This study addresses atmospheric pollution, specifically in urban areas such as Beijing, China, focusing on PM2.5 particles. The importance of China in air pollution research and its correlation with meteorological factors and chemical compounds are emphasized. A forecasting model based on a state-space modeling approach is proposed to predict air pollution variation, utilizing data collected between 2013 and 2017 from various monitoring stations in Beijing. The theoretical analysis includes key concepts of air pollution, previous studies on PM2.5, as well as an introduction to time series analysis and state-space models. The results show that variables related to atmospheric pressure and wind speed are significant for predicting air pollution, although further exploration of additional methods for more precise variable selection is suggested. Furthermore, it is concluded that the proposed model is effective for short-term forecasts but may require refinement for longer periods.