Enhancing road traffic flow in sustainable cities through transformer models: Advancements and challenges

Abstract

Efficient traffic flow is crucial for sustainable cities, as it directly impacts energy consumption, pollution levels, and overall quality of life. The integration of superficial intelligence, particularly transformer models, plays a significant role in enhancing the predictive capabilities for traffic management, thereby supporting sustainable urban development. In this survey, we explored the application of transformer models to predict and optimize traffic flow in sustainable cities. These models leverage advanced machine learning to capture intricate spatiotemporal patterns,thereby providing valuable insights for urban planners and traffic management centers. By systematically reviewing the literature, we emphasize the importance of transformer models in urban planning and sustainable resource use. Our study demonstrates how transformer models can learn complex spatiotemporal patterns from traffic data by incorporating both real-time and historical data to enhance prediction accuracy. This improved predictive capability aids the development of smart cities by reducing traffic congestion, facilitating smoother movement for city dwellers and tourists, and ultimately contributing to the sustainability goals of urban areas. This comprehensive review highlights the transformative potential of predictive modeling using transformer models, underscoring their critical role in optimizing urban infrastructure and promoting sustainable city development.