ABSTRACT
This chapter provides an overview of the evolution from classical transportation planning models to contemporary approaches powered by artificial intelligence (AI) and machine learning (ML). It examines the limitations of traditional methods – such as aggregate focus, sequential modelling, and static data requirements – highlighting their challenges in addressing the complexity, scale, and dynamic nature of modern urban transport systems. The chapter introduces key AI/ML technologies and models relevant to transportation, including neural networks, evolutionary algorithms, fuzzy logic, agent-based models, and generative AI. Through a systematic comparison, this chapter explains the strengths of AI/ML in handling diverse datasets, enabling real-time adaptation, and solving optimisation problems, while also addressing the challenges of data dependency, interpretability, computational cost, and ethical considerations. The discussion extends to emerging trends such as deep learning, digital twins, and explainable AI, and outlines research priorities for integrating these technologies into sustainable, resilient, and equitable urban mobility systems. By synthesising current knowledge and highlighting practical and theoretical implications, this chapter serves as a resource for researchers, practitioners, and policymakers engaged in the future of intelligent transportation planning.
