Improving Air Quality in Cities: Forecasting Air Pollution in Kuwait

As urbanisation and industrial activities continue to accelerate, air quality, particularly the concentration of particulate matter (PM), has become a major environmental concern. PM2.5, in particular, presents significant health risks, especially in regions like Kuwait where rapid urban growth and industrial activities contribute to high pollution levels.

This research aims to enhance the accuracy of PM2.5 concentration forecasts in Kuwait by employing advanced deep learning models, addressing the limitations of traditional machine learning (ML) methods. The primary contribution will be to compare traditional ML techniques with deep learning models to determine which approach provides more precise, region-specific predictions. In doing so, we will address existing challenges such as data quality and the need for improved datasets for long-term forecasting, enabling a better understanding of the complex spatial-temporal patterns of air pollution. By improving forecasting accuracy, the study will provide valuable insights for the development of effective environmental policies and public health interventions, ultimately contributing to improved air quality management and environmental sustainability in Kuwait and the wider GCC region. This research will advance the understanding of AI applications in air pollution prediction, providing a foundation for more informed decision-making to protect public health and the environment. 

This project forms part of Kuwait Programme, funded by the Kuwait Foundation for the Advancement of Science (KFAS).

_{A hot and dusty scene with Kuwaiti flags hazy in the orange sand. (C) Imago / Alamy Stock Photo}