PhD Candidate: Olsi Miraçi
PhD supervisor and co-supervisor: Prof. Asoc. Dr. Lekë Pepkolaj (University Metropolitan Tirana), Prof. Asoc. Dr. Salvatore Romano (University of Salento, Italy)
Air quality degradation represents one of the most critical environmental and public health challenges worldwide, driven by increasing urbanization, industrial activities, and climate change. While recent advances in environmental sensing technologies have enabled high-resolution monitoring of atmospheric pollutants, conventional monitoring and modeling approaches remain limited by high costs, sparse spatial coverage, delayed response times, and difficulties in handling large, heterogeneous datasets. Recent studies have shown that Artificial Intelligence (AI) can significantly enhance environmental monitoring by improving pollution detection, forecasting capabilities, and decision-making processes, particularly in resource-constrained or data-intensive contexts (Olawade et al., 2024; Chadalavada et al., 2024).
This PhD research focuses on the integration of multi-source environmental sensors with Artificial Intelligence and Machine Learning (ML) techniques to advance air quality monitoring and forecasting. In detail, the research aims to combine data from ground-based sensor networks, meteorological observations, and, where available, remote sensing platforms to improve the detection, characterization, and prediction of atmospheric pollutants such as particulate matter and gaseous species. AI-driven approaches have demonstrated strong potential in extracting complex, non-linear relationships between pollutant concentrations and environmental drivers, enabling more accurate and adaptive air quality assessment compared to traditional deterministic models (Neo et al., 2023; Bainomugisha et al., 2024).
The proposed methodology is based on deep learning and unsupervised learning algorithms to automatically identify spatio-temporal patterns, anomalous pollution events, and evolving emission behaviors within heterogeneous datasets. Unlike conventional approaches, AI-based systems can continuously learn from incoming data streams, enhancing predictive performance over time with limited human intervention. This adaptive capability is particularly relevant for real-time monitoring and short- to medium-term forecasting in dynamic urban environments, as highlighted by recent AI-assisted smart city air quality frameworks (Neo et al., 2023; Nguyen et al., 2024).
A key objective of this research is the development of computationally efficient, scalable, and interpretable AI models suitable for near-real-time applications. Special attention will be devoted to feature optimization, data fusion strategies, and the use of synthetic data generation techniques to address data gaps, sensor uncertainties, and the limited availability of extreme pollution events, which have been identified as critical challenges in current AI-based environmental monitoring systems (Olawade et al., 2024; Alotaibi & Nassif, 2024).
Overall, this work aims to contribute to the development of intelligent, sensor-based environmental monitoring and forecasting systems that support early warning mechanisms, evidence-based policy making, and sustainable air quality management in urban and regional contexts.
Keywords: Air Quality Monitoring, Artificial Intelligence, Environmental Sensors, Machine Learning, Forecasting, Environmental Analysis
References:
Alotaibi, E., Nassif, N. (2024). Artificial intelligence in environmental monitoring: in-depth analysis. Discov Artif Intell 4, 84. https://doi.org/10.1007/s44163-024-00198-1
Bainomugisha, E., Warigo, P. A., Daka, F. B., Nshimye, A., Birungi, M., & Okure, D. (2024). AI-driven environmental sensor networks and digital platforms for urban air pollution monitoring and modelling. Societal Impacts, 3, Article 100044. https://doi.org/10.1016/j.socimp.2024.100044
Chadalavada, S., Faust, O., Salvi, M., Seoni, S., Raj, N., Raghavendra, U., Gudigar, A., Barua, P. D., Molinari, F., & Acharya, R. (2024). Application of artificial intelligence in air pollution monitoring and forecasting: A systematic review. Environmental Modelling & Software, 173, 106312. https://doi.org/10.1016/j.envsoft.2024.106312
Neo, E. X., Hasikin, K., Lai, K. W., Mokhtar, M. I., Azizan, M. M., Hizaddin, H. F., Razak, S. A., & Yanto. (2023). Artificial intelligence-assisted air quality monitoring for smart city management. PeerJ Computer Science, 9, e1306. https://doi.org/10.7717/peerj-cs.1306
Nguyen, A.T., Pham, D.H., Oo, B. et al. (2024). Predicting air quality index using attention hybrid deep learning and quantum-inspired particle swarm optimization. J Big Data 11, 71. https://doi.org/10.1186/s40537-024-00926-5
Olawade, D. B., Wada, O. Z., Ige, A. O., Egbewole, B. I., Olojo, A., & Oladapo, B. I. (2024). Artificial intelligence in environmental monitoring: Advancements, challenges, and future directions. Hygiene and Environmental Health Advances, 12, Article 100114. https://doi.org/10.1016/j.heha.2024.100114
