Air pollution can originate from both natural sources, such as wildfires and volcanic eruptions, and human activities, including vehicular traffic and industrial processes. These activities release particulate matter (PM2.5, PM10) and gases like CO2 and volatile organic compounds (VOCs) into the environment, negatively impacting health and productivity in workplace settings. In response to this issue, an intelligent system is proposed that utilizes IoT sensors and machine learning algorithms to monitor, predict, and control air quality in real time within automotive workshops. A data storage and processing architecture was implemented using Amazon Web Services (AWS), and the predictive model was trained using the Random Forest algorithm. The system also incorporates automated air filters that are activated when critical pollutant concentrations are detected. Preliminary results show high accuracy in pollutant prediction, highlighting the system’s potential to effectively monitor environmental conditions in industrial environments.