This study presents a comprehensive framework for deploying Artificial Intelligence (AI) to advance smart and sustainable urban transportation, using Isfahan, Iran, as a case study. The research designs and proposes a multi-model AI architecture, utilizing Graph Neural Networks (GNNs) with LSTM layers for high-accuracy (target >80%) short-term traffic prediction, Deep Reinforcement Learning for adaptive signal control that incorporates BRT priority, and XGBoost for passenger demand forecasting. A phased implementation plan is outlined, integrating these models with Isfahan's existing BRT data infrastructure through a microservices architecture. The projected environmental impact, calculated via a tailored emissions model, indicates targeted reductions of 20% in CO₂ emissions and 18% in fuel consumption. A socio-economic cost-benefit analysis forecasts a substantial benefit-cost ratio (BCR > 2.5) by optimizing travel time, safety, and operational costs. The study critically addresses implementation challenges, including data governance, computational demands, and algorithmic bias, providing a replicable blueprint for AI-driven urban mobility that balances efficiency, equity, and environmental sustainability.
Ghaderi et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: