1110069108 2012

  27. [27]

    Rahwan, I.et al.Machine behaviour.Nature568, 477– 486, DOI: 10.1038/s41586-019-1138-y (2019)

    work page doi:10.1038/s41586-019-1138-y 2019

  28. [28]

    D., Shklarsh, A

    Karpas, E. D., Shklarsh, A. & Schneidman, E. Informa- tion socialtaxis and efficient collective behavior emerging in groups of information-seeking agents.Proc. Natl. Acad. Sci.114, 5589–5594, DOI: 10.1073/pnas.1618055114 (2017). 29.Koopman, P. & Wagner, M. Autonomous vehicle safety: An interdisciplinary challenge.IEEE Intell. Transp. Syst. Mag.9, 90–96, D...

    work page doi:10.1073/pnas.1618055114 2017

  29. [29]

    & Caulfield, B

    Rezaei, A. & Caulfield, B. Safety of autonomous vehicles: What are the insights from experienced industry profes- sionals?Transp. Res. Part F: Traffic Psychol. Behav.81, 472–489, DOI: 10.1016/j.trf.2021.07.005 (2021)

    work page doi:10.1016/j.trf.2021.07.005 2021

  30. [30]

    Ibanez-Guzman, J. & Li, Y . Lidar and cameras in au- tonomous driving.Nat. Rev. Electr. Eng.1–2, DOI: 10.1038/s44287-025-00176-4 (2025)

    work page doi:10.1038/s44287-025-00176-4 2025

  31. [31]

    & Ciuffo, B

    He, Y ., Montanino, M., Mattas, K., Punzo, V . & Ciuffo, B. Physics-augmented models to simulate commercial adaptive cruise control (acc) systems.Transp. Res. Part C: Emerg. Technol.139, 103692, DOI: 10.1016/j.trc.2022. 103692 (2022)

    work page doi:10.1016/j.trc.2022 2022

  32. [32]

    & Di Francesco, L

    Cascetta, E., Cartenì, A. & Di Francesco, L. Do au- tonomous vehicles drive like humans? a turing ap- proach and an application to sae automation level 2 cars. Transp. Res. Part C: Emerg. Technol.134, 103499, DOI: 10.1016/j.trc.2021.103499 (2022)

    work page doi:10.1016/j.trc.2021.103499 2021

  33. [33]

    Gunter, G., Janssen, C., Barbour, W., Stern, R. E. & Work, D. B. Model-based string stability of adaptive cruise control systems using field data.IEEE Transactions on Intell. Veh.5, 90–99, DOI: 10.1109/TIV .2019.2955368 (2020)

    work page doi:10.1109/tiv 2019

  34. [34]

    Tian, J.et al.On the role of speed adaptation and spac- ing indifference in traffic instability: Evidence from car- following experiments and its stochastic model.Transp. Res. Part B: Methodol.129, 334–350, DOI: 10.1016/j.trb. 2019.09.014 (2019)

    work page doi:10.1016/j.trb 2019

  35. [35]

    & Xie, Y

    Li, T., Chen, D., Zhou, H., Laval, J. & Xie, Y . Car- following behavior characteristics of adaptive cruise con- trol vehicles based on empirical experiments.Transp. Res. Part B: Methodol.147, 67–91, DOI: 10.1016/j.trb.2021. 03.003 (2021)

    work page doi:10.1016/j.trb.2021 2021

  36. [36]

    Dey, P. P. & Chandra, S. Desired time gap and time headway in steady-state car-following on two-lane roads. J. Transp. Eng.135, 687–693, DOI: 10.1061/(ASCE) 0733-947X(2009)135:10(687) (2009)

    work page doi:10.1061/(asce 2009

  37. [37]

    R.et al.Human-agent collectives.Commun

    Jennings, N. R.et al.Human-agent collectives.Commun. ACM57, 80–88, DOI: 10.1145/2629559 (2014)

    work page doi:10.1145/2629559 2014

  38. [38]

    Dafoe, A.et al.Cooperative ai: machines must learn to find common ground.Nature593, 33–36, DOI: 10.1038/ d41586-021-01170-0 (2021)

    2021

  39. [39]

    Jiang, R.et al.Experimental and empirical investiga- tions of traffic flow instability.Transp. Res. Part C: Emerg. Technol.94, 83–98, DOI: 10.1016/j.trc.2017.08. 024 (2018)

    work page doi:10.1016/j.trc.2017.08 2017

  40. [40]

    Jabari, S. E. & Liu, H. X. A stochastic model of traf- fic flow: Theoretical foundations.Transp. Res. Part B: Methodol.46, 156–174, DOI: 10.1016/j.trb.2011.09.006 (2012)

    work page doi:10.1016/j.trb.2011.09.006 2011

  41. [41]

    Tian, J.et al.Car following behavioral stochasticity analysis and modeling: Perspective from wave travel 13/14 time.Transp. Res. Part B-Methodological143, 160–176, DOI: 10.1016/j.trb.2020.11.008 (2021)

    work page doi:10.1016/j.trb.2020.11.008 2020

  42. [42]

    Sci.51, 807–822, DOI: 10.1287/trsc.2015.0667 (2017)

    Tian, J.et al.Cellular automaton model with dynamical 2d speed-gap relation.Transp. Sci.51, 807–822, DOI: 10.1287/trsc.2015.0667 (2017)

    work page doi:10.1287/trsc.2015.0667 2015

  43. [43]

    & Cousineau, D

    Lacouture, Y . & Cousineau, D. How to use matlab to fit the ex-gaussian and other probability functions to a distribution of response times.Tutorials Quant. Methods for Psychol.4, 35–45 (2008)

    2008

  44. [44]

    Models of single lane time headway distri- butions.Transp

    Branston, D. Models of single lane time headway distri- butions.Transp. Sci.10, 125–148, DOI: 10.1287/trsc.10. 2.125 (1976)

    work page doi:10.1287/trsc.10 1976

  45. [45]

    He, Y .et al.The energy impact of adaptive cruise con- trol in real-world highway multiple-car-following sce- narios.Eur. Transp. Res. Rev.12, 17, DOI: 10.1186/ s12544-020-00406-w (2020)

    2020

  46. [46]

    A., Kouvelas, A

    Apostolakis, T., Makridis, M. A., Kouvelas, A. & Am- pountolas, K. Energy-based assessment and driving be- havior of ACC systems and humans inside platoons.IEEE Transactions on Intell. Transp. Syst.24, 12726–12735, DOI: 10.1109/TITS.2023.3285296 (2023)

    work page doi:10.1109/tits.2023.3285296 2023

  47. [47]

    A., Toth, C

    Laval, J. A., Toth, C. S. & Zhou, Y . A parsimonious model for the formation of oscillations in car-following models.Transp. Res. Part B: Methodol.70, 228–238, DOI: 10.1016/j.trb.2014.09.004 (2014)

    work page doi:10.1016/j.trb.2014.09.004 2014

  48. [48]

    & Gao, Z.-Y

    Bouadi, M., Jia, B., Jiang, R., Li, X. & Gao, Z.-Y . Stochastic factors and string stability of traffic flow: An- alytical investigation and numerical study based on car- following models.Transp. Res. Part B: Methodol.165, 96–122, DOI: 10.1016/j.trb.2022.09.007 (2022)

    work page doi:10.1016/j.trb.2022.09.007 2022

  49. [49]

    & Kesting, A

    Treiber, M. & Kesting, A. The intelligent driver model with stochasticity—new insights into traffic flow oscilla- tions.Transp. Res. Part B: Methodol.117, 613–623, DOI: 10.1016/j.trb.2017.08.012 (2018)

    work page doi:10.1016/j.trb.2017.08.012 2017

  50. [50]

    A., Zheng, Z

    Chen, D., Laval, J. A., Zheng, Z. & Ahn, S. A behav- ioral car-following model that captures traffic oscilla- tions.Transp. Res. Part B: Methodol.46, 744–761, DOI: 10.1016/j.trb.2012.01.009 (2012). Data availability The NGSIM trajectory data are publicly available from the US Federal Highway Administration (FHWA) at https: //ops.fhwa.dot.gov/trafficanalys...

    work page doi:10.1016/j.trb.2012.01.009 2012