AI-Enabled IoT System for Real-Time Indoor Air Quality Monitoring to Safeguard the Health and Safety of Industrial Workers

Abstract

The rapid advancement of technological innovations has led to the pervasive digitization infiltrating every aspect of modern life, often termed as Industry 4.0. This transformative paradigm incorporates digital technologies within professional environments, engendering the emergence of “smart factories” and fundamentally altering conventional manufacturing methodologies. Consequently, the establishment of robust Occupational Health and Safety (OHS) protocols has become imperative in the digital era to ensure the welfare of the workforce. Indoor Air Quality (IAQ) assumes a pivotal role in OHS, given that substandard air quality presents considerable health hazards to individuals in occupational settings. This paper describes the RDSA (Risk Detection and Safety Assistance) system, an integrated framework that leverages the Internet of Things (IoT) and artificial intelligence (AI) to monitor hazardous conditions in real time and proactively assess risks. The RDSA system combines cost-effective IoT sensors with cloud-based dashboards to aggregate and clarify environmental data, while machine learning algorithms predict indoor air quality (IAQ) levels and facilitate early identification of pollution spikes. In addition, a convolutional neural network (CNN) is used for instant face mask detection to promote compliance with safety measures. Among the models examined for IAQ prediction, logistic regression, decision tree, and random forest classifiers showed the highest effectiveness, each achieving 99.9% accuracy on the evaluation dataset. In the context of face mask detection, the CNN model achieved an accuracy of 0.99. The proposed system demonstrates strong performance for all classifiers and significantly improves workplace safety by promoting data-driven decision-making. The empirical results validate the effectiveness and scalability of the RDSA system for practical industrial implementation.