A Survey on AQI Prediction and Analysis of Traffic Impact on Air Quality

Abstract

Air pollution, a matter of grave concern, has caught the cities by storm, and transportation is taking a major role in that process. Therefore, accurate Air Quality Index prediction is necessary for successful management of air pollution, urban planning, and public awareness. A number of approaches, mostly using machine learning algorithms, have been proposed recently for precise prediction of the Air Quality Index by describing complex correlations between air pollutant concentration and atmospheric variables. However, the role of traffic-related variables in Air Quality Index prediction is often ignored. This survey provides a comprehensive overview of the relevant methods of machine learning and deep learning applied to air quality index (AQI) forecasting, primarily focusing on the works that involve or address parameters of vehicle traffic, like density and speed. The relevant literature is provided with a systematic arrangement on the basis of methods of models, sources of data, and methods of combination of traffic parameters. A comparative study of the research conducted earlier shows the key results and current deficiencies. In addition to that, the survey points out the key research deficiencies and research avenues that can be addressed to develop effective traffic-aware AQI forecasting systems.

Introduction

Air pollution has become a major environmental and public health challenge due to rapid industrialization, urbanization, and population growth. Vehicular emissions are a key contributor, releasing harmful pollutants such as PM2.5, PM10, NOx, CO, and VOCs, which are linked to serious health impacts including respiratory and cardiovascular diseases. Accurate Air Quality Index (AQI) prediction is therefore essential for public awareness, policy planning, and preventive action.

Conventional AQI prediction models rely mainly on meteorological and environmental parameters but often fail to capture the sudden pollution spikes caused by traffic congestion during rush hours. This limitation reduces their accuracy and usefulness for real-time decision-making. Machine learning (ML) techniques—such as Random Forest, Support Vector Regression, Artificial Neural Networks, LSTM, CNN, and hybrid models—have demonstrated superior performance in modeling nonlinear and temporal pollutant patterns. However, many existing models suffer from high complexity, limited interpretability, regional dependence, or incomplete feature integration.

Recent research highlights the importance of incorporating traffic-related factors like vehicle density, congestion rate, and average speed into AQI prediction models. While some studies have integrated traffic data using regression and ensemble techniques and achieved improved prediction accuracy, they remain limited by region-specific datasets and insufficient handling of dynamic urban conditions. Overall, the literature emphasizes the need for intelligent, traffic-aware AQI prediction models that integrate environmental, meteorological, and traffic data to support effective smart city air quality monitoring and management.

Conclusion

This review paper comprehensively discussed machine learning-based approaches for the prediction of Air Quality Index, with the primary concentration on the impact that vehicle traffic patterns have on air quality. By integrating these research results from over fifty peer-reviewed publications, a conclusion was drawn that vehicle congestion level, vehicle density, and speed-related factors improve the accuracy of AQI prediction when combined with prevailing AQI data. From the literature review, it has been found that ensemble methods and deep learning techniques have greater accuracy in nonlinear and dynamic variations of air quality than compared with traditional statistical methods. Some issues still exist, such as a lack of interpretability for proposed methods, regional adaptability of methods, or real-time lack of standardized traffic data. Further studies are required in improving regional model interpretability, traffic data merging methods, or discovering relationships between traffic patterns and concentration levels of certain air pollutants.

References

[1] Y?ld?r?m Özüpak, F., Alpsalaz, F., & Aslan, E. (2025). “Air quality forecasting using machine learning: Comparative analysis and ensemble strategies for enhanced prediction”. Volume 236, Article 464. [2] Kumari, P. M. K., Manjaiha, D. H., & Ashwini, K. M. (2024). “Traffic prediction based on air quality in IoT-based smart city using regression and ensemble techniques”. In IIP Series, Volume 3, Book 2, Part 1, Chapter 7 (e-ISBN: 978-93-6252-830-8). [3] Naz, F., Yousuf, M., Zia, I., & Tasadduq, I. A. (2023). “Comparative analysis of deep learning and statistical models for air pollutants prediction in urban areas”. IEEE Access. https://doi.org/10.1109/ACCESS.2023.3289153 [4] Maltare, N., & Vahora, S. (2023). “Air quality index prediction using machine learning for Ahmedabad city”. Data in Chemical Engineering, 100093. https://doi.org/10.1016/j.dche.2023.100093 [5] Setiawan, A., et al. (2023). “The impact of traffic conditions on air quality: A case study of IoT monitoring and machine learning analysis”. https://doi.org/10.1109/ICISS59129.2023.10291616 [6] A. Bose, I. Roy Chowdhury “Investigating the association between air pollutants\' concentration and meteorological parameters in a rapidly growing urban center of West Bengal, India: a statistical modeling-based approach” Model. Earth Syst. Environ., 1 (2023), pp. 1-16, 10.1007/S40808-022-01670-6/FIGURES/9 [7] K. Chaudhari, D. Joshi, P. Harugade, K. Jambusariya and V. Tiwari, “Predicting Mumbai\'s Air Quality Index by Machine Learning,” INCET, Belgaum, India, 2023, pp. 1–5. [8] M. S. Ram, C. Reshmasri, S. Shahila and J. V. P. Saketh, “Air Quality Prediction using Machine Learning Algorithm,” ICSCDS, Erode, India, 2023, pp. 316–321. [9] Alkabbani, H., Ramadan, A., Zhu, Q. and Elkamel, A., 2022. An improved air quality index machine learning-based forecasting with multivariate data imputation approach. Atmosphere, 13 ( 7 ), p. 1144. [10] Balogun A.-L., Tella A. “Modelling and investigating the impacts of climatic variables on ozone concentration in Malaysia using correlation analysis with random forest, decision tree regression, linear regression, and support vector regression” Chemosphere, 299 (2022), Article 134250, 10.1016/j.chemosphere.2022.134250 [11] A. Bekkar, B. Hssina, S. Douzi, K. Douzi “Air-pollution prediction in smart city, deep learning approach” J. Big Data, 8 (2021), pp. 1-21, 10.1186/S40537-021-00548-1/FIGURES/17 [12] Saikiran, K., Lithesh, G., Srinivas, B. and Ashok, S., 2021, September. Prediction of Air Quality Index Using Supervised Machine Learning Algorithms. In 2021 2nd International Conference on Advances in Computing, Communication, Embedded and Secure Systems (ACCESS) (pp. 1 - 4 ). IEEE. [13] Kekulanadara, K. M. O. V. K., Kumara, B. T. G. S., & Kuhaneswaran, B. (2021). “Air quality forecasting and classification using machine learning based on Indian urban pollutant”. [14] Gopu P., Panda R.R., Nagwani N.K.Time series analysis using ARIMA model for air pollution prediction in hyderabad city of India Advances in Intelligent Systems and Computing, Springer, Singapore (2021), pp. 47-56, 10.1007/978-981-33- 6912-2_5 [15] Murugan, R., & Palanichamy, N. (2021). “Smart city air quality prediction using machine learning”. In 2021 5th International Conference on Intelligent Computing and Control Systems (ICICCS). https://doi.org/10.1109/ICICCS51141.2021.9432074 [16] Gowri, D., & Anandhasilambarasan, D. (2021). “Prediction of air pollution in smart cities using machine learning techniques. International Journal for Research in Applied Science and Engineering Technology”, 9(12). https://doi.org/10.22214/ijraset.2021.39241 [17] Madan, T., Sagar, S., & Virmani, D. (2020). “Air quality forecasting and classification using machine learning based on Indian urban pollutant”. In 2020 2nd International Conference on Advances in Computing, Communication Control and Networking (ICACCCN). https://doi.org/10.1109/ICACCCN51052.2020.9362912 [18] Dobrea, D., et al. (2020). “Air quality prediction using machine learning algorithms – A review”. [19] Halsana, S. (2020). “Air quality prediction model using supervised machine learning algorithms. International Journal of Scientific Research in Computer Science, Engineering and Information Technology”, 8, 190–201. https://doi.org/10.32628/CSEIT206435 [20] Chang Y.-S., Chiao H.-T., Abimannan S., Huang Y.-P., Tsai Y.-T., Lin K.-M. “An LSTM-based aggregated model for air pollution forecasting Atmos”. Pollut. Res., 11 (8) (2020), pp. 1451-1463, 10.1016/j.apr.2020.05.015 [21] World Health Organization, \"Ambient (outdoor) air pollution,\" WHO, 2020. [Online]. Available: https://www.who.int/airpollution/ambient/about/en/. [Accessed: 06-Jan-2026]. [22] Microsoft, \"A fresher breeze: How AI can help improve air quality,\" Microsoft News, 20-Aug-2019. [Online]. Available: https://news.microsoft.com/europe/2019/08/20/a-fresherbreeze-how-ai-can-help-improve-air-quality/. [Accessed: 06-Jan-2026]. [23] G. Kaur, J. Gao, S. Chiao, S. Lu, and G. Xie, \"Air Quality Prediction: Big Data and Machine Learning Approaches,\" International Journal of Environmental Science and Development, vol. 9, no. 1, pp. 8-16, 2018. [24] \"Artificial Intelligence for Cleaner Air in Smart Cities,\" PDX Engineering, 28-Mar-2019. [Online]. Available: https://www.pdxeng.ch/2019/03/28/artificial-intelligence-for-cleanerair-in-smart-cities/.[Accessed: 06-Jan-2026]. [25] S. Garzon, S. Walther, B. Deva, and A. Küpper, \"Urban Air Pollution Alert Service for Smart Cities,\" 2018. [26] A. Dorogush, V. Ershov, and A. Gulin, \"CatBoost: Gradient Boosting with Categorical Features Support,\" arXiv preprint, vol. arXiv:1810.11363, 2018. [27] N. Alharbi and B. Soth, \"Roles and Challenges of Network Sensors in Smart Cities,\" IOP Conference Series: Earth and Environmental Science, vol. 322, no. 1, 2019. [28] R. Sánchez-Corcuera, A. Nuñez-Marcos, J. Sesma-Solance, A. Bilbao Jayo, R. Mulero, U. Zulaika, and A. Almeida, \"Smart Cities Survey: Technologies, Application Domains and Challenges for the Cities of the Future,\" International Journal of Distributed Sensor Networks, vol. 15, no. 6, 2019. [29] \"Smart Cities Air Quality,\" Aeroqual. [Online]. Available: https://www.aeroqual.com/smart-cities-air-quality. [Accessed: 06-Jan-2026]. [30] \"Beijing Multi-Site Air Quality Data,\" UCI Machine Learning Repository. [Online]. Available: https://archive.ics.uci.edu/ml/datasets/Beijing+Multi-Site+Air-Quality+Data. [Accessed: 06-Jan-2026].

Copyright

Copyright © 2026 Ashwini Bhavsar, Damayanti Kale, Rajvi Damrekar, Mrunali Bawane, Arya Madhe. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.