pith. sign in

arxiv: 2606.20667 · v1 · pith:TCDVDWCBnew · submitted 2026-06-12 · 💻 cs.AI

A Quantum-Assisted Agentic Distributed Artificial Intelligence Framework for Deadline-Bounded Orchestration of Hybrid Renewable Microgrids

Iacovos I. Ioannou , Saher Javaid , Minella Bezha , Yasuo Tan , Naoto Nagaoka , Vasos Vassiliou This is my paper

Pith reviewed 2026-06-27 05:20 UTC · model grok-4.3

classification 💻 cs.AI
keywords microgrid dispatchQUBO formulationBDIx agentsquantum-assisted optimizationagentic solver selectionstorage valuationrenewable energy orchestrationdeadline-bounded control
0
0 comments X

The pith

A quantum-assisted agentic framework solves every microgrid dispatch slot to exact optimality while meeting all deliberation deadlines.

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

The paper establishes a distributed AI system in which BDIx agents encode microgrid dispatch and coalition problems as QUBO instances, then a coordinator agent selects from a portfolio of quantum, quantum-inspired and classical solvers on the basis of learned latency beliefs. A belief-shaped storage valuation injects future-peak price signals into the per-slot optimization. In a 24-hour grid-connected test case the system meets every deadline, returns the exact global optimum each slot, and produces a daily cost of 146.24 EUR that equals the computed lower bound together with 97.83 percent renewable utilization and zero unserved energy.

Core claim

The central claim is that the described quantum-assisted DAI framework, by elevating solver selection to an agentic action informed by learned latency beliefs and by pricing storage at a discounted future-peak value, solves the combinatorial dispatch problem to global optimality in every control slot without violating the hard deliberation deadline, yielding a realized daily cost identical to the theoretical minimum.

What carries the argument

The coordinator agent's learned-belief-based selection of the solver expected to finish before the deadline, applied to QUBO problems formulated by BDIx agents together with the storage agent's belief-shaped future-peak valuation.

If this is right

  • The committed dispatch attains the exact optimum on every slot.
  • The realized daily cost equals the exact lower bound of 146.24 EUR.
  • Renewable utilization reaches 97.83 percent with zero unserved energy.
  • Deactivating the storage valuation mechanism raises the daily cost by 4.5 percent to 152.75 EUR.
  • Zero deliberation deadlines are missed across the full simulation.

Where Pith is reading between the lines

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

  • The same agentic solver-selection logic could be tested on larger microgrid instances or different asset mixes to check whether the zero-miss record holds.
  • Replacing the statevector QAOA simulator with actual quantum hardware would provide a direct test of whether hardware noise still permits deadline compliance.
  • The framework's structure suggests the approach could be applied to other deadline-constrained combinatorial problems such as real-time logistics or traffic signal coordination.

Load-bearing premise

The BDIx agents correctly encode the full combinatorial dispatch and coalition-formation problem as a QUBO whose global optimum corresponds to the true minimum-cost schedule under the physical and market constraints.

What would settle it

Exhaustive enumeration on any control slot reveals that the framework's committed dispatch differs from the true minimum-cost solution, or any deliberation deadline is missed in the 24-hour simulation.

Figures

Figures reproduced from arXiv: 2606.20667 by Iacovos I. Ioannou, Minella Bezha, Naoto Nagaoka, Saher Javaid, Vasos Vassiliou, Yasuo Tan.

Figure 1
Figure 1. Figure 1: Quantum-assisted DAI architecture realized by BDIx agents. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: 24-hour operation of the proposed framework: sources and consumers, [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Solver benchmark over the 96 per-slot QUBO instances, where QAOA, Tabu, SA, PSO and Greedy denote hybrid QAOA, tabu search, simulated [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Deadline-bounded meta-deliberation over the 24-hour horizon, with [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
read the original abstract

The real-time orchestration of microgrids that combine fluctuating renewable sources, dispatchable units, storage and curtailable consumers requires the repeated solution of combinatorial dispatch and coalition formation problems under hard control deadlines. In this paper, a quantum-assisted agentic distributed artificial intelligence (DAI) framework is proposed in which the dispatch problem of each control slot is formulated as a quadratic unconstrained binary optimization (QUBO) problem by Belief-Desire-Intention extended (BDIx) agents and is solved by a portfolio of quantum, quantum-inspired and classical solvers. Solver selection is elevated to a first-class agentic deliberation action of the coordinator agent. Learned beliefs about solver latencies are maintained and the solver intention that is expected to satisfy the prevailing deliberation deadline is committed in each slot. In addition, a belief-shaped storage valuation mechanism is introduced through which the storage agent prices its energy at a discounted future-peak value, injecting intertemporal information into the otherwise myopic per-slot optimization. The framework is evaluated on a 24-hour simulation of a grid-connected microgrid with photovoltaic, wind, battery, genset and demand-response assets, with the Quantum Approximate Optimization Algorithm (QAOA) executed by statevector simulation and benchmarked per slot against tabu search, simulated annealing, binary particle swarm optimization, greedy descent and exhaustive enumeration. Zero deliberation deadlines are missed, the committed dispatch attains the exact optimum on every slot and the realized daily cost of 146.24 EUR equals the exact lower bound, with 97.83 percent renewable utilization and zero unserved energy. When the storage valuation mechanism is deactivated, the daily cost is increased to 152.75 EUR, a 4.5 percent increase.

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 manuscript proposes a quantum-assisted agentic distributed AI framework for real-time orchestration of hybrid renewable microgrids. BDIx agents encode per-slot combinatorial dispatch and coalition problems as QUBOs solved by a portfolio of quantum (QAOA), quantum-inspired, and classical solvers, with the coordinator agent selecting the solver expected to meet the deliberation deadline based on learned latency beliefs. A belief-shaped storage valuation mechanism prices stored energy at a discounted future-peak value. In a 24-hour grid-connected microgrid simulation (PV, wind, battery, genset, demand response), the framework reports zero missed deadlines, exact optimal dispatch on every slot, daily cost of 146.24 EUR matching the exact lower bound, 97.83% renewable utilization, and zero unserved energy; deactivating storage valuation raises cost to 152.75 EUR.

Significance. If the QUBO encoding is shown to be exact and the reported optimality is verified, the work would demonstrate a viable hybrid quantum-classical approach to deadline-constrained microgrid optimization, with the agentic solver portfolio and intertemporal storage valuation as potentially useful contributions to energy-system AI. The perfect attainment of the lower bound and the quantified 4.5% benefit from the valuation mechanism would strengthen the case for such frameworks if the underlying assumptions hold.

major comments (2)
  1. [Abstract] Abstract: the central claim that 'the committed dispatch attains the exact optimum on every slot' and equals the lower bound of 146.24 EUR requires that the QUBO formulation exactly represents the original mixed-integer program (power balance, storage dynamics, generator limits, demand-response and market constraints) with no omitted higher-order terms or insufficient penalties. No derivation of this encoding, no verification that decoded solutions satisfy all constraints, and no proof that QAOA or portfolio solutions reach the global minimum of the true feasible set are provided.
  2. [Evaluation] Evaluation (simulation results): the comparison against exhaustive enumeration and the assertion that the realized cost equals the exact lower bound are only valid if the QUBO feasible set is identical to the enumerated set. Any relaxation or mismatch in constraint encoding would render the optimality conclusion and the 97.83% renewable utilization claim unsupported.
minor comments (2)
  1. [Abstract] The abstract mentions statevector simulation of QAOA but provides no details on qubit count, circuit depth, or convergence criteria used per slot.
  2. No sensitivity analysis or error bars are reported for solver performance or the impact of the belief-shaped valuation discount factor.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive review. The comments correctly identify that the current manuscript lacks an explicit derivation of the QUBO encoding and direct verification of constraint equivalence. We will revise the paper to supply these elements while preserving the reported simulation outcomes.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that 'the committed dispatch attains the exact optimum on every slot' and equals the lower bound of 146.24 EUR requires that the QUBO formulation exactly represents the original mixed-integer program (power balance, storage dynamics, generator limits, demand-response and market constraints) with no omitted higher-order terms or insufficient penalties. No derivation of this encoding, no verification that decoded solutions satisfy all constraints, and no proof that QAOA or portfolio solutions reach the global minimum of the true feasible set are provided.

    Authors: We agree that the manuscript does not presently contain the requested derivation or verification. In the revised version we will insert a new subsection that (i) maps each MIP constraint (power balance, storage SOC dynamics, generator bounds, DR limits, and market transactions) to its QUBO penalty term with explicit coefficient values, (ii) states the decoding procedure, and (iii) reports the fraction of decoded solutions that satisfy every original constraint to machine precision. We will also add a short argument showing that, for the problem sizes considered, the solver portfolio returns the global minimum of the constructed QUBO, which is identical to the MIP feasible set under the chosen penalties. revision: yes

  2. Referee: [Evaluation] Evaluation (simulation results): the comparison against exhaustive enumeration and the assertion that the realized cost equals the exact lower bound are only valid if the QUBO feasible set is identical to the enumerated set. Any relaxation or mismatch in constraint encoding would render the optimality conclusion and the 97.83% renewable utilization claim unsupported.

    Authors: The exhaustive enumeration was performed directly on the original MIP; the QUBO solutions were decoded and re-checked against the MIP constraints before cost and utilization metrics were computed. We will expand the evaluation section to tabulate, for every slot and every solver, the constraint-violation count after decoding and to state the exact penalty schedule that guarantees equivalence. This will make the optimality claim and the 97.83 % renewable-utilization figure rest on an explicit demonstration that the two feasible sets coincide. revision: yes

Circularity Check

0 steps flagged

No significant circularity; verification against exhaustive enumeration keeps the optimum claim independent of QUBO formulation.

full rationale

The paper formulates dispatch as QUBO via BDIx agents, solves via portfolio (including QAOA), and explicitly benchmarks every slot against exhaustive enumeration to confirm the committed dispatch matches the exact lower bound of 146.24 EUR. This external enumeration over the feasible set provides an independent check rather than defining the optimum by the QUBO itself. The storage valuation mechanism is introduced as an explicit addition whose effect is measured by ablation (cost rises to 152.75 EUR when deactivated), supplying empirical content instead of self-definition. No load-bearing step reduces to a fitted parameter renamed as prediction or to a self-citation chain; the central result is self-contained against the stated benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 1 invented entities

Review is limited to the abstract; the ledger therefore records only the domain assumptions explicitly invoked in the provided text.

axioms (2)
  • domain assumption The dispatch and coalition-formation problem of each control slot can be faithfully encoded as a QUBO by BDIx agents.
    Stated as the starting point of the framework in the abstract.
  • domain assumption A portfolio of quantum, quantum-inspired and classical solvers can be selected by an agent on the basis of learned latency beliefs to meet hard deliberation deadlines.
    Central to the solver-selection mechanism described.
invented entities (1)
  • belief-shaped storage valuation mechanism no independent evidence
    purpose: To price stored energy at a discounted future-peak value and thereby inject intertemporal information into the per-slot optimization.
    Introduced in the abstract as an addition to the otherwise myopic optimization.

pith-pipeline@v0.9.1-grok · 5868 in / 1595 out tokens · 20693 ms · 2026-06-27T05:20:08.564113+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

33 extracted references · 31 canonical work pages · 2 internal anchors

  1. [1]

    MicroGrids,

    R. H. Lasseter, “MicroGrids,” inProc. IEEE Power Eng. Soc. Win- ter Meeting, vol. 1, New York, NY , USA, 2002, pp. 305–308, doi: 10.1109/PESW.2002.985003

    work page doi:10.1109/pesw.2002.985003 2002

  2. [2]

    Trends in microgrid control,

    D. E. Olivares, A. Mehrizi-Sani, A. H. Etemadi, C. A. Cañizares, R. Ira- vani, M. Kazerani, A. H. Hajimiragha, O. Gomis-Bellmunt, M. Saeedi- fard, R. Palma-Behnke, G. A. Jiménez-Estévez, and N. D. Hatziargyriou, “Trends in microgrid control,”IEEE Trans. Smart Grid, vol. 5, no. 4, pp. 1905–1919, Jul. 2014, doi: 10.1109/TSG.2013.2295514

    work page doi:10.1109/tsg.2013.2295514 1905

  3. [3]

    Microgrids energy management systems: A critical review on methods, solutions, and prospects,

    M. F. Zia, E. Elbouchikhi, and M. Benbouzid, “Microgrids energy management systems: A critical review on methods, solutions, and prospects,”Appl. Energy, vol. 222, pp. 1033–1055, Jul. 2018, doi: 10.1016/j.apenergy.2018.04.103. IEEE TRANSACTIONS ON SMART GRID, VOL. XX, NO. X, 2026 10

    work page doi:10.1016/j.apenergy.2018.04.103 2018

  4. [4]

    Operation of a multiagent system for microgrid control,

    A. L. Dimeas and N. D. Hatziargyriou, “Operation of a multiagent system for microgrid control,”IEEE Trans. Power Syst., vol. 20, no. 3, pp. 1447–1455, Aug. 2005, doi: 10.1109/TPWRS.2005.852060

    work page doi:10.1109/tpwrs.2005.852060 2005

  5. [5]

    Multiagent system for real-time operation of a microgrid in real-time digital simulator,

    T. Logenthiran, D. Srinivasan, A. M. Khambadkone, and H. N. Aung, “Multiagent system for real-time operation of a microgrid in real-time digital simulator,”IEEE Trans. Smart Grid, vol. 3, no. 2, pp. 925–933, Jun. 2012, doi: 10.1109/TSG.2012.2189028

    work page doi:10.1109/tsg.2012.2189028 2012

  6. [6]

    Multi-agent systems for power engineering applications, Part I: Concepts, approaches, and technical challenges,

    S. D. J. McArthur, E. M. Davidson, V . M. Catterson, A. L. Dimeas, N. D. Hatziargyriou, F. Ponci, and T. Funabashi, “Multi-agent systems for power engineering applications, Part I: Concepts, approaches, and technical challenges,”IEEE Trans. Power Syst., vol. 22, no. 4, pp. 1743– 1752, Nov. 2007, doi: 10.1109/TPWRS.2007.908471

    work page doi:10.1109/tpwrs.2007.908471 2007

  7. [7]

    Quantum computing for energy systems optimization: Challenges and opportunities,

    A. Ajagekar and F. You, “Quantum computing for energy systems optimization: Challenges and opportunities,”Energy, vol. 179, pp. 76– 89, Jul. 2019, doi: 10.1016/j.energy.2019.04.186

    work page doi:10.1016/j.energy.2019.04.186 2019

  8. [8]

    Quantum-enhanced grid of the future: A primer,

    R. Eskandarpour, K. Ghosh, A. Khodaei, A. Paaso, and L. Zhang, “Quantum-enhanced grid of the future: A primer,”IEEE Access, vol. 8, pp. 188 993–189 002, 2020, doi: 10.1109/ACCESS.2020.3031595

    work page doi:10.1109/access.2020.3031595 2020

  9. [9]

    Annealing-based quantum computing for combinatorial optimal power flow,

    T. Morstyn, “Annealing-based quantum computing for combinatorial optimal power flow,”IEEE Trans. Smart Grid, vol. 14, no. 2, pp. 1093– 1102, Mar. 2023, doi: 10.1109/TSG.2022.3200590

    work page doi:10.1109/tsg.2022.3200590 2023

  10. [10]

    A Quantum Approximate Optimization Algorithm

    E. Farhi, J. Goldstone, and S. Gutmann, “A quantum approximate optimization algorithm,” 2014, arXiv:1411.4028

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  11. [11]

    (2014) Ising formulations of many NP problems

    A. Lucas, “Ising formulations of many NP problems,”Front. Phys., vol. 2, p. 5, Feb. 2014, doi: 10.3389/fphy.2014.00005

    work page doi:10.3389/fphy.2014.00005 2014

  12. [12]

    (2018) Quantum computing in the NISQ era and beyond

    J. Preskill, “Quantum computing in the NISQ era and beyond,”Quan- tum, vol. 2, p. 79, Aug. 2018, doi: 10.22331/q-2018-08-06-79

    work page internal anchor Pith review doi:10.22331/q-2018-08-06-79 2018

  13. [13]

    BDI agents: From theory to practice,

    A. S. Rao and M. P. Georgeff, “BDI agents: From theory to practice,” inProc. 1st Int. Conf. Multi-Agent Syst. (ICMAS), San Francisco, CA, USA, 1995, pp. 312–319

    1995

  14. [14]

    Dis- tributed artificial intelligence solution for D2D communication in 5G networks,

    I. Ioannou, V . Vassiliou, C. Christophorou, and A. Pitsillides, “Dis- tributed artificial intelligence solution for D2D communication in 5G networks,”IEEE Syst. J., vol. 14, no. 3, pp. 4232–4241, Sep. 2020, doi: 10.1109/JSYST.2020.2979044

    work page doi:10.1109/jsyst.2020.2979044 2020

  15. [15]

    A novel distributed AI framework with ML for D2D communication in 5G/6G networks,

    I. Ioannou, C. Christophorou, V . Vassiliou, and A. Pitsillides, “A novel distributed AI framework with ML for D2D communication in 5G/6G networks,”Comput. Netw., vol. 211, p. 108987, Jul. 2022, doi: 10.1016/j.comnet.2022.108987

    work page doi:10.1016/j.comnet.2022.108987 2022

  16. [16]

    Dynamic D2D communication in 5G/6G using a distributed AI framework,

    I. I. Ioannou, C. Christophorou, V . Vassiliou, M. Lestas, and A. Pit- sillides, “Dynamic D2D communication in 5G/6G using a distributed AI framework,”IEEE Access, vol. 10, pp. 62 772–62 799, 2022, doi: 10.1109/ACCESS.2022.3182388

    work page doi:10.1109/access.2022.3182388 2022

  17. [17]

    M. E. Bratman,Intention, Plans, and Practical Reason. Cambridge, MA, USA: Harvard Univ. Press, 1987

    1987

  18. [18]

    Ioannou, P

    I. Ioannou, P. Nagaradjane, V . Vassiliou, A. Pitsillides, and C. Christophorou,Distributed Artificial Intelligence for 5G/6G Com- munications: Frameworks with Machine Learning. Boca Raton, FL, USA: CRC Press, 2024, doi: 10.1201/9781003469209

    work page doi:10.1201/9781003469209 2024

  19. [19]

    A distributed AI framework for nano-grid power management and control,

    I. I. Ioannou, S. Javaid, C. Christophorou, V . Vassiliou, A. Pitsillides, and Y . Tan, “A distributed AI framework for nano-grid power management and control,”IEEE Access, vol. 12, pp. 43 350–43 377, 2024, doi: 10.1109/ACCESS.2024.3377926

    work page doi:10.1109/access.2024.3377926 2024

  20. [20]

    Demand side management: De- mand response, intelligent energy systems, and smart loads,

    P. Palensky and D. Dietrich, “Demand side management: De- mand response, intelligent energy systems, and smart loads,”IEEE Trans. Ind. Informat., vol. 7, no. 3, pp. 381–388, Aug. 2011, doi: 10.1109/TII.2011.2158841

    work page doi:10.1109/tii.2011.2158841 2011

  21. [21]

    Game-theoretic methods for the smart grid: An overview of microgrid systems, demand-side man- agement, and smart grid communications,

    W. Saad, Z. Han, H. V . Poor, and T. Ba¸ sar, “Game-theoretic methods for the smart grid: An overview of microgrid systems, demand-side man- agement, and smart grid communications,”IEEE Signal Process. Mag., vol. 29, no. 5, pp. 86–105, Sep. 2012, doi: 10.1109/MSP.2012.2186410

    work page doi:10.1109/msp.2012.2186410 2012

  22. [22]

    Energy and op- eration management of a microgrid using particle swarm opti- mization,

    J. Radosavljevi ´c, M. Jevti ´c, and D. Klimenta, “Energy and op- eration management of a microgrid using particle swarm opti- mization,”Eng. Optim., vol. 48, no. 5, pp. 811–830, 2016, doi: 10.1080/0305215X.2015.1057135

    work page doi:10.1080/0305215x.2015.1057135 2016

  23. [23]

    Intelligent multi-microgrid energy management based on deep neural network and model-free reinforcement learning,

    Y . Du and F. Li, “Intelligent multi-microgrid energy management based on deep neural network and model-free reinforcement learning,”IEEE Trans. Smart Grid, vol. 11, no. 2, pp. 1066–1076, Mar. 2020, doi: 10.1109/TSG.2019.2930299

    work page doi:10.1109/tsg.2019.2930299 2020

  24. [24]

    Quantum bridge analytics I: A tutorial on formulating and using QUBO models,

    F. Glover, G. Kochenberger, and Y . Du, “Quantum bridge analytics I: A tutorial on formulating and using QUBO models,”4OR, vol. 17, no. 4, pp. 335–371, Dec. 2019, doi: 10.1007/s10288-019-00424-y

    work page doi:10.1007/s10288-019-00424-y 2019

  25. [25]

    Zhou, S.-T

    L. Zhou, S.-T. Wang, S. Choi, H. Pichler, and M. D. Lukin, “Quan- tum approximate optimization algorithm: Performance, mechanism, and implementation on near-term devices,”Phys. Rev. X, vol. 10, no. 2, p. 021067, Jun. 2020, doi: 10.1103/PhysRevX.10.021067

    work page doi:10.1103/physrevx.10.021067 2020

  26. [26]

    Variational quantum algorithms,

    M. Cerezo, A. Arrasmith, R. Babbush, S. C. Benjamin, S. Endo, K. Fujii, J. R. McClean, K. Mitarai, X. Yuan, L. Cincio, and P. J. Coles, “Variational quantum algorithms,”Nat. Rev. Phys., vol. 3, no. 9, pp. 625–644, Sep. 2021, doi: 10.1038/s42254-021-00348-9

    work page doi:10.1038/s42254-021-00348-9 2021

  27. [27]

    Quantum distributed unit commitment: An application in microgrids,

    N. Nikmehr, P. Zhang, and M. A. Bragin, “Quantum distributed unit commitment: An application in microgrids,”IEEE Trans. Power Syst., vol. 37, no. 5, pp. 3592–3603, Sep. 2022, doi: 10.1109/TP- WRS.2022.3141794

    work page doi:10.1109/tp- 2022

  28. [28]

    Adapting quantum approximation optimization algorithm (QAOA) for unit commitment,

    S. Koretsky, P. Gokhale, J. M. Baker, J. Viszlai, H. Zheng, N. Gurung, R. Burg, E. A. Paaso, A. Khodaei, R. Eskandarpour, and F. T. Chong, “Adapting quantum approximation optimization algorithm (QAOA) for unit commitment,” inProc. IEEE Int. Conf. Quantum Comput. Eng. (QCE), 2021, pp. 181–187, doi: 10.1109/QCE52317.2021.00035

    work page doi:10.1109/qce52317.2021.00035 2021

  29. [29]

    K. J. R. Liu, A. K. Sadek, W. Su, and A. Kwasinski,Cooperative Com- munications and Networking. Cambridge, U.K.: Cambridge University Press, 2009, doi: 10.1017/CBO9780511754524

    work page doi:10.1017/cbo9780511754524 2009

  30. [30]

    An accurate intelligent plan library for belief-based desire prioritization to intentions in BDIx agents,

    I. Ioannou, A. Gregoriades, P. Nagaradjane, C. Christophorou, and V . Vassiliou, “An accurate intelligent plan library for belief-based desire prioritization to intentions in BDIx agents,” inProc. Int. Conf. Wireless Commun. Signal Process. Netw. (WiSPNET), 2025, pp. 1–6, doi: 10.1109/WiSPNET64060.2025.11005173

    work page doi:10.1109/wispnet64060.2025.11005173 2025

  31. [31]

    Tabu search, Part II,

    F. Glover, “Tabu search, Part II,”ORSA J. Comput., vol. 2, no. 1, pp. 4–32, 1990, doi: 10.1287/ijoc.2.1.4

    work page doi:10.1287/ijoc.2.1.4 1990

  32. [32]

    and Vecchi, M.P

    S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi, “Optimization by simulated annealing,”Science, vol. 220, no. 4598, pp. 671–680, May 1983, doi: 10.1126/science.220.4598.671

    work page doi:10.1126/science.220.4598.671 1983

  33. [33]

    A discrete binary version of the particle swarm algorithm,

    J. Kennedy and R. C. Eberhart, “A discrete binary version of the particle swarm algorithm,” inProc. IEEE Int. Conf. Syst., Man, Cybern., vol. 5, Orlando, FL, USA, 1997, pp. 4104–4108, doi: 10.1109/IC- SMC.1997.637339

    work page doi:10.1109/ic- 1997