Pith. sign in

REVIEW 3 major objections 5 minor 34 references

Joint Optimization of Electric Vehicle Routes and Charging Locations through Learning Charge Constraints Using QUBO Solvers

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Charging-station placement and battery-constrained EV routing can be jointly optimized by an outer Bayesian loop that learns a quadratic surrogate, with the inner route solver using only a soft battery penalty.

desk verdict Plausible BOCS+QUBO framework for EV charging placement, but the abstract overclaims and the battery equation has a concrete bug. read the letter →

arxiv 2506.04687 v2 pith:LPQZRYZD submitted 2025-06-05 quant-ph

Akihisa Okada , Keisuke Otaki , Hiroaki Yoshida This is my paper
classification quant-ph
keywords electricvehicleroutingchargingstationplacementquadraticunconstrainedbinaryoptimizationBayesiancombinatorialstructuresbatteryconstraintsquantumannealingblack-box
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 claims that the hard part of EV charging logistics—choosing where to put charging stations while respecting battery limits on a delivery tour—can be solved without ever encoding the battery constraint in the optimization model. The authors wrap a quadratic unconstrained binary optimization (QUBO) route solver in a Bayesian-optimization loop (BOCS) that treats station placement as a black-box binary input, learns a quadratic model of the mapping from station choices to route cost, and proposes new station configurations from that model. The inner route solver uses a soft penalty that keeps the battery level near a reference value, while the true “battery must stay between 0 and max” check is applied only when scoring each proposed configuration. On a 20-location problem with 16 candidate station sites, the loop reliably found battery-feasible configurations close to the best known solutions within 300 evaluations, using either quantum or simulated annealing for the QUBO steps.

What carries the argument

The central mechanism is the BOCS outer loop with a QUBO surrogate y = s^T A s for the map from the station-placement bit string s to the evaluated tour cost. Inside each evaluation, the route is produced by minimizing H_total = H_TSP + λ4 H_Battery, where H_Battery = Σ_t (Q_t − Q_standard)^2 is the soft battery term that keeps the battery near Q_standard without enforcing the hard [0, Qmax] interval. The true constraint enters only in the scoring step, where a violated interval at any time step adds a fixed penalty to the cost. This separation is what lets the framework “learn” the battery constraint rather than formulate it.

What would settle it

For the 20-location instance, enumerate all $2^{16}$ charging-station subsets (or solve them exactly with a mixed-integer program) and compute each one's true optimal battery-feasible tour cost. If the best configuration found by the BOCS loop after 300 iterations is not among the near-optimal subsets, or if the learned quadratic surrogate ranks configurations differently from their true evaluated costs near the optimum, the central claim fails.

Watch

Extended reading notes

Core claim

The central claim is that the coupled problem—where to install charging stations and what route the EV should take—can be separated into an outer and inner optimization, connected by learning. Given a station placement s, a QUBO route-search problem is solved with a deliberately simplified battery objective that only encourages the battery to stay near a reference level Qstandard. The resulting tour is then scored under the true battery constraint (0 ≤ Qt ≤ Qmax), with a fixed penalty per violated time step added to the travel cost. A Bayesian optimizer (BOCS) collects these scores and fits a QUBO surrogate y = s^T A s, then proposes the next station placement by minimizing that surrogate with quantum annealing or simulated annealing. The paper reports that this cycle finds feasible, low-cost station placements and tours on a 20-location asymmetric-cost instance within 300 iterations, and the abstract additionally reports small-scale exhaustive-enumeration validation showing that discovered configurations sit close to the global optimum.

Load-bearing premise

The load-bearing premise is that the mapping from charging-station choices to evaluated route cost is well approximated by a quadratic expression in those binary choices in the neighborhood of the optimum; if that approximation is poor, the outer loop will propose poor station configurations and the learned-constraint mechanism stops working.

Editorial extensions

If this is right

  • Routing and charging-station placement can be optimized together with only a few hundred evaluations, far fewer than the 2^16 possible station configurations.
  • Battery constraints can be enforced without the large number of auxiliary bits that a high-resolution binary encoding of the continuous battery level would require.
  • The obtained routes differ from pure shortest-cost tours, showing that station placement actively shapes the route.
  • The same “soft inner solver plus penalty-scored outer learner” pattern may extend to other range-type constraints, though the paper notes that the Q_standard term is domain-specific.
  • Quantum annealing and simulated annealing gave statistically similar outer-loop results on this instance, with quantum annealing slightly better on average.

Reading between the lines

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

  • If the quadratic-surrogate assumption holds generally, the framework should transfer to other constrained combinatorial problems whose inner solver can accept a soft version of the constraint, such as time-window or inventory constraints.
  • A natural testable extension is to add a penalty for the number of installed stations to the learned objective; the paper itself notes that the final charging station in its best solution may be redundant, so such a term could push toward sparser, cheaper deployments.
  • The efficiency claim would be stronger if the framework were compared against a classical exact solver on the same 20-location instance; the paper only compares quantum and simulated annealing as outer solvers, not the whole framework against a conventional baseline.
  • Because the inner route search is itself heuristic, part of the observed convergence could come from the route solver rather than the learned constraint; isolating that contribution would require fixing the inner solver and varying only the outer learner.
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

3 major / 5 minor

Summary. The paper proposes a black-box optimization framework for joint charging-station placement and battery-constrained EV routing. The outer loop uses BOCS to learn a quadratic surrogate y = s^T A s over the binary station-placement vector s, where y is the sum of the route cost (from an inner QUBO solver) plus a hand-set penalty for battery violations. The inner QUBO includes a soft penalty that keeps battery level near a reference Qstandard instead of strictly enforcing the [0, Qmax] range. Experiments on a 20-location instance with 16 candidate station locations compare quantum annealing (QA) and simulated annealing (SA) within BOCS, reporting convergence to feasible low-cost tours within 300 iterations.

Significance. If the framework works as claimed, it would offer a practical way to apply QUBO solvers to constrained combinatorial problems without explicitly encoding the constraint in the QUBO, which is a genuine contribution to the black-box QUBO optimization literature. The paper has strengths: it clearly separates the inner soft-constraint QUBO from the outer feasible-evaluation procedure (Steps 1–6 in the section 'Penalty for Satisfying Battery Constraints'), the experimental setup is described with explicit parameters in Table 1, and the comparison between QA and SA is a useful sanity check. However, the significance is substantially weakened by the absence of the validation promised in the abstract (exhaustive enumeration, runtime/QUBO-size analyses) and by the lack of any direct test of the quadratic surrogate that is the core of the learning claim.

major comments (3)
  1. [Abstract and Results] The abstract states that 'Small-scale validation experiments using exhaustive enumeration show that the framework reliably discovers feasible configurations close to the global optimum, while runtime and QUBO-size analyses clarify its computational characteristics.' The body of the paper contains neither of these. The Results section reports only search histories (Fig. 6), best-so-far curves (Fig. 7), and a cost table (Table 2); there is no exhaustive enumeration over the 2^16 station configurations, no comparison to the global optimum, and no runtime or QUBO-size measurements anywhere. This is load-bearing because the claim of efficiency (300 searches vs. 2^16 configurations) is only meaningful against an exhaustive or random-search baseline. Please either add the missing experiments or remove the corresponding abstract claims; as submitted, the central claim is unsupported.
  2. [Eq. (6) and Table 1] Equation (6) defines the battery level as Q_t = Q_1 - sum_{t'=1}^{t-1} [ sum_{i,j} C_{ij} x_{ijt'} - sum_{i,j} s_i x_{ijt'} ], i.e., each visit to a station adds one unit of charge (because s_i is binary). However, Table 1 sets Qcharge = 3, the actual charge per station visit used in the explicit battery simulation of Section 'Penalty for Satisfying Battery Constraints'. The inner QUBO's soft constraint (Eq. 7 with Qstandard = 3) is therefore inconsistent with the true evaluation model: the solver is encouraged to keep a battery model that charges by 1, while the evaluation penalizes violations of a battery that charges by 3. This mismatch undermines the reported numerical results and the claim that the soft penalty 'encourages' battery-feasible behavior. Please correct Eq. (6) to use Qcharge * s_i (or otherwise clarify), and re-run the experiments if necessary.
  3. [Discussion (surrogate approximation) and Results] The entire method rests on the assumption that the evaluation y(s) can be approximated by a quadratic form y = s^T A s over the 16-bit space, at least near the optimum. The evaluation y is discontinuous: it includes the inner solver's route cost plus a hard penalty ypenalty = 10 for every time step where the battery constraint is violated. The only justification offered is the Discussion's assertion that 'the expressive power of the QUBO formulation may be sufficiently high in the vicinity of the optimal solutions,' supported only by analogy to Taylor expansion and by reference to broad applicability of black-box QUBO optimization. No evidence is provided that this approximation holds: there is no held-out prediction error, no comparison of BOCS proposals against a random-search baseline, and no verification that the discovered best configurations are actually near the global optimum (which the abstract claims to have checked with exhaustive enumeration). Without such a test, the 'consistent convergence' in Figs. 6–7 could be the result of random exploration of the 2^16 space rather than genuine learning. Please include a quantitative diagnostic of surrogate quality (e.g., correlation or RMSE on held-out configurations) and a random-search or exhaustive-enumeration baseline; without these, the learning claim is not established.
minor comments (5)
  1. [Abstract] The phrase 'learning charge constraints' is misleading because the battery constraint is not learned; it is enforced by the hand-set penalty ypenalty and the soft reference level Qstandard, as the authors themselves acknowledge in the Discussion. Consider rephrasing to 'learning the mapping from station placements to feasible tour cost' or similar.
  2. [Methods, 'Penalty for Satisfying Battery Constraints'] The sentence 'It is important to note that, in computing the battery level at each time step, increases by charge (Qcharge) and travel-based consumption should be handled separately' is a valuable detail, but the order of operations in the actual simulation is not specified. Please state explicitly whether charging is applied after consumption at each visit and whether the battery is allowed to exceed Qmax during the step before a penalty is applied.
  3. [Figure 4] The caption says 'The color represents pseudo-elevation, where higher elevations are assumed to be unsuitable for charging station installation,' but the figure is grayscale in the manuscript and no color legend is provided. Please add a legend or describe the color mapping in the text.
  4. [Minor typos] There are several typos: 'promlems' (Introduction), 'leaning scheme' (Methods), 'the number of searching' (Fig. 3 caption), and inconsistent use of singular/plural 'solver/solvers' (e.g., 'QUBO solver' vs. 'specialized solvers'). These do not affect the technical content but should be fixed.
  5. [Data availability] The paper states data are available 'on reasonable request' but provides no code or data repository. Given that the abstract claims reproducible validation experiments, please consider making the code and data publicly available to support reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the evaluation y is computed from an external battery-constrained simulation, and BOCS merely fits a quadratic surrogate to that data in the standard Bayesian-optimization loop.

full rationale

The paper's derivation chain is self-contained with respect to circularity. The evaluation value y is not a renamed fit or a predicted quantity derived from the surrogate: it is computed by (1) solving the inner QUBO route problem for a given station placement s, (2) evaluating the true travel cost a via Eq. (1), (3) simulating the battery level step by step, and (4) adding an explicit hard penalty b for constraint violations (Methods, 'Penalty for Satisfying Battery Constraints', steps 1–6). BOCS then fits the quadratic model y = s^T A s to the collected (s, y) pairs and proposes the next s; this is the standard Bayesian-optimization loop from Baptista and Poloczek (ref. 33), an external method, not an input recycled as an output. The authors' self-citations (refs. 15, 19, 32) are used only as examples of black-box QUBO applications and are not load-bearing for the proposed framework. No uniqueness theorem, no ansatz smuggled through prior author work, and no quantity fitted to data is presented as a prediction. The Discussion explicitly admits that the effectiveness of the quadratic surrogate is not formally proven ('it is not possible to formally explain why the proposed framework effectively functions'), and the 'expressive power' argument is an unverified correctness assumption rather than a circular step. Similarly, the abstract's promise of exhaustive-enumeration validation and runtime analyses is not fulfilled in the Results, and the absence of random-search baselines weakens the empirical claim, but these are evidence/correctness concerns, not circularity. Therefore no circular step can be exhibited, and the appropriate score is 0.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central claim rests on an unproven approximation of the placement-to-cost function by a quadratic, on heuristic penalty choices (Qstandard, ypenalty), and on the assumption that the inner annealer returns representative routes. The battery equation in the inner QUBO is inconsistent with the stated Qcharge, weakening the model further.

free parameters (4)
  • Qstandard = 3
    Reference battery level in the soft penalty; set to half of Qmax with no sensitivity study.
  • ypenalty = 10
    Penalty added per time step with battery violation in the evaluation score; chosen by hand.
  • Penalty coefficients lambda_1..lambda_4 = not stated
    Described as 'sufficiently large' in Eq 5 and Eq 9; numeric values not reported.
  • BOCS hyperparameters NInit and Nsearch = 10 and 300
    Initial sample count and iteration budget; heuristic choices, no tuning analysis.
assumptions (4)
  • ad hoc to paper The objective over station configurations can be approximated by a quadratic form y = s^T A s in the relevant region.
    Used in the Discussion to justify BOCS; no approximation guarantee is given.
  • ad hoc to paper The soft battery penalty around Qstandard is sufficient to guide the inner route solver toward tours that are close to battery-feasible.
    The inner QUBO never enforces the true [0, Qmax] constraint; the method depends on this heuristic.
  • domain assumption The inner QUBO solver (Fixstars Amplify) provides good-enough routes that the evaluation y represents the placement quality.
    No solver statistics or optimality gaps are reported; noise in y would flow into the BOCS model.
  • domain assumption Battery evolution follows the paper's formula with a fixed recharge per station.
    The problem defines Qcharge=3, but Eq 6 uses s_i instead of Qcharge*s_i, so the implemented inner model differs from the stated dynamics.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Joint Optimization of Electric Vehicle Routes and Charging Locations through Learning Charge Constraints Using QUBO Solvers." pith.science (2026). https://pith.science/paper/LPQZRYZD

@misc{pith2026250604687,
  author       = {Pith},
  title        = {Pith review of: Joint Optimization of Electric Vehicle Routes and Charging Locations through Learning Charge Constraints Using QUBO Solvers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LPQZRYZD}},
  note         = {Machine review of arXiv:2506.04687}
}
read the original abstract

Optimal routing problems of electric vehicles (EVs) have attracted much attention in recent years, and installation of charging stations is an important issue for EVs. Hence, we focus on the joint optimization of the location of charging stations and the routing of EVs. When routing problems are formulated in the form of quadratic unconstrained binary optimization (QUBO), specialized solvers such as quantum annealers are expected to provide optimal solutions with high speed and accuracy. However, battery capacity constraints make it hard to formulate into QUBO form without a large number of auxiliary qubits. Here, we propose a sequential optimization method utilizing the Bayesian inference and QUBO solvers, in which the battery capacity constraints are automatically learned. This method enables us to optimize the number and location of charging stations and the routing of EVs with a small number of searches. Applying this method to a routing problem of 20 locations, we observed consistent convergence toward battery-feasible solutions across independent runs, demonstrating stable learning behavior of the proposed framework. Small-scale validation experiments using exhaustive enumeration show that the framework reliably discovers feasible configurations close to the global optimum, while runtime and QUBO-size analyses clarify its computational characteristics.

Figures

Figures reproduced from arXiv: 2506.04687 by the authors.

Figure 1
Figure 1. Problem setting. Locations where charging stations cannot be installed are represented as black dots, while locations where installation is possible are shown as circles. The objective is to determine a vehicle route that visits all required locations while maintaining a positive battery level throughout the tour and minimizing total travel cost. Simultaneously, the placement of charging stations must be determined—… view at source ↗
Figure 2
Figure 2. Optimization Cycle Using BOCS. The BOCS algorithm performs sequential optimization for a system in which the relationship between the binary input vector s (a bit string) and the corresponding output y is unknown. The process consists of the following three steps, which are repeated iteratively: (1) Obtain a data pair (s, y) by evaluating the output y for a given input s; (2) Learn the relationship between inputs s … view at source ↗
Figure 3
Figure 3. Flowchart of the sequential framework for determining the charging station placement s and the optimal tour z. Given a proposed station configuration s, the optimal tour z is obtained by minimizing the energy consumption a without strictly enforcing battery constraints. If the resulting tour violates battery constraints, a penalty is added to a, and the total becomes the evaluation value y. The relationship between … view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Two-dimensional spatial arrangement of the locations used in this study. The numbers next to the markers indicate the location labels. The color represents pseudo-elevation, where higher elevations are assumed to be unsuitable for charging station installation. In this…
Figure 6
Figure 6. Figure 6: Representative search trajectories of BOCS-QA under different random seeds. The vertical axis labeled ’Cost’ represents the final evaluation value y from Procedure 6. While the learning dynamics differ across runs, it is consistently observed that the search becomes mo…
Figure 7
Figure 7. Figure 7: Update history of the best-so-far tour cost with respect to the number of search iterations. The solid lines represent the average over 10 independent runs, and the shaded regions indicate the range between the maximum and minimum values. Results obtained using Quantum…
Figure 8
Figure 8. Figure 8: Battery level history during the tour that yielded the best tour cost under QA. (a) (b) Cost 3 2 1 0 [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 9
Figure 9. Figure 9: Effect of battery constraints on routing. (a) Route obtained by minimizing travel cost only. (b) Route obtained while considering battery level constraints. The black dots indicate locations where charging stations cannot be installed, while the gray circles represent …

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

34 extracted references · 14 canonical work pages

  1. [1]

    & Krems, J

    Rauh, N., Franke, T. & Krems, J. F. Understanding the impact of electric vehicle driving experience on range anxi- ety.Hum. Factors57, 177–187, DOI: 10.1177/0018720814546372 (2015). PMID: 25790577, https://doi.org/10.1177/ 0018720814546372

    work page doi:10.1177/0018720814546372 2015

  2. [2]

    & and, S

    Coffman, M., Bernstein, P. & and, S. W. Electric vehicles revisited: a review of factors that affect adoption.Transp. Rev. 37, 79–93, DOI: 10.1080/01441647.2016.1217282 (2017). https://doi.org/10.1080/01441647.2016.1217282

    arXiv 2017

  3. [3]

    Pamidimukkala, A., Kermanshachi, S., Rosenberger, J. M. & Hladik, G. Evaluation of barriers to electric vehicle adoption: A study of technological, environmental, financial, and infrastructure factors.Transp. Res. Interdiscip. Perspectives22, 100962, DOI: https://doi.org/10.1016/j.trip.2023.100962 (2023)

    arXiv 2023

  4. [4]

    & Goeke, D

    Schneider, M., Stenger, A. & Goeke, D. The electric vehicle-routing problem with time windows and recharging stations. Transp. Sci.48, 500–520 (2014)

    work page 2014

  5. [5]

    & Margaritis, D

    Afroditi, A., Boile, M., Theofanis, S., Sdoukopoulos, E. & Margaritis, D. Electric vehicle routing problem with industry constraints: Trends and insights for future research.Transp. Res. Procedia3, 452–459, DOI: https://doi.org/10.1016/j.trpro. 2014.10.026 (2014). 17th Meeting of the EURO Working Group on Transportation, EWGT2014, 2-4 July 2014, Sevilla, Spain

    doi:10.1016/j.trpro 2014

  6. [6]

    & Çatay, B

    Keskin, M. & Çatay, B. Partial recharge strategies for the electric vehicle routing problem with time windows.Transp. Res. Part C: Emerg. Technol.65, 111–127, DOI: https://doi.org/10.1016/j.trc.2016.01.013 (2016)

    doi:10.1016/j.trc.2016.01.013 2016

  7. [7]

    Montoya, A., Guéret, C., Mendoza, J. E. & Villegas, J. G. The electric vehicle routing problem with nonlinear charging function.Transp. Res. Part B: Methodol.103, 87–110, DOI: https://doi.org/10.1016/j.trb.2017.02.004 (2017). Green Urban Transportation

    doi:10.1016/j.trb.2017.02.004 2017

  8. [8]

    Lam, A. Y . S., Leung, Y .-W. & Chu, X. Electric vehicle charging station placement: Formulation, complexity, and solutions. IEEE Transactions on Smart Grid5, 2846–2856, DOI: 10.1109/TSG.2014.2344684 (2014)

    arXiv 2014

Show all 34 references
  1. [9]

    & Phung, Q

    Moghaddam, Z., Ahmad, I., Habibi, D. & Phung, Q. V . Smart charging strategy for electric vehicle charging stations.IEEE Transactions on Transp. Electrification4, 76–88, DOI: 10.1109/TTE.2017.2753403 (2018)

    2018

  2. [10]

    W.et al.Quantum annealing with manufactured spins.Nature473, 194–198, DOI: 10.1038/nature10012 (2011)

    Johnson, M. W.et al.Quantum annealing with manufactured spins.Nature473, 194–198, DOI: 10.1038/nature10012 (2011)

    2011 doi

  3. [11]

    In2015 IEEE International Solid-State Circuits Conference - (ISSCC) Digest of Technical Papers, 1–3, DOI: 10.1109/ISSCC.2015

    Yamaoka, M.et al.20k-spin ising chip for combinational optimization problem with cmos annealing. In2015 IEEE International Solid-State Circuits Conference - (ISSCC) Digest of Technical Papers, 1–3, DOI: 10.1109/ISSCC.2015. 7063111 (2015)

    2015 doi

  4. [12]

    Tatsumura, K., Dixon, A. R. & Goto, H. FPGA-based simulated bifurcation machine. In2019 29th International Conference on Field Programmable Logic and Applications (FPL), 59–66, DOI: 10.1109/FPL.2019.00019 (2019)

    2019

  5. [13]

    In 2020 25th Asia and South Pacific Design Automation Conference (ASP-DAC), 667–672, DOI: 10.1109/ASP-DAC47756

    Matsubara, S.et al.Digital annealer for high-speed solving of combinatorial optimization problems and its applications. In 2020 25th Asia and South Pacific Design Automation Conference (ASP-DAC), 667–672, DOI: 10.1109/ASP-DAC47756. 2020.9045100 (2020). 10/12

    2020

  6. [14]

    & Muramatsu, M

    Endo, K., Matsuda, Y ., Tanaka, S. & Muramatsu, M. A phase-field model by an Ising machine and its application to the phase-separation structure of a diblock polymer.Sci. Reports12, 10794, DOI: 10.1038/s41598-022-14735-4 (2022)

    2022 doi

  7. [15]

    & Yoshida, H

    Otaki, K., Okada, A. & Yoshida, H. Experimental study on the information disclosure problem: Branch-and-bound and qubo solver.Front. Appl. Math. Stat.9, DOI: 10.3389/fams.2023.1150921 (2023)

    2023

  8. [16]

    Ding, Y .et al.Toward prediction of financial crashes with a d-wave quantum annealer.Entropy25, DOI: 10.3390/e25020323 (2023)

    2023 doi

  9. [17]

    & Schmitt, S

    Yarkoni, S., Raponi, E., Bäck, T. & Schmitt, S. Quantum annealing for industry applications: introduction and review. Reports on Prog. Phys.85, 104001, DOI: 10.1088/1361-6633/ac8c54 (2022)

    2022 doi

  10. [18]

    B., Khan, F

    Hussain, H., Javaid, M. B., Khan, F. S., Dalal, A. & Khalique, A. Optimal control of traffic signals using quantum annealing.Quantum Inf. Process.19, DOI: 10.1007/s11128-020-02815-1 (2020)

    2020 doi

  11. [19]

    & Yoshida, H

    Inoue, D., Okada, A., Matsumori, T., Aihara, K. & Yoshida, H. Traffic signal optimization on a square lattice with quantum annealing.Sci. Reports11, 3303, DOI: 10.1038/s41598-021-82740-0 (2021)

    2021 doi

  12. [20]

    Nishimura, N., Tanahashi, K., Suganuma, K., Miyama, M. J. & Ohzeki, M. Item listing optimization for e-commerce websites based on diversity.Front. Comput. Sci.1, 2, DOI: 10.3389/fcomp.2019.00002 (2019)

    2019

  13. [21]

    Ohzeki, M., Miki, A., Miyama, M. J. & Terabe, M. Control of automated guided vehicles without collision by quantum annealer and digital devices.Front. Comput. Sci.1, 9, DOI: 10.3389/fcomp.2019.00009 (2019)

    2019

  14. [22]

    & Crispin, A

    Syrichas, A. & Crispin, A. Large-scale vehicle routing problems: Quantum annealing, tunings and results.Comput. & Oper. Res.87, 52–62, DOI: 10.1016/j.cor.2017.05.014 (2017)

    2017 doi

  15. [23]

    I.et al.Evaluation of quantum annealer performance via the massive mimo problem.IEEE Access9, 131658– 131671, DOI: 10.1109/ACCESS.2021.3114543 (2021)

    Tabi, Z. I.et al.Evaluation of quantum annealer performance via the massive mimo problem.IEEE Access9, 131658– 131671, DOI: 10.1109/ACCESS.2021.3114543 (2021)

    2021

  16. [24]

    J., Sanches, F., Ide, T., Kamiya, K

    Weinberg, S. J., Sanches, F., Ide, T., Kamiya, K. & Correll, R. Supply chain logistics with quantum and classical annealing algorithms.Sci. Reports13, 4770, DOI: 10.1038/s41598-023-31765-8 (2023)

    2023 doi

  17. [25]

    & Pan, W

    Ye, Z., Qian, X. & Pan, W. Quantum topology optimization via quantum annealing.IEEE Transactions on Quantum Eng. 4, 1–15, DOI: 10.1109/TQE.2023.3266410 (2023)

    2023

  18. [26]

    ICT 6, DOI: 10.3389/fict.2019.00013 (2019)

    Feld, S.et al.A hybrid solution method for the capacitated vehicle routing problem using a quantum annealer.Front. ICT 6, DOI: 10.3389/fict.2019.00013 (2019)

    2019

  19. [27]

    & Taguchi, S

    Irie, H., Wongpaisarnsin, G., Terabe, M., Miki, A. & Taguchi, S. Quantum annealing of vehicle routing problem with time, state and capacity. In Feld, S. & Linnhoff-Popien, C. (eds.)Quantum Technology and Optimization Problems, 145–156 (Springer International Publishing, Cham, 2019)

    2019

  20. [28]

    Kitai, K.et al.Designing metamaterials with quantum annealing and factorization machines.Phys. Rev. Res.2, 013319, DOI: 10.1103/PhysRevResearch.2.013319 (2020)

    2020 doi

  21. [29]

    & Ambai, M

    Kadowaki, T. & Ambai, M. Lossy compression of matrices by black box optimisation of mixed integer nonlinear programming.Sci. Reports12, 15482, DOI: 10.1038/s41598-022-19763-8 (2022)

    2022 doi

  22. [30]

    Express30, 43503–43512, DOI: 10.1364/OE.476839 (2022)

    Inoue, T.et al.Towards optimization of photonic-crystal surface-emitting lasers via quantum annealing.Opt. Express30, 43503–43512, DOI: 10.1364/OE.476839 (2022)

    2022 doi

  23. [31]

    & Kadowaki, T

    Matsumori, T., Taki, M. & Kadowaki, T. Application of qubo solver using black-box optimization to structural design for resonance avoidance.Sci. Reports12, 12143, DOI: 10.1038/s41598-022-16149-8 (2022)

    2022 doi

  24. [32]

    Okada, A.et al.Design optimization of noise filter using quantum annealer.IEEE Access11, 44343–44349, DOI: 10.1109/ACCESS.2023.3271969 (2023)

    2023

  25. [33]

    & Poloczek, M

    Baptista, R. & Poloczek, M. Bayesian optimization of combinatorial structures. In Dy, J. & Krause, A. (eds.)Proceedings of the 35th International Conference on Machine Learning, vol. 80 ofProceedings of Machine Learning Research, 462–471 (PMLR, 2018)

    2018

  26. [34]

    & Togawa, N

    Shirai, T. & Togawa, N. Post-processing variationally scheduled quantum algorithm for constrained combinatorial optimization problems.IEEE Transactions on Quantum Eng.1–14, DOI: 10.1109/TQE.2024.3376721 (2024). 35.D-Wave Systems Inc. dwave-neal. Https://github.com/dwavesystems...

    2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.