Integrated AI Framework for Sustainable Environmental Management: Multivariate Air Pollution Interpretation and Prediction Using Ensemble and Deep Learning Models

Accurate prediction, forecasting and interpretability of air pollutant concentrations are important for sustainable environmental management and protecting public health. An integrated artificial intelligence (AI) framework is proposed to predict, forecast and analyse six major air pollutants, such as particulate matter concentrations (PM 2.5 and PM 10 ), ground-level ozone (O 3 ), carbon monoxide (CO), nitrogen dioxide (NO 2 ), and sulphur dioxide (SO 2 ), using a combination of ensemble and deep learning models. Five years of hourly air quality and meteorological data are analysed through correlation and Granger causality tests to uncover pollutant interdependencies and driving factors. The results of the Pearson correlation analysis reveal strong positive associations among primary pollutants (PM 2.5 –PM 10 , CO–nitrogen oxides NO x and VOCs) and inverse correlations between O 3 and NO x (NO and NO 2 ), confirming typical photochemical behaviour. Granger causality analysis further identified NO 2 and NO as key causal drivers influencing other pollutants, particularly O 3 formation. Among the 23 tested AI models for prediction, XGBoost, Random Forest, and Convolutional Neural Networks (CNNs) achieve the best performance for different pollutants. NO 2 prediction using CNNs displays the highest accuracy in testing (R 2 = 0.999, RMSE = 0.66 µg/m 3 ), followed by PM 2.5 and PM 10 with XGBoost (R 2 = 0.90 and 0.79 during testing, respectively). The Air Quality Index (AQI) analysis shows that SO 2 and PM 10 are the dominant contributors to poor air quality episodes, while ozone peaks occur during warm, high-radiation periods. The interpretability analysis based on Shapley Additive exPlanations (SHAP) highlights the key influence of relative humidity, temperature, solar brightness, and NO x species on pollutant concentrations, confirming their meteorological and chemical relevance. Finally, a deep-NARMAX model was applied to forecast the next horizons for the six air pollutants studied. Six formulas were elaborated using input data at times (t, t − 1, t − 2, …, t − n) to forecast a horizon of (t + 1) hours for single-step forecasting. For multi-step forecasting, the forecast is extended iteratively to (t + 2) hours and beyond. A recursive strategy is adopted for this purpose, whereby the forecast at (t + 1) is fed back as an input to generate the forecasts at (t + 2), and so forth. Overall, this integrated framework combines predictive accuracy with physical interpretability, offering a powerful data-driven tool for air quality assessment and policy support. This approach can be extended to real-time applications for sustainable environmental monitoring and decision-making systems.