Open Access

ARTICLE

Deep Multi-Agent Stochastic Optimization for Traffic Management in IoT-Enabled Transportation Networks

Nada Alasbali^*

Department of Informatics and Computer Systems, College of Computer Science, King Khalid University, Abha, 61421, Saudi Arabia

* Corresponding Author: Nada Alasbali. Email:

(This article belongs to the Special Issue: Intelligent Vehicles and Emerging Automotive Technologies: Integrating AI, IoT, and Computing in Next-Generation in Electric Vehicles)

Computers, Materials & Continua 2025, 85(3), 4943-4958. https://doi.org/10.32604/cmc.2025.068330

Received 26 May 2025; Accepted 12 September 2025; Issue published 23 October 2025

Abstract

Intelligent Traffic Management (ITM) has progressively developed into a critical component of modern transportation networks, significantly enhancing traffic flow and reducing congestion in urban environments. This research proposes an enhanced framework that leverages Deep Q-Learning (DQL), Game Theory (GT), and Stochastic Optimization (SO) to tackle the complex dynamics in transportation networks. The DQL component utilizes the distribution of traffic conditions for epsilon-greedy policy formulation and action and choice reward calculation, ensuring resilient decision-making. GT models the interaction between vehicles and intersections through probabilistic distributions of various features to enhance performance. Results demonstrate that the proposed framework is a scalable solution for dynamic optimization in transportation networks.

---

Keywords

---