Air pollution poses a significant threat to global health and the environment, with automobile emissions being a primary contributor. Motor vehicles are among the major sources of urban air pollution. Traditional methods for monitoring vehicular air pollutants, such as Portable Emissions Measurement Systems (PEMS), are limited by their inability to transmit data to the cloud for comprehensive analysis and predictive modeling. The purpose of this study is to develop a framework for real-time monitoring of various pollutants, including carbon monoxide (CO), nitrogen oxides (NOx), sulfur dioxide (SO2), volatile organic compounds (VOCs), and particulate matter (PM), using Internet of Things (IoT) sensors. This study aims to leverage the collected data for predictive insights, enabling effective preventive maintenance and mitigation strategies for air pollution control. In this paper, we discuss in detail the framework consisting of seven layers: (i) sensing, (ii) processing and local storage, (iii) edge-based computing, (iv) network, (v) cloud-based storage, (vi) cloud-based computing, (vii) application. As a proof of concept, we implement and discuss an IoT solution using an MQ135 sensor, connected to a cloud server via WiFi or GSM network, where the data is stored and analyzed. Additionally, we integrate machine learning algorithms to predict future pollutant levels and enhance preparedness for preventive maintenance and air pollution reduction. The implementation of this framework is expected to enhance the accuracy of air quality assessments, provide actionable insights for improving urban air quality, and effectively reduce air pollution by monitoring and detecting pollutants early at their source. This enables vehicle owners, regulators, and fleet managers to take appropriate measures based on forecasted insights.