Author: Ostanaqulov, Khikmatillo Akmaljon ogli

Annotation: This paper presents a theoretical framework for developing a decision-support methodology in Intelligent Transportation Systems (ITS) using Artificial Intelligence (AI) and Digital Twin (DT) technologies. The study applies a conceptual and analytical approach grounded in recent scientific literature (2020–2025). The methodology integrates five functional components: data acquisition, digital simulation, AI reasoning, decision and recommendation, and feedback synchronization.The framework demonstrates how AI algorithms and DT environments create a continuous feedback loop that enables real-time prediction, optimization, and adaptive control of traffic networks. Reinforcement learning and graph neural networks are identified as core AI mechanisms, while explainable AI (XAI) ensures transparency and interpretability of decisions. The results of theoretical synthesis confirm that the integration of AI and DT transforms traditional traffic management into a predictive and prescriptive ecosystem. The proposed architecture aligns with ISO/IEC 30173:2023 and Directive (EU) 2023/2661, providing a foundation for scalable, interoperable, and sustainable ITS design. The research contributes a structured methodological model that can guide future experimental and applied studies in smart mobility systems.

Keywords: Artificial Intelligence (AI); Digital Twin (DT); Intelligent Transportation Systems (ITS); Decision-Support Methodology; Reinforcement Learning; Graph Neural Networks; Explainable AI; Smart Mobility; Predictive Control; ISO/IEC 30173; Urban Transport Optimization.

Pages in journal: 122 - 131