Generative AI in Transportation Planning: Emerging Applications, Opportunities and Challenges

Abstract

Transportation planning is becoming more complex due to rapid urbanization, increase in vehicle ownership, climate concerns and dynamic travel behaviour of people. Traditional modelling and simulation approaches often require large manual effort and sometimes fails to represent real-world uncertainty. Recently, Generative Artificial Intelligence (Generative AI) has shown significant potential to transform how transportation systems are analysed, designed and optimized. Generative AI models such as Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), Large Language Models (LLMs) and diffusion models can generate realistic traffic patterns, synthetic mobility data, road network designs and even support decision making in policy formulation. This paper presents a comprehensive review on the role of Generative AI in transportation planning. The study discusses various techniques, applications in traffic prediction, demand modelling, infrastructure planning, autonomous mobility, and smart city integration. Benefits, limitations and future research directions are also highlighted. The review indicates that Generative AI can significantly improve data-driven transportation planning while reducing cost and time, however issues related to data quality, interpretability and ethics needs further attention.

KEYWORDS: Generative AI, Transportation Planning, Traffic Modelling, GAN, Smart Mobility, Urban Transport, Synthetic Data