With the increasing demand for efficient and reliable public transportation systems, the integration of the Internet of Things (IoT) and machine learning models has emerged as a transformative approach. This paper aims to understand the possible improvements of using IoT and machine learning in optimizing the functionality of smart public transport systems. It outlines an extensive architecture that makes use of the real-time data collected through the IoT sensors to apply the machine learning algorithms to different areas of public transport such as the schedules, the routes, and the maintenance. With the ability to study the current flow of passengers and the movement of the vehicles, the proposed system will act to minimize delays, increase service delivery, and control passenger satisfaction levels. This paper outlines the method for deploying the integrated system and analyzes the results based on the simulations carried out in this research study; this paper also addresses the implications of the study on future smart city developments.