Sarcouncil Journal of Engineering and Computer Sciences

Abstract: This work represents a systematic review of the applications of Artificial Intelligence (AI) in environments defined as smart cities, with an emphasis on urban mobility and the open traffic datasets. The conducted review examines the existing studies published between 2019 and 2025, outlining the applications of ML, DL, RL and hybrid AI ranging from critical mobility-related tasks including traffic forecasting, route planning, demand estimation, anomaly detection and intelligent traffic signal control. The characteristics, availability, interoperability and features/limitations of open traffic datasets in smart cities that can be used to support AI-based transport systems are also examined by the study. Furthermore, we also performed an empirical exploration of the open traffic dataset in New York City (NYC), to showcase practical traffic patterns, time-based congestion behavior and segment-level traffic distribution. It concludes that AI-based mobility systems can increase the efficiency, sustainability and decision-making in transportation planning when they are backed by high quality open datasets. This is very important, but there are still serious challenges that remain in terms of standardization, privacy, interoperability and ethical governance by October 2023. The paper ends with unresolved research challenges and outlooks towards resilient, sustainable and data-driven urban mobility in smart cities.