UMAF: Unified Multi-Agent Framework for Real-Time Public Transport Dispatching

References

[1] S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Computation, vol. 9, no. 8, pp. 1735–1780, Nov. 1997, doi: 10.1162/neco.1997.9.8.1735.

[2] R. S. Sutton and A. G. Barto, Reinforcement Learning: An Introduction, 2nd ed. Cambridge, MA: MIT Press, 2018.

[3] P. Veličković et al., "Graph Attention Networks," in Proc. 6th Int. Conf. Learning Representations (ICLR 2018), Vancouver, BC, Canada, Apr. 2018.

[4] B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, "Communication-Efficient Learning of Deep Networks from Decentralized Data," in Proc. 20th Int. Conf. Artificial Intelligence and Statistics (AISTATS), Fort Lauderdale, FL, Apr. 2017, pp. 1273–1282.

[5] C. Dwork, "Differential Privacy," in Automata, Languages and Programming, vol. 4052, Lecture Notes in Computer Science, Springer, Berlin, Heidelberg, 2006, pp. 1–12.

[6] J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov, "Proximal Policy Optimization Algorithms," arXiv preprint arXiv:1707.06347, Jul. 2017.

[7] A. Goodfellow et al., "Generative Adversarial Nets," in Advances in Neural Information Processing Systems 27 (NIPS 2014), pp. 2672–2680, 2014.

[8] M. Pecht, "Battery Management Systems for Large Lithium-Ion Battery Packs," Microelectronics Reliability, vol. 52, no. 7, pp. 1440–1445, 2012.

[9] X. Liang, X. Shen, J. Liu, N. Li, W. Yu, and Y. Zhuang, "Scheduling of Virtual Reality Applications in D2D Networks," IEEE Trans. Wireless Commun., vol. 17, no. 6, pp. 4153–4168, Jun. 2018.

[10] R. H. Thaler and C. R. Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness. New Haven, CT: Yale Univ. Press, 2008.

[11] T. Lütjen, T. Feige, and J. Hauenschild, "Federated Learning for Smart City Transportation," IEEE Trans. Intell. Transport. Syst., vol. 23, no. 8, pp. 12201–12213, Aug. 2022.

[12] Y. Liang and C. L. Philip Chen, "Multi-Agent Reinforcement Learning for Public Transit," IEEE Trans. Cybernetics, vol. 52, no. 3, pp. 1893–1905, Mar. 2022.

@article{1718018,
      author = {Hemanth Kumar H},
      title = {UMAF: Unified Multi-Agent Framework for Real-Time Public Transport Dispatching},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {9},
      number = {11},
      pages = {2941-2951},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1718018.pdf},
      abstract = {This paper proposes a Unified Multi-Agent Framework (UMAF) for real-time public transport dispatching in smart cities. The framework integrates eight critical technical domains: spatial flux modeling, 5G V2X communication, battery State-of-Health (SoH) monitoring, human-centric automation, federated edge privacy, adversarial AI defense, behavioral economics, and multi-agent reinforcement learning. Current dispatching systems suffer from fragmentation — they optimize for traffic flow while ignoring hardware constraints or cybersecurity threats. By utilizing a Multi-Agent Reinforcement Learning (MARL) engine cross-verified against chemistry-aware battery models and lightweight adversarial filters, the proposed system demonstrates a 15–18% reduction in deadheading and a 20% extension in electric vehicle (EV) fleet longevity. Furthermore, the implementation of a federated edge architecture ensures 100% privacy compliance with minimal computational latency, providing a secure and sustainable foundation for next-generation smart city mobility. This research uniquely bridges the gap between academic AI theory and real-world operational constraints by proposing a framework that is simultaneously technically rigorous, privacy-preserving, and economically viable. The results indicate that UMAF outperforms existing state-of-the-art dispatch systems across all evaluated performance metrics.},
      month = {May},
      doi = {https://doi.org/10.64388/IREV9I11-1718018}
  }