Authors: Anam Firoz Shaikh, Professor Bangar Abhishek P, Professor Bhosale Sachin B, Professor Mandlik S.Y.
Abstract: Air pollution is a major environmental concern because it affects human health, ecosystems, and the quality of life in urban and industrial areas. Traditional monitoring systems often depend on fixed stations and manual sampling. Although these methods provide useful measurements, they may have high operating costs, limited spatial coverage, and delays in reporting. Recent developments in Artificial Intelligence (AI), Machine Learning (ML), the Internet of Things (IoT), and wireless sensors offer ways to monitor air quality continuously and forecast pollution levels. This review examines four supplied papers covering AI and ML models with wireless sensors, IoT-based monitoring, comparisons between traditional and AI-based methods, and a proposed smart-city forecasting framework. The reviewed literature describes applications such as missing-data estimation, sensor calibration, anomaly detection, Air Quality Index (AQI) estimation, and short-term pollution forecasting. Models including regression, Random Forest, Support Vector Machines, neural networks, and deep-learning methods are discussed. The papers also identify challenges related to data quality, sensor reliability, cost, scalability, model transfer between locations, and system integration. One proposed framework reports improved prediction results, but these figures are treated as results reported by that paper rather than independently verified findings. Overall, AI can support earlier warnings and better-informed pollution management when it is combined with reliable sensors , quality data, and local validation.
