Multi-Agent Reinforcement Learning for Traffic Signal Optimization in Tier-2 Cities

Mousa, Sura; Sahana, Mahmood

doi:10.25673/123729

Proceedings of International Conference on Applied Innovation in IT · 2026/03/31 · Vol. 14 · Issue 1 · pp. 1279–1286

Multi-Agent Reinforcement Learning for Traffic Signal Optimization in Tier-2 Cities

Sura Hamed Mousa, Mahmood Anees Ahmed and Subrata Sahana

📄 Download PDF DOI: 10.25673/123729

Abstract

Traffic jamming in cities is also one major issue of concern with the major concern being the situation in Tier-2 cities where urbanization has been taking an overgrowth without the development of the infrastructure. The older fixed time and actuated traffic signal systems are not usually able to adjust to irregular and mixed traffic, which causes long delays, more fuel use as well as higher emissions. In order to solve these problems, this paper suggests a Multi-Agent Reinforcement Learning (MARL)-based adaptive traffic signal control scheme. According to the suggested model, every intersection will be described as a self-governed agent, which monitors the state of local traffic, chooses the best possible signal phases, and modifies its policy according to the information provided by the environment. The system was tested with the realistic simulation environment of a Tier-2 city with mixed vehicle vehicles, pedestrian movement and noisy sensor data. Findings indicate that the MARL framework can reduce vehicle delays as well as pedestrian wait times by a significant factor than fixed-time, actuated, and single-agent DRL models in enhancing throughput and decreasing emissions. In addition, ablation experiments ratified the significance of multi-objective reward design in the attainment of a balanced optimization. This study identifies the opportunity of using MARL as a scalable and cost-effective solution to enhance the management of traffic in resource-limited urban settings, and this research paper prepares the way to deploy it in real-world Tier-2 city networks in the future.

Keywords

Multi-Agent Reinforcement Learning (MARL) Traffic Signal Optimization Tier-2 Cities Adaptive Traffic Control Intelligent Transportation Systems (ITS) Deep Reinforcement Learning (DRL).

References

L. Li, Y. Lv, and F. Y. Wang, “Traffic signal timing via deep reinforcement learning,” IEEE/CAA Journal of Automatica Sinica, vol. 3, no. 3, pp. 247-254, 2016.
Y. Zhao, Z. Zhang, K. Huang, and X. Li, “A reinforcement learning-based approach for coordinated traffic signal control in urban environments,” Applied Soft Computing, vol. 97, p. 106836, 2020.
T. Chu, J. Wang, L. Codeca, and Z. Li, “Multi-agent deep reinforcement learning for large-scale traffic signal control,” IEEE Transactions on Intelligent Transportation Systems, vol. 21, no. 3, pp. 1086-1095, 2019.
H. Wei, N. Xu, H. Zhang, G. Zheng, X. Zang, C. Chen, and Z. Li, “CoLight: Learning network-level cooperation for traffic signal control,” in Proceedings of the 28th ACM International Conference on Information and Knowledge Management, pp. 1913-1922, 2019.
H. Wei, C. Chen, G. Zheng, K. Wu, V. Gayah, K. Xu, and Z. Li, “PressLight: Learning max pressure control to coordinate traffic signals in arterial network,” in Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp. 1290-1298, 2019.
L. Kuyer, S. Whiteson, B. Bakker, and N. Vlassis, “Multiagent reinforcement learning for urban traffic control using coordination graphs,” in Joint European Conference on Machine Learning and Knowledge Discovery in Databases, pp. 656-671, Springer, Berlin, Heidelberg, 2008.
S. Whiteson, “Multiagent Reinforcement Learning for Urban Traffic Control using Coordination Graphs,” 2008.
P. Michailidis, I. Michailidis, C. R. Lazaridis, and E. Kosmatopoulos, “Traffic Signal Control via Reinforcement Learning: A Review on Applications and Innovations,” Infrastructures, vol. 10, no. 5, p. 114, 2025.
W. Jia and M. Ji, “Multi-Agent Deep Reinforcement Learning for Large-Scale Traffic Signal Control with Spatio-Temporal Attention Mechanism,” Applied Sciences, vol. 15, no. 15, p. 8605, 2025.
C. Wang, Y. Li, J. Chen, J. Zhang, and Y. Xue, “Cooperative traffic signal control for a partially observed vehicular network using multi-agent reinforcement learning,” Engineering Applications of Artificial Intelligence, vol. 160, p. 111813, 2025.
K. Othman, X. Wang, A. Shalaby, and B. Abdulhai, “Multimodal adaptive traffic signal control: A decentralized multiagent reinforcement learning approach,” Multimodal Transportation, vol. 4, no. 1, p. 100190, 2025.
M. Kolat, B. Kővári, T. Bécsi, and S. Aradi, “Multi-agent reinforcement learning for traffic signal control: A cooperative approach,” Sustainability, vol. 15, no. 4, p. 3479, 2023.
A. Agafonov, A. Yumaganov, and V. Myasnikov, “Cooperative control for signalized intersections in intelligent connected vehicle environments,” Mathematics, vol. 11, no. 6, p. 1540, 2023.
D. Li, F. Zhu, J. Wu, Y. D. Wong, and T. Chen, “Managing mixed traffic at signalized intersections: An adaptive signal control and CAV coordination system based on deep reinforcement learning,” Expert Systems with Applications, vol. 238, p. 121959, 2024.
G. Zhang, F. Chang, J. Jin, F. Yang, and H. Huang, “Multi-objective deep reinforcement learning approach for adaptive traffic signal control system with concurrent optimization of safety, efficiency, and decarbonization at intersections,” Accident Analysis & Prevention, vol. 199, p. 107451, 2024.
A. Alshammari, M. Alsalmi, N. Almetleqem, and N. Alajmi, “Exploring the Potential of Manufacturing Bioplastics from Waste and Wastewater Sources: A Review,” Journal of Techniques, vol. 7, no. 2, pp. 67-74, 2025, [Online]. Available: https://doi.org/10.51173/jt.v7i2.2676.
S. M. Ferhan and H. Agahi, “Multi-Objective Optimization of Hybrid Energy Systems,” Electrical Engineering Technical Journal, vol. 2, no. 2, pp. 1-16, 2025, [Online]. Available: https://doi.org/10.51173/eetj.v2i2.22.