Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach

Beyza Yıldız; Enes Furkan Erkan

doi:10.21541/apjess.1968813

Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach

Abstract

Air pollution remains a major challenge for environmental sustainability and public health. PM2.5, defined as fine particulate matter, is a critical air pollution indicator because its small particle size can adversely affect the respiratory and cardiovascular systems. Monitoring and predicting PM2.5 levels are essential for air quality policy, risk reduction, and health-based decision-making. This study predicts PM2.5 levels in OECD countries using environmental, socioeconomic, urban, and transportation-related indicators. A country-year-level dataset comprising 264 observations from 24 OECD countries during 2013–2023 was constructed. PM2.5 was used as the dependent variable, while built-up area ratio, green coverage ratio, industrial indicators, population density, urbanization rate, economic density, environmental protection, environmental regulation, internal combustion engine vehicles, and electric vehicles were included as independent variables. Given the temporal structure, observations from 2013–2021 were used for model development, while observations from 2022–2023 constituted the holdout set. An AutoML approach was employed to compare 18 regression algorithms using performance metrics. The Extra Trees Regressor achieved the best cross-validated training performance, with an R² of 0.914, and obtained an R² of 0.971 on the temporal holdout set. SHAP analysis identified electric vehicles, population density, environmental protection, green coverage ratio, and internal combustion engine vehicles as the most influential predictors. Scenario analysis showed that increasing electric vehicles and green coverage while reducing internal combustion engine vehicles and built-up areas decreased predicted PM2.5 level by 5.42%. Overall, PM2.5 should be evaluated through the combined effects of transportation, demographic, environmental, and urban structural indicators rather than a single variable.

Keywords

Ethical Statement

This study did not involve any experiments on humans or animals. The research was conducted using country-level secondary data obtained from publicly available databases. Therefore, ethical approval was not required. The authors declare that the study was carried out in accordance with academic integrity and ethical research principles.

References

World Health Organization, “Ambient (outdoor) air pollution.” Accessed: Jun. 10, 2026. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/ambient-(outdoor)-air-quality-and-health
C. A. Pope and D. W. Dockery, “Health effects of fine particulate air pollution: lines that connect,” J. Air Waste Manag. Assoc., vol. 56, no. 6, pp. 709–742, Jun. 2006, doi: 10.1080/10473289.2006.10464485.
R. D. Brook et al., “Particulate matter air pollution and cardiovascular disease: An update to the scientific statement from the American Heart Association,” Circulation, vol. 121, no. 21, pp. 2331–2378, Jun. 2010, doi: 10.1161/CIR.0b013e3181dbece1.
A. J. Cohen et al., “Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases Study 2015,” The Lancet, vol. 389, no. 10082, pp. 1907–1918, May 2017, doi: 10.1016/S0140-6736(17)30505-6.
G. Hoek et al., “A review of land-use regression models to assess spatial variation of outdoor air pollution,” Atmos. Environ., vol. 42, no. 33, pp. 7561–7578, Oct. 2008, doi: 10.1016/j.atmosenv.2008.05.057.
J. Li et al., “The Cause of China’s Haze Pollution: City Level Evidence Based on the Extended STIRPAT Model,” Int. J. Environ. Res. Public. Health, vol. 19, no. 8, p. 4597, Apr. 2022, doi: 10.3390/ijerph19084597.
H. Alizadeh and V. Hosseini, “Contributions of local weather characteristics to elevated urban air pollution in northern cities in Canada,” Atmos. Environ., vol. 378, p. 122046, Aug. 2026, doi: 10.1016/j.atmosenv.2026.122046.
C. Thornton, F. Hutter, H. H. Hoos, and K. Leyton-Brown, “Auto-WEKA: Combined Selection and Hyperparameter Optimization of Classification Algorithms,” Mar. 06, 2013, arXiv: arXiv:1208.3719. doi: 10.48550/arXiv.1208.3719.

F. Hutter, L. Kotthoff, and J. Vanschoren, Eds., Automated Machine Learning: Methods, Systems, Challenges. Cham: Springer, 2019.
X. He, K. Zhao, and X. Chu, “AutoML: A Survey of the State-of-the-Art,” Knowl.-Based Syst., vol. 212, p. 106622, Jan. 2021, doi: 10.1016/j.knosys.2020.106622.
Y.-G. Lee, P.-H. Lee, S.-M. Choi, M.-H. An, and A.-S. Jang, “Effects of Air Pollutants on Airway Diseases,” Int. J. Environ. Res. Public. Health, vol. 18, no. 18, p. 9905, Sep. 2021, doi: 10.3390/ijerph18189905.
T. Zhang, G. Ahn, L. Wang, and X. Mao, “Protective effects of natural products against lung damage caused by fine particulate matter,” Environ. Pollut., vol. 384, p. 126942, Nov. 2025, doi: 10.1016/j.envpol.2025.126942.
J. Lelieveld, J. S. Evans, M. Fnais, D. Giannadaki, and A. Pozzer, “The contribution of outdoor air pollution sources to premature mortality on a global scale,” Nature, vol. 525, no. 7569, pp. 367–371, Sep. 2015, doi: 10.1038/nature15371.
R. Burnett et al., “Global estimates of mortality associated with long-term exposure to outdoor fine particulate matter,” Proc. Natl. Acad. Sci., vol. 115, no. 38, pp. 9592–9597, Sep. 2018, doi: 10.1073/pnas.1803222115.
V. Gupta, R. Choudhary, and A. Agarwal, “Integrating land use and traffic to spatial prediction of particulate matter,” Urban Clim., vol. 54, p. 101844, Mar. 2024, doi: 10.1016/j.uclim.2024.101844.
J. R. Krall et al., “Short-Term Associations of Road Density and Road Features with In-Vehicle PM2.5 during Daily Trips in the Washington, D.C. Metro Area,” Environments, vol. 11, no. 7, p. 135, Jul. 2024, doi: 10.3390/environments11070135.
X. Hu et al., “Estimating ground-level PM2.5 concentrations in the Southeastern United States using MAIAC AOD retrievals and a two-stage model,” Remote Sens. Environ., vol. 140, pp. 220–232, Jan. 2014, doi: 10.1016/j.rse.2013.08.032.
H. Karimian et al., “Evaluation of Different Machine Learning Approaches to Forecasting PM2.5 Mass Concentrations,” Aerosol Air Qual. Res., vol. 19, no. 6, pp. 1400–1410, 2019, doi: 10.4209/aaqr.2018.12.0450.
P. Patel et al., “A systematic study on PM2.5 and PM10 concentration prediction in air pollution using machine learning and deep learning model,” Environ. Chem. Ecotoxicol., vol. 7, pp. 1401–1415, Jan. 2025, doi: 10.1016/j.enceco.2025.07.001.
K. Kumar and B. P. Pande, “Air pollution prediction with machine learning: a case study of Indian cities,” Int. J. Environ. Sci. Technol., vol. 20, no. 5, pp. 5333–5348, May 2023, doi: 10.1007/s13762-022-04241-5.
A. Bekkar, B. Hssina, S. Douzi, and K. Douzi, “Air-pollution prediction in smart city, deep learning approach,” J. Big Data, vol. 8, no. 1, p. 161, Dec. 2021, doi: 10.1186/s40537-021-00548-1.
M. Lee et al., “Forecasting Air Quality in Taiwan by Using Machine Learning,” Sci. Rep., vol. 10, no. 1, p. 4153, Mar. 2020, doi: 10.1038/s41598-020-61151-7.
M. Méndez, M. G. Merayo, and M. Núñez, “Machine learning algorithms to forecast air quality: a survey,” Artif. Intell. Rev., vol. 56, no. 9, pp. 10031–10066, Sep. 2023, doi: 10.1007/s10462-023-10424-4.
A. Makhdoomi, M. Sarkhosh, and S. Ziaei, “PM2.5 concentration prediction using machine learning algorithms: an approach to virtual monitoring stations,” Sci. Rep., vol. 15, no. 1, p. 8076, Mar. 2025, doi: 10.1038/s41598-025-92019-3.
S. Z. Shogrkhodaei, S. V. Razavi-Termeh, and A. Fathnia, “Spatio-temporal modeling of PM2.5 risk mapping using three machine learning algorithms,” Environ. Pollut., vol. 289, p. 117859, Nov. 2021, doi: 10.1016/j.envpol.2021.117859.
J. Peng, H. Han, Y. Yi, H. Huang, and L. Xie, “Machine learning and deep learning modeling and simulation for predicting PM2.5 concentrations,” Chemosphere, vol. 308, p. 136353, Dec. 2022, doi: 10.1016/j.chemosphere.2022.136353.
K. Gohari, A. Sheidaei, M. Yitshak-sade, E. Colicino, and I. Kloog, “Exploring multivariate machine learning frameworks to parallelize PM2.5 simultaneous estimations across the continental United States,” Environ. Pollut., vol. 374, p. 126161, Jun. 2025, doi: 10.1016/j.envpol.2025.126161.
M. Feurer, A. Klein, K. Eggensperger, J. Springenberg, M. Blum, and F. Hutter, “Efficient and Robust Automated Machine Learning,” in Advances in Neural Information Processing Systems, Curran Associates, Inc., 2015. Accessed: Jun. 10, 2026. [Online]. Available: https://papers.nips.cc/paper_files/paper/2015/hash/11d0e6287202fced83f79975ec59a3a6-Abstract.html
R. S. Olson, N. Bartley, R. J. Urbanowicz, and J. H. Moore, “Evaluation of a Tree-based Pipeline Optimization Tool for Automating Data Science,” Mar. 20, 2016, arXiv: arXiv:1603.06212. doi: 10.48550/arXiv.1603.06212.
S. Lundberg and S.-I. Lee, “A Unified Approach to Interpreting Model Predictions,” Nov. 25, 2017, arXiv: arXiv:1705.07874. doi: 10.48550/arXiv.1705.07874.
S. M. Lundberg et al., “From local explanations to global understanding with explainable AI for trees,” Nat. Mach. Intell., vol. 2, no. 1, pp. 56–67, Jan. 2020, doi: 10.1038/s42256-019-0138-9.
S. Rao et al., “Future air pollution in the Shared Socio-economic Pathways,” Glob. Environ. Change, vol. 42, pp. 346–358, Jan. 2017, doi: 10.1016/j.gloenvcha.2016.05.012.
R. A. Silva, Z. Adelman, M. M. Fry, and J. J. West, “The Impact of Individual Anthropogenic Emissions Sectors on the Global Burden of Human Mortality due to Ambient Air Pollution,” Environ. Health Perspect., vol. 124, no. 11, pp. 1776–1784, Nov. 2016, doi: 10.1289/EHP177.
Q. Zhang, S. Ye, T. Ma, X. Fang, Y. Shen, and L. Ding, “Influencing factors and trend prediction of PM2.5 concentration based on STRIPAT-Scenario analysis in Zhejiang Province, China,” Environ. Dev. Sustain., vol. 25, no. 12, pp. 14411–14435, Dec. 2023, doi: 10.1007/s10668-022-02672-1.
OECD, “OECD Data Explorer.” Accessed: Jun. 10, 2026. [Online]. Available: https://data-explorer.oecd.org/
International Energy Agency, “Global EV Outlook 2026 – Analysis,” 2025. Accessed: Jun. 10, 2026. [Online]. Available: https://www.iea.org/reports/global-ev-outlook-2026
J. Benesty, J. Chen, Y. Huang, and I. Cohen, “Pearson Correlation Coefficient,” in Noise Reduction in Speech Processing, I. Cohen, Y. Huang, J. Chen, and J. Benesty, Eds., Berlin, Heidelberg: Springer, 2009, pp. 1–4. doi: 10.1007/978-3-642-00296-0_5.
J. M. Wooldridge, Introductory Econometrics: A Modern Approach. Mason, Ohio: Cengage Learning, 2013.
T. Chai and R. R. Draxler, “Root mean square error (RMSE) or mean absolute error (MAE)?,” Feb. 28, 2014, Copernicus GmbH. doi: 10.5194/gmdd-7-1525-2014.
D. Chicco, M. J. Warrens, and G. Jurman, “The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation,” PeerJ Comput. Sci., vol. 7, p. e623, Jul. 2021, doi: 10.7717/peerj-cs.623.
T. O. Hodson, “Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not,” Geosci. Model Dev., vol. 15, no. 14, pp. 5481–5487, Jul. 2022, doi: 10.5194/gmd-15-5481-2022.
S. Singh, V. Singh, A. Kumar, A. Singh, A. K. Srivastava, and V. Pathak, “A Decadal Study of PM2.5 Concentrations over Delhi using MERRA-2 and Ground Measurements: Predictive Insights via Machine Learning,” Indian J. Pure Appl. Phys. IJPAP, vol. 62, no. 9, pp. 764–778, Sep. 2024, doi: 10.56042/ijpap.v62i9.11443.
R. Ranjan, S. Chandel, D. Dhiman, H. Kadian, and G. Saxena, “Air Quality Prediction Using Ensemble Gradient Boosting Model,” in 2025 Global Conference on Information Technology and Communication Networks (GITCON), Aug. 2025, pp. 1–5. doi: 10.1109/GITCON65266.2025.11376870.
A. Makhdoomi, M. Sarkhosh, and S. Ziaei, “PM2.5 concentration prediction using machine learning algorithms: an approach to virtual monitoring stations,” Sci. Rep., vol. 15, no. 1, p. 8076, Mar. 2025, doi: 10.1038/s41598-025-92019-3.

Details

Primary Language

English

Subjects

Machine Learning Algorithms

Journal Section

Research Article

Authors

Beyza Yıldız This is me
0009-0003-9462-928X
Türkiye

Enes Furkan Erkan ^*
0000-0002-5470-8333
Türkiye

Publication Date

October 1, 2026

Submission Date

June 11, 2026

Acceptance Date

September 11, 2026

Published in Issue

Year 2026 Volume: 14 Number: 3

DOI

https://doi.org/10.21541/apjess.1968813

IZ

https://izlik.org/JA32JB76FE

Cite

RIS / Bibtex

APA

Yıldız, B., & Erkan, E. F. (2026). Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach. Academic Platform Journal of Engineering and Smart Systems, 14(3), 178-189. https://doi.org/10.21541/apjess.1968813

AMA

1.Yıldız B, Erkan EF. Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach. APJESS. 2026;14(3):178-189. doi:10.21541/apjess.1968813

Chicago

Yıldız, Beyza, and Enes Furkan Erkan. 2026. “Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach”. Academic Platform Journal of Engineering and Smart Systems 14 (3): 178-89. https://doi.org/10.21541/apjess.1968813.

EndNote

Yıldız B, Erkan EF (October 1, 2026) Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach. Academic Platform Journal of Engineering and Smart Systems 14 3 178–189.

IEEE

[1]B. Yıldız and E. F. Erkan, “Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach”, APJESS, vol. 14, no. 3, pp. 178–189, Oct. 2026, doi: 10.21541/apjess.1968813.

ISNAD

Yıldız, Beyza - Erkan, Enes Furkan. “Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach”. Academic Platform Journal of Engineering and Smart Systems 14/3 (October 1, 2026): 178-189. https://doi.org/10.21541/apjess.1968813.

JAMA

1.Yıldız B, Erkan EF. Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach. APJESS. 2026;14:178–189.

MLA

Yıldız, Beyza, and Enes Furkan Erkan. “Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach”. Academic Platform Journal of Engineering and Smart Systems, vol. 14, no. 3, Oct. 2026, pp. 178-89, doi:10.21541/apjess.1968813.

Vancouver

1.Beyza Yıldız, Enes Furkan Erkan. Machine Learning-Based Prediction of PM2.5 Levels in OECD Countries: An Integrated AutoML, SHAP, and Scenario Analysis Approach. APJESS. 2026 Oct. 1;14(3):178-89. doi:10.21541/apjess.1968813