Pith. sign in

REVIEW 4 major objections 6 minor 29 references

A United Framework for Planning Electric Vehicle Charging Accessibility

T0 review · 4 major / 6 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read EV charging station placement can be made equity-aware: the paper claims a 15–20% cut in travel-time variability from MEM-based siting in NYC simulation.

desk verdict The accessibility term is constant under the paper's own constraints, so MEM and NoMEM reduce to the same cost-only p-median; the reported 15-20% improvement has no mechanism in the stated model. read the letter →

arxiv 2508.05827 v1 pith:KFMABSP5 submitted 2025-08-07 eess.SY cs.SY

Tony Kinchen , Panagiotis Typaldos , Andreas A. Malikopoulos This is my paper · ORCID
classification eess.SYcs.SY
keywords EVcharginginfrastructureMobilityEquityMetricstationsitingaccessibilitytraveltimevariabilitySUMOsimulationfacilitylocationurban
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

The paper proposes a unified optimization framework for placing EV charging stations in dense cities, combining traffic simulation, energy consumption modeling, and a mobility equity measure called the Mobility Equity Metric (MEM). It claims this is the first framework to embed MEM directly into station siting, and that in a New York City case study accessibility-aware planning cuts travel-time variability by 15–20% while increasing average travel distance by less than 0.1%. The authors argue that the tunable parameter λ lets planners balance efficiency against equity in a predictable way. If the claim holds, cities with equity mandates could use this method to choose station locations that serve under-resourced neighborhoods without much efficiency loss.

What carries the argument

The central object is the accessibility-weighted objective of Problem 1, specifically the term $\lambda\sum_{i}\sum_{j}\varepsilon_i x_{ij}y_j$, built from the Mobility Index $\varepsilon_i$ defined by the Mobility Equity Metric (MEM). The objective sits inside a binary facility-location problem with constraints (1)–(6): each demand point assigned to exactly one open station, exactly $p$ stations selected, capacity limits respected, and total capacity meeting total demand. In the simulation pipeline, accessibility scores are precomputed, the integer program selects stations, and SUMO simulates the resulting trips; the paper claims $\lambda$ acts as a monotonic control knob for the efficiency

What would settle it

Run the optimization on the same NYC data with $\lambda=0$ and with $\lambda>0$ (e.g., $\lambda=4$) and compare the chosen station sets and objective values. Under the constraints as written, the accessibility term is constant, so the two runs should return identical or equivalent solutions; if they do not, the claimed 15–20% travel-time variability reduction cannot be attributed to the equity term and would need to be traced to the simulation pipeline or assignment procedure instead.

Watch

Extended reading notes

Core claim

The central claim is that accessibility can be engineered into charging infrastructure placement by adding a reward term $\lambda\sum_{i\in I}\sum_{j\in J}\varepsilon_i x_{ij}y_j$ to a facility-location objective. Here $\varepsilon_i$ is a Mobility Index score for demand location $i$, $x_{ij}$ indicates assignment of demand $i$ to station $j$, $y_j$ indicates whether station $j$ is selected, and $\lambda$ weights equity against travel cost. The paper proves feasibility when total station capacity meets total demand and asserts that as $\lambda$ goes from 0 to infinity, the solution interpolates from pure travel-cost minimization to pure accessibility maximization. In SUMO simulations of New

Load-bearing premise

The equity term only matters if choosing one station over another changes the value of $\sum_{i}\sum_{j}\varepsilon_i x_{ij}y_j$; under constraints (1) and (3) every demand point is assigned to exactly one open station, so this sum equals $\sum_i\varepsilon_i$, a constant, and the optimization sees no accessibility signal.

Editorial extensions

If this is right

  • Planners can choose station sets that reduce travel-time variability by 15–20% relative to distance-only siting, based on the NYC simulation.
  • The parameter $\lambda$ gives a principled tuning mechanism; larger values push toward accessibility-oriented placements with longer trips but better battery retention and lower energy consumption.
  • The formulation scales to large urban networks and can be solved with standard integer-programming solvers, making it usable within existing planning tools.
  • For new infrastructure deployments, equity concerns can be incorporated at modest cost compared with retrofitting existing station networks.
  • The study is a proof of concept; installation costs, grid impacts, and richer user behavior are listed by the authors as needed extensions before deployment.

Reading between the lines

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

  • A natural testable variant is to attach the accessibility score to candidate station locations ($\varepsilon_j$) rather than demand points; then the equity term genuinely shifts with station choice, and the claimed 15–20% variability reduction could be checked directly against the demand-level formulation.
  • If the demand-level accessibility term is invariant across feasible station selections, the paper's strongest interpretable evidence is the simulated travel-time distribution under different deployments rather than the station-selection mechanism itself.
  • The same MEM-as-objective idea could be applied to other shared-infrastructure siting problems such as e-bike docks, transit stops, or suburban EV charging, though the NYC calibration would need to be redone for each city type.
  • The stated limitations point to incremental extensions rather than reformulations: adding installation and grid costs, heterogeneous vehicle types, and user preferences such as charging speed and payment method.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes an optimization framework for siting EV charging stations in dense urban networks, with an objective (Eq. (7)) that minimizes travel cost and subtracts a lambda-weighted accessibility score. Constraints (1)-(5) require each demand point to be assigned to exactly one open station, limit the number of open stations, enforce x_ij <= y_j, impose capacity, and require total selected capacity to cover total demand. The authors prove feasibility and a lambda trade-off theorem, then use SUMO simulations of a New York City network to report that accessibility-aware planning reduces travel-time variability by 15-20% at modest distance cost. The paper's central claim is that this is the first framework to integrate the Mobility Equity Metric into charging-station optimization.

Significance. The intended contribution - an equity-aware, scalable charging-siting framework - is well motivated and, if realized, would be a useful planning tool. The MEM formulation and the use of SUMO with NYC data are appropriate. However, the stated optimization does not realize the contribution. Because x_ij <= y_j and sum_j x_ij = 1 imply sum_j x_ij y_j = 1 for every i, the accessibility term is constant, so Problem 1 is exactly equivalent to cost-only minimization for every lambda. The trade-off theorem and Pareto results are therefore vacuous with respect to the stated model, and the reported improvements have no mechanism. The paper ships no code or data, and the simulation section does not specify the station-selection rule that produced the MEM and NoMEM deployments. These issues are internal inconsistencies, not merely disagreements with the literature.

major comments (4)
  1. [II, Eq. (7) with constraints (1) and (3)] For every feasible (x,y), x_ij in {0,1} and x_ij <= y_j imply x_ij y_j = x_ij. Summing over j and applying constraint (1) gives sum_j x_ij y_j = 1 for each i. Hence the accessibility term lambda * sum_i sum_j epsilon_i x_ij y_j equals lambda * sum_i epsilon_i, a constant on the feasible set. Problem 1 is therefore identical to the cost-only problem for every lambda >= 0 and every epsilon. Proposition 1 in Sections III identifies this collapse only for uniform epsilon, but the collapse is general. Consequently, the MEM-vs-NoMEM comparison in Section IV and the Pareto curves in Fig. 2 cannot be solutions of the stated model; lambda cannot change the station set. The paper must either change the model (e.g., attach accessibility to stations via epsilon_j y_j or use a non-constant term) or explicitly state that the simulations use a different assignment/selection rule.
  2. [III, Theorem 2 and Remark 1] Theorem 2 claims lambda -> 0 recovers cost minimization and lambda -> infinity maximizes accessibility. Since f_access(x,y) in Eq. (22) is constant on the feasible set, the theorem is vacuous: the objective is f_cost(x) - lambda * const, so the argmin is the same for every finite lambda. The proof's term-wise comparison never establishes that f_access varies; it only rewrites the objective. Remark 1's asserted 'monotonic trade-off' therefore does not exist in this model. In addition, writing lim_{lambda -> infinity} argmin is not a well-defined set limit without a precise definition; in any case the optimal set does not depend on lambda.
  3. [III, Theorem 1 proof and constraints (4)-(5)] The feasibility proof treats d_i as a capacity consumption amount in Eq. (12), r_j <- r_j - d_i, while constraint (4) counts assignments via binary x_ij and constraint (5) uses weighted demand d_i. If d_i is a continuous weight, a binary x_ij cannot represent fractional demand; if d_i is an integer number of EVs, the proof requires d_i <= 1 for a single binary assignment. Moreover, condition (8), total capacity >= total demand, is not sufficient for indivisible assignments: with demand weights 6 and 6 and station capacities 10 and 2, total capacity 12 equals total demand 12, yet no assignment satisfies (4) with binary x. Thus the theorem's feasibility condition is not sufficient for the stated model, and the mismatch between (4) and (5) needs correction.
  4. [IV, Simulation setup and results] The pipeline first solves the optimization, then 'Assignment and Simulation: Assign EVs to nearest available stations' (Section IV.A). This means the reported travel times, distances, and energy values are not the objective value or assignment of Problem 1; they come from a post-hoc nearest-station heuristic. The paper does not state how lambda changes the set of open stations in that heuristic, nor how the MEM and NoMEM station sets differ. Without this information, the claimed 15-20% reduction in travel-time variability and the trade-offs in Fig. 2 are not reproducible from the model or the experimental description.
minor comments (6)
  1. [Title and throughout] The title uses 'United Framework'; the intended word is 'Unified Framework'.
  2. [Throughout] Typographical errors: 'technoclogical', 'accessiblity', 'V oronoi' in Section I.A, and inconsistent uses of 'accessibility score' for epsilon_i.
  3. [II, Table I and text] The text says 'each charging station j is also assigned an accessibility score epsilon_i', but epsilon_i is indexed by demand point i. Clarify whether MEM scores are attached to demand locations or candidate stations; the current constant-term issue depends on this indexing.
  4. [II, constraints] Problem 1 refers to constraints (1)-(6), but only (1)-(5) are explicitly defined. State constraint (6) (presumably binary restrictions) explicitly for completeness.
  5. [IV.B, Table II and Fig. 2] The reported numbers and Pareto plots lack error bars, confidence intervals, or repeated-trial statistics. Since the claim is about variability, a measure of variability across simulation runs is needed.
  6. [V.A] The discussion states 'Average trip distances increased by less than 0.1%', while Section IV.B reports MEM distances of 122 m vs 100 m and 153 m vs 135 m, which are 20% or more. The summary statistic is inconsistent with the reported simulation numbers.

Circularity Check

1 steps flagged · score 8.0 of 10

Accessibility term in objective is constant under constraints (1)-(3), so MEM cannot influence station selection; claimed trade-off and 15-20% improvement are artifacts of the model definition.

  1. other [Section II, Definition 2 / Problem 1 (Eq. 7) and constraints (1), (3); Section III, Theorem 2 and Remark 1; Section IV-B, NoMEM baseline definition]
    "min ∑_{i∈I} ∑_{j∈J} d_i c_ij x_ij − λ ∑_{i∈I} ∑_{j∈J} ε_i x_ij y_j (7) s.t. constraints (1)–(6). ... ∑_{j∈J} x_ij = 1, ∀i ∈ I, (1) ... x_ij ≤ y_j, ∀i ∈ I, ∀j ∈ J, (3) ... NoMEM refers to infrastructure placements optimized without considering the MEM framework, i.e., we consider only the first part of the objective function in (18)."

    With binary x and constraint (3), x_ij y_j = x_ij for every pair (i,j): if x_ij=1 then y_j must be 1; if x_ij=0 the product is 0. Summing over j and applying (1), ∑_j x_ij y_j = 1 for every i. Hence the accessibility term λ∑_i∑_j ε_i x_ij y_j equals λ∑_i ε_i, a constant independent of the decision variables x and y. Therefore objective (7) differs from the pure cost objective by only a constant, and the argmin is identical for every λ ≥ 0; the MEM term cannot influence station selection. The 'NoMEM baseline' (first part of (18)) is thus the same optimization as the MEM objective up to a constant, so the reported MEM-vs-NoMEM improvements and Pareto fronts have no mechanism in the stated model. Theorem 2's λ→∞ limit 'max_{x,y} ∑_{i,j} ε_i x_ij y_j' is vacuous because f_access is constant on

full rationale

The paper's central claim is that accessibility-aware planning reduces travel-time variability by 15–20% and that λ trades off efficiency against accessibility. But the paper's own equations destroy this mechanism: constraints (1) and (3), together with binary x, force the accessibility term in (7) to be λ∑_i ε_i, a constant. Thus Problem 1 is equivalent to the cost-only problem for every λ, and the MEM-vs-NoMEM comparison compares the same optimization problem against itself. The paper even proves the relevant identity in Proposition 1, but restricts it to uniform ε_i; the proof actually works for arbitrary ε_i because ∑_j x_ij y_j = 1 follows from (1)–(3) regardless of ε_i. Theorem 2 and Remark 1 therefore assert a trade-off that does not exist in the formulation: f_access is constant, so 'maximizing accessibility' is trivially satisfied by every feasible solution, and λ cannot interpolate between efficiency and equity. The 15–20% reduction is presented as an empirical result, but since the stated model cannot select different stations for different λ or for MEM vs NoMEM, the simulation comparison is unexplained by the model and the headline prediction reduces to an artifact. Self-citation is not the load-bearing issue here; the defect is internal to the equations. Score 8 reflects that the central claimed trade-off is forced by definition and the empirical claim lacks a mechanism in the stated model.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The framework relies on standard facility-location assumptions, homogeneity assumptions in SUMO, the authors' prior MEM formulation, and an unproven feasibility condition. The most consequential assumption is the single-assignment rule plus demand-fixed accessibility scores, which makes the equity term constant.

free parameters (5)
  • lambda (accessibility weight) = values 0,1,2,4,8 tested; authors recommend [1,4]
    Chosen by hand; claimed to control the equity-efficiency trade-off, but has no effect because the accessibility term is constant under constraint (1).
  • kappa_i (price sensitivity in MI)
    Appears in Definition 1 of the MI; no estimation procedure reported, presumably fixed from prior MEM work.
  • beta_s (service priority weights)
    Weights for service types in Definition 1; no calibration details are given.
  • tau_m (time thresholds)
    Time thresholds in the normalized service accessibility term of Definition 1; no values reported.
  • SOC threshold and initial SOC = not reported
    Simulation triggers charging below a set SOC threshold; values are not given, so energy results are not reproducible.
assumptions (5)
  • domain assumption Each demand point is assigned to exactly one selected station (constraints (1)-(3)).
    This is the standard single-assignment facility location assumption; it is also what makes the accessibility term constant, since sum_j x_ij y_j = 1 for every i.
  • domain assumption All vehicles are homogeneous: same initial charge, same battery capacity, same consumption rate, same vehicle type (Assumptions 1-4).
    Stated in Section III; used to simplify SUMO energy modeling.
  • domain assumption The Mobility Equity Metric from [24]-[26] correctly measures spatial accessibility equity.
    Adopted from the authors' prior work without external validation or comparison in this paper.
  • domain assumption SUMO with OpenStreetMap data for NYC faithfully represents traffic and charging behavior.
    Section IV-A; no validation against real data or error analysis.
  • ad hoc to paper Total candidate capacity >= total demand is sufficient for feasibility of indivisible assignments.
    Theorem 1 relies on this, but it is false when no single station has enough remaining capacity for a demand point; it also ignores the mismatch between weighted capacity constraint (5) and unweighted constraint (4).

how reviews work

0 comments
Cite this review

Pith. "Pith review of A United Framework for Planning Electric Vehicle Charging Accessibility." pith.science (2026). https://pith.science/paper/KFMABSP5

@misc{pith2026250805827,
  author       = {Pith},
  title        = {Pith review of: A United Framework for Planning Electric Vehicle Charging Accessibility},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KFMABSP5}},
  note         = {Machine review of arXiv:2508.05827}
}
read the original abstract

The shift towards electric vehicles (EVs) is crucial for establishing sustainable and low-emission urban transportation systems. However, the success of this transition depends on the strategic placement of the charging infrastructure. This paper addresses the challenge of optimizing charging station locations in dense urban environments while balancing efficiency with spatial accessibility. We propose an optimization framework that integrates traffic simulation, energy consumption modeling, and a mobility equity measure to evaluate the social reach of each potential charging station. Using New York City as a case study, we demonstrate consistent improvements in accessibility (15-20% reduction in travel time variability). Our results provide a scalable methodology for incorporating equity considerations into EV infrastructure planning, although economic factors and grid integration remain important areas for future development.

Figures

Figures reproduced from arXiv: 2508.05827 by the authors.

Figure 1
Figure 1. Distribution of total trip distances under MEM and [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Pareto analysis of efficiency-accessibility trade-offs [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Distribution of regenerative brake energy recovered [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Distribution of total energy consumption per trip [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 28 canonical work pages

  1. [1]

    Towards sustainable mobility: An optimization-based approach for smart ev charging reducing grid and environmental impact,

    A. Gonz ´alez-Garrido, F. Cort ´es, and M. Santos-Mugica, “Towards sustainable mobility: An optimization-based approach for smart ev charging reducing grid and environmental impact,” in 2022 IEEE International Smart Cities Conference (ISC2) , 2023, pp. 1–7

    work page 2022

  2. [2]

    Improving mobility toward sustainability,

    Suryani, Hendrawan, Adipraja, Widodo, Chou, and Zahra, “Improving mobility toward sustainability,” in 2023 IEEE 7th International Conference on Information Technology, Information Systems and Electrical Engineering (ICITISEE) . IEEE, 2023, pp. 30–35

    work page 2023

  3. [3]

    Electric vehicle charging station placement: Formulation, complexity, and solutions,

    A. Y . S. Lam, Y .-W. Leung, and X. Chu, “Electric vehicle charging station placement: Formulation, complexity, and solutions,” in IEEE Transactions on Smart Grid , vol. 5, no. 6, 2014, pp. 2846–2856

    work page 2014

  4. [4]

    Environmentally aware allocation of electric vehicle charging stations by analyzing locational marginal emissions,

    K. Santacruz and Y . Sang, “Environmentally aware allocation of electric vehicle charging stations by analyzing locational marginal emissions,” in 2022 North American Power Symposium (NAPS), 2022, pp. 1–6

    work page 2022

  5. [5]

    Locational marginal emission evaluation for electric vehicle charging facility planning,

    Y . Sang and I. Santiago, “Locational marginal emission evaluation for electric vehicle charging facility planning,” Center for Advancing Research in Transportation Emissions, Energy, and Health (CARTEEH), University of Texas at El Paso, Tech. Rep. UTEP-05-46, 2023

    work page 2023

  6. [6]

    Future trends in smart green iov: V2x, edge, and blockchain for sustainable development,

    F. Al-Turjman and A. Malekloo, “Future trends in smart green iov: V2x, edge, and blockchain for sustainable development,” IEEE Communications Magazine, vol. 59, no. 12, pp. 88–94, 2021

    work page 2021

  7. [7]

    Optimization of electric vehicle charging stations using renewable energy sources,

    E. Hossain, S. Uddin, K. Mahmud, M. Arif, and F. Fernandez, “Optimization of electric vehicle charging stations using renewable energy sources,” Energies, vol. 12, no. 20, p. 3790, 2019

    work page 2019

  8. [8]

    New york state transportation electrification report,

    New York State Energy Research and Development Authority, “New york state transportation electrification report,” 2021, accessed: 2025-04-12. [Online]. Available: https://www.nyserda.ny.gov/-/ media/Project/Nyserda/Files/Publications/Research/Transportation/ 21-06-New-York-State-Transportation-Electrification-Report.pdf

    work page 2021

Show all 29 references
  1. [9]

    Smart mobility for sustainable development goals: Enablers and barriers,

    G. Perboli and M. Rosano, “Smart mobility for sustainable development goals: Enablers and barriers,” Transportation Research Part A: Policy and Practice , vol. 153, pp. 1–15, 2021

    2021

  2. [10]

    Sumo– simulation of urban mobility: An overview,

    M. Behrisch, L. Bieker, J. Erdmann, and D. Krajzewicz, “Sumo– simulation of urban mobility: An overview,” in Proceedings of the Third International Conference on Advances in System Simulation (SIMUL 2011). ThinkMind, 2011, pp. 63–68

    2011

  3. [11]

    Charging stations for electric vehicles in sumo simulation environment and their impact on the traffic flow,

    E. Spachtholz, M. Glomsda, I. Kranefeld, F. E. Kracht, and D. Schramm, “Charging stations for electric vehicles in sumo simulation environment and their impact on the traffic flow,” in Proceedings of the 2024 International Conference on Traffic Simulation. IEEE, 2024

    2024

  4. [12]

    A sumo framework for deep reinforcement learning experiments solving electric vehicle charging dispatching problem,

    Y . Song, H. Zhao, R. Luo, L. Huang, Y . Zhang, and R. Su, “A sumo framework for deep reinforcement learning experiments solving electric vehicle charging dispatching problem,” arXiv preprint arXiv:2209.02921, 2022

    2022 arXiv

  5. [13]

    System modeling and simulation as a tool for developing a vision for future hybrid electric vehicle drivetrain configurations,

    C. Hoffmann, A. L ¨ocken, and B. M ¨uller, “System modeling and simulation as a tool for developing a vision for future hybrid electric vehicle drivetrain configurations,” in Proceedings of the SUMO 2013 Conference, 2013, pp. 43–50

    2013

  6. [14]

    Quantifying the environmental impacts of battery electric vehicles from a criticality perspective,

    F. P. V , M. F. D. R, A. Pehlken, and B. Koch, “Quantifying the environmental impacts of battery electric vehicles from a criticality perspective,” IEEE Xplore, June 2022

    2022

  7. [15]

    Eco-routing of electric vehicles in large networks: An aggregation- based approach,

    B. Asamer, M. Reinthaler, J. Ruthmair, M. Straub, and G. Fl ¨otter¨od, “Eco-routing of electric vehicles in large networks: An aggregation- based approach,” Transportation Research Part C: Emerging Technologies, vol. 67, pp. 431–448, 2016

    2016

  8. [16]

    Energy-efficient route navigation (eco-routing) for electric vehicles in sumo,

    I. Sagaama, A. Kchiche, W. Trojet, and F. Kamoun, “Energy-efficient route navigation (eco-routing) for electric vehicles in sumo,” IEEE Canadian Journal of Electrical and Computer Engineering , vol. 47, no. 4, pp. 187–198, 2024

    2024

  9. [17]

    A model-based eco-routing strategy for electric vehicles in large urban networks,

    G. D. Nunzio, L. Thibault, and A. Sciarretta, “A model-based eco-routing strategy for electric vehicles in large urban networks,” Transportation Research Part C: Emerging Technologies, vol. 78, pp. 98–115, 2017

    2017

  10. [18]

    Software-defined electric vehicle (ev)-to-ev charging framework with mobile aggregator,

    L. Chen, Y . Zhang, J. Wu, and K. Li, “Software-defined electric vehicle (ev)-to-ev charging framework with mobile aggregator,” in 2022 IEEE 95th Vehicular Technology Conference (VTC2022-Spring) , 2022, pp. 1–6

    2022

  11. [19]

    Distributed Optimization for Traffic Light Control and Connected Automated Vehicle Coordination in Mixed-Traffic Intersections,

    V .-A. Le and A. A. Malikopoulos, “Distributed Optimization for Traffic Light Control and Connected Automated Vehicle Coordination in Mixed-Traffic Intersections,” IEEE Control Systems Letters , vol. 8, pp. 2721–2726, 2024

    2024

  12. [20]

    Controller Adaptation via Learning Solutions of Contextual Bayesian Optimization,

    ——, “Controller Adaptation via Learning Solutions of Contextual Bayesian Optimization,” arXiv preprint arXiv:2403.04881 , 2024

    2024 arXiv

  13. [21]

    Experimental validation of a real-time optimal controller for coordination of cavs in a multi-lane roundabout,

    B. Chalaki, L. E. Beaver, and A. A. Malikopoulos, “Experimental validation of a real-time optimal controller for coordination of cavs in a multi-lane roundabout,” in 31st IEEE Intelligent Vehicles Symposium (IV), 2020, pp. 504–509

    2020

  14. [22]

    Routing guidance for emerging transportation systems with improved dynamic trip equity,

    T. Bai, A. Li, G. Xu, C. G. Cassandras, and A. A. Malikopoulos, “Routing guidance for emerging transportation systems with improved dynamic trip equity,” 2025, in 64th IEEE Conference on Decision and Control (in review), arXiv preprint arXiv: 2503.12601, 2025

    2025 arXiv

  15. [23]

    Combining cooperative re- routing with intersection coordination for connected and automated vehicles in urban networks,

    P. Typaldos and A. A. Malikopoulos, “Combining cooperative re- routing with intersection coordination for connected and automated vehicles in urban networks,” arXiv preprint arXiv:2503.10004 , 2025

    2025 arXiv

  16. [24]

    On mobility equity and the promise of emerging transportation systems,

    H. Bang, A. Dave, F. Tzortzoglou, S. Wang, and A. A. Malikopoulos, “On mobility equity and the promise of emerging transportation systems,” IEEE Transactions on Intelligent Transportation Systems , 2024 (in press)

    2024

  17. [25]

    A mobility equity metric for multi-modal intelligent transportation systems,

    H. Bang, A. Dave, F. Tzortzoglou, and A. A. Malikopoulos, “A mobility equity metric for multi-modal intelligent transportation systems,” Proceedings of the 17th IFAC Symposium on Control in Transportation Systems (CTS) , pp. 114–119, 2024, iFAC Young Author Award Finalist

    2024

  18. [26]

    Routing in mixed transportation systems for mobility equity,

    ——, “Routing in mixed transportation systems for mobility equity,” Proceedings of the 2024 American Control Conference (ACC) , pp. 1486–1491, 2024

    2024

  19. [27]

    Combining learning and control in linear systems,

    A. A. Malikopoulos, “Combining learning and control in linear systems,” European Journal of Control, vol. 80, no. Part A, p. 101043, 2024

    2024

  20. [28]

    Stochastic time-optimal trajectory planning for connected and automated vehicles in mixed-traffic merging scenarios,

    V .-A. Le, B. Chalaki, F. N. Tzortzoglou, and A. A. Malikopoulos, “Stochastic time-optimal trajectory planning for connected and automated vehicles in mixed-traffic merging scenarios,” IEEE Transactions on Control Systems Technology , 2024

    2024

  21. [29]

    An optimization case study for solving a transport robot scheduling problem on quantum-hybrid and quantum- inspired hardware,

    D. Leib, T. Seidel, S. J ¨ager, R. Heese, C. Jones, A. Awasthi, A. Niederle, and M. Bortz, “An optimization case study for solving a transport robot scheduling problem on quantum-hybrid and quantum- inspired hardware,” arXiv preprint arXiv:2309.09736 , 2023

    2023 arXiv

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.