Nowadays, air pollution is a worldwide challenging issue. Hence, its changing scenario must be assessed regularly. Given rising concerns over environmental pollution and its impact on health, accurate air quality prediction is crucial for urban planning and public health. The objective of this research is to evaluate the effectiveness of the autoregressive integrated moving average (ARIMA) model in predicting the concentration of key air pollutants: respirable suspended particulate matter (RSPM), sulfur dioxide (SO2), and oxides of nitrogen (NOx), over a 30-day period. For this analysis, all concerned datasets were adopted from the manual ground monitoring station, Albert Ekka Chowk (Ranchi), which was found to be one of the major polluting locations in Jharkhand. Later, data preprocessing, such as handling missing values and seasonal decomposition, is also done, and finally, the datasets are prepared for modeling. The steps involved while performing this predictive model are the application of the ARIMA model over historical air quality data, followed by validation against observed values, and finally, the model’s parameters fine-tuned to optimize its predictive accuracy. The output showed that for RSPM, SO2, and NOx, the predicted values got around 105 µg/m3, 17.2 µg/m3, and 36.2 µg/m3 closely matched its observed range of 102–106 µg/m3, 16.8–17.5 µg/m3, and 35.6 µg/m3, respectively. The study also discussed the model’s limitations in handling nonlinear patterns. Hence, in this research, the ARIMA model demonstrated its efficacy in terms of predicting short-term air quality very well, providing valuable insights for policymakers and environmental agencies. This research contributes to environmental science by offering a robust framework for air quality trend assessment and prediction, supporting sustainable urban development. However, in future, the accuracy of its predictions could be enhanced with a large dataset that is marked as constrained by the precision of the current predictions.