PREDICTION OF CARBON DIOXIDE EMISSIONS FROM MOTOR VEHICLES USING GRU AND DBSCAN
DOI:
https://doi.org/10.15282/Keywords:
Climate Change, CO2 emissions, Deep learning, DBSCAN, Gated Recurrent UnitsAbstract
Extreme climatic conditions resulting from climate change pose significant challenges to mankind, prompting urgent action across various fronts. With vehicle emissions contributing substantially to carbon dioxide emissions, there is a pressing need for efficient prediction methods for informed mitigation strategies. While some research has been carried out on predicting carbon dioxide using machine learning, there have been few studies that explore Recurrent Neural Networks particularly those that can be modelled with a smaller dataset to improve emissions prediction. The study aimed to improve the prediction of carbon dioxide emissions by employing a new approach. The study was conducted on a Canadian dataset with 7,385 samples featuring motor vehicle parameters. Therefore, this research employs a deep learning model known as Gated Recurrent Units (GRU) for predictive modelling coupled with another algorithm called the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) for robust outlier detection and subsequent removal. Other algorithms that were used for preprocessing include label encoding and min-max normalisation. The study was evaluated using four metrics, namely root MSE (RMSE), mean square error (MSE), mean absolute error (MAE) and determination coefficient (R2). Compared with the highest-performing results from prior studies, our approach achieved a 4.59% increase in R² and a 2.14% increase in R, alongside reductions in MSE and RMSE by 0.0008766 and 0.0015804, respectively. Thus, showing superior handling of data irregularities and temporal patterns. These results highlight the potential for improved emissions predictions, which can be instrumental in refining carbon tax calculations and informing environmental policy. Future works will involve expanding the dataset to include the age of the motor vehicle and using hybrid deep learning models.
Downloads
References
[1] Jin H. Prediction of direct carbon emissions of Chinese provinces using artificial neural networks. PLoS One. 2021;16(5):e0236685.
[2] Shah IH, Manzoor MA, Jinhui W, Li X, Hameed MK, Rehaman A, et al. Comprehensive review: Effects of climate change and greenhouse gases emission relevance to environmental stress on horticultural crops and management. J Environ Manage. 2024;351:119978.
[3] Bajracharya TR, Shakya SR, Sharma A. Climate change and sustainable energy systems. In: Handbook of Energy and Environmental Security [Internet]. Elsevier; 2022 [cited 2024 Feb 29]. p. 531–545. Available from: https://linkinghub.elsevier.com/retrieve/pii/B9780128240847240010
[4] Goswami A, Kapoor HS, Jangir RK, Ngigi CN, Nowrouzi-Kia B, Chattu VK. Impact of economic growth, trade openness, urbanization and energy consumption on carbon emissions: A study of India. Sustainability. 2023;15(11):9025.
[5] Aminzadegan S, Shahriari M, Mehranfar F, Abramović B. Factors affecting the emission of pollutants in different types of transportation: A literature review. Energy Rep. 2022;8:2508–2529.
[6] Emami Javanmard M, Tang Y, Wang Z, Tontiwachwuthikul P. Forecast energy demand, CO2 emissions and energy resource impacts for the transportation sector. Appl Energy. 2023;338:120830.
[7] Bartlett VL, Doernberg H, Mooghali M, Gupta R, Wallach JD, Nyhan K, et al. Published research on the human health implications of climate change between 2012 and 2021: A cross-sectional study. BMJ Med. 2024;3(1):e000627.
[8] Boogaard H, Patton AP, Atkinson RW, Brook JR, Chang HH, Crouse DL, et al. Long-term exposure to traffic-related air pollution and selected health outcomes: A systematic review and meta-analysis. Environ Int. 2022;164:107262.
[9] Al-Nefaie AH, Aldhyani THH. Predicting CO2 emissions from traffic vehicles for sustainable and smart environment using a deep learning model. Sustainability. 2023;15(9):7615.
[10] Zhang R, Wang Y, Pang Y, Zhang B, Wei Y, Wang M, et al. A deep learning micro-scale model to estimate the CO2 emissions from light-duty diesel trucks based on real-world driving. Atmosphere. 2022;13(9):1466.
[11] Azeez O, Pradhan B, Shafri H, Shukla N, Lee CW, Rizeei H. Modeling of CO emissions from traffic vehicles using artificial neural networks. Appl Sci. 2019;9(2):313.
[12] Chadha AS, Shinde Y, Sharma N, De PK. Predicting CO2 emissions by vehicles using machine learning. In: Data Management, Analytics and Innovation [Internet]. Singapore: Springer; 2023 [cited 2024 Mar 4]. p. 197–207. Available from: https://link.springer.com/10.1007/978-981-19-2600-6_14
[13] Shah S, Thakar S, Jain K, Shah B, Dhage S. A comparative study of machine learning and deep learning techniques for prediction of CO2 emission in cars [Internet]. arXiv; 2022 [cited 2023 Nov 1]. Available from: http://arxiv.org/abs/2211.08268
[14] Jia T, Zhang P, Chen B. A microscopic model of vehicle CO2 emissions based on deep learning: A spatiotemporal analysis of taxicabs in Wuhan, China. IEEE Trans Intell Transp Syst. 2022;23(10):18446–18455.
[15] Ghahramani M, Pilla F. Analysis of carbon dioxide emissions from road transport using taxi trips. IEEE Access. 2021;9:98573–98580.
[16] Rivera-Campoverde ND, Muñoz-Sanz JL, Arenas-Ramirez BDV. Estimation of pollutant emissions in real driving conditions based on data from OBD and machine learning. Sensors. 2021;21(19):6344.
[17] Li X, Ren A, Li Q. Exploring patterns of transportation-related CO2 emissions using machine learning methods. Sustainability. 2022;14(8):4588.
[18] Natarajan Y, Wadhwa G, Sri Preethaa KR, Paul A. Forecasting carbon dioxide emissions of light-duty vehicles with different machine learning algorithms. Electronics. 2023;12(10):2288.
[19] Abdulmalik R, Srivastava G. Forecasting of transportation-related CO2 emissions in Canada with different machine learning algorithms. Adv Artif Intell Mach Learn. 2023;3(3):1295–1312.
[20] Mangla H, Sharma A. Fuel emission detection using machine learning. Int J Innov Res Technol. 2021;3(6):1–6.
[21] Mądziel M. Liquified petroleum gas-fuelled vehicle CO2 emission modelling based on portable emission measurement system, on-board diagnostics data and gradient-boosting machine learning. Energies. 2023;16(6):2754.
[22] Bappon SD, Dey A, Sabuj SM, Das A. Toward a machine learning approach to predict the CO2 rating of fuel-consuming vehicles in Canada. In: 2022 25th International Conference on Computer and Information Technology (ICCIT). IEEE; 2022. p. 384–389.
[23] Sami O, Elsheikh Y, Almasalha F. The role of data preprocessing techniques in improving machine learning accuracy for predicting coronary heart disease. Int J Adv Comput Sci Appl. 2021;12(6).
[24] Dahouda MK, Joe I. A deep-learned embedding technique for categorical features encoding. IEEE Access. 2021;9:114381–114391.
[25] Kosaraju N, Sankepally SR, Mallikharjuna Rao K. Categorical data: Need, encoding, selection of encoding method and its emergence in machine learning models. In: Proceedings of International Conference on Data Science and Applications. Singapore: Springer; 2023. p. 369–382.
[26] Sinsomboonthong S. Performance comparison of new adjusted min-max with decimal scaling and statistical column normalisation methods for artificial neural network classification. Int J Math Math Sci. 2022;2022:1–9.
[27] Rodriguez MZ, Comin CH, Casanova D, Bruno OM, Amancio DR, Costa LDF, et al. Clustering algorithms: A comparative approach. PLoS One. 2019;14(1):e0210236.
[28] Lopes HEG, Gosling MDS. Cluster analysis in practice: Dealing with outliers in managerial research. Rev Adm Contemp. 2021;25(1):e200081.
[29] Yigit G, Amasyali MF. Simple but effective GRU variants. In: 2021 International Conference on Innovations in Intelligent Systems and Applications (INISTA). IEEE; 2021. p. 1–6.
[30] Hasnain M, Jeong SR, Pasha MF, Ghani I. Performance anomaly detection in web services: An RNN-based approach using dynamic quality of service features. Comput Mater Contin. 2020;64(2):729–752.
[31] Carvalho TP, Soares FAAMN, Vita R, Francisco RDP, Basto JP, Alcalá SGS. A systematic literature review of machine learning methods applied to predictive maintenance. Comput Ind Eng. 2019;137:106024.
[32] Hodson TO. Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not. Geosci Model Dev. 2022;15(14):5481–5490.
[33] Zameer A, Jaffar F, Shahid F, Muneeb M, Khan R, Nasir R. Short-term solar energy forecasting: Integrated computational intelligence of LSTMs and GRU. PLoS One. 2023;18(10):e0285410.
[34] Kvalseth TO. Note on the R² measure of goodness of fit for nonlinear models. Bull Psychon Soc. 1983;21(1):79–80.
[35] Haut N, Banzhaf W, Punch B. Correlation versus RMSE loss functions in symbolic regression tasks [Internet]. arXiv; 2022 [cited 2023 Nov 14]. Available from: https://arxiv.org/abs/2205.15990
[36] Handelman GS, Kok HK, Chandra RV, Razavi AH, Huang S, Brooks M, et al. Peering into the black box of artificial intelligence: Evaluation metrics of machine learning methods. AJR Am J Roentgenol. 2019;212(1):38–43.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 The Author(s)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
