Machine learning in transportation: key applications and methods

Abstract

Machine learning (ML) has become a foundational technology in modern transportation systems, enabling data-driven decision-making across traffic operations, safety, logistics, and system-level planning. With the rapid growth of sensing technologies, connected vehicles, and large-scale mobility data, ML techniques are increasingly used to model complex, non-linear, and dynamic transport processes that are difficult to address using traditional analytical or rule-based approaches [1]–[4]. These methods support improvements in safety, operational efficiency, environmental sustainability, and user experience in both passenger and freight transportation.