ARIMA and Machine Learning–Based Forecasting of Electric and Hybrid Vehicle Registrations in Türkiye

Year 2025, Volume: 8 Issue: 2, 121 - 131, 26.12.2025

Mehmet Ümit Ak , Ramazan Öz

https://doi.org/10.70030/sjmakeu.1838078

https://izlik.org/JA67XF94HM

Abstract

The rapid proliferation of electric and hybrid vehicles has become one of the fundamental components of Turkey's sustainable transportation policies. This trend necessitates reliable future projections in terms of energy infrastructure planning, charging station placement, environmental impact reduction, and transformation within the automotive sector. In this study, time series analyses of the number of electric (EV) and hybrid (HV) cars registered in Turkey monthly between 2020M01-2025M10 were obtained, and a forecasting model was developed for monthly periods in 2026. The Autoregressive Integrated Moving Average (ARIMA) model and machine learning algorithms such as Prophet, LSTM, and SVR were used as methods. The findings show that machine learning-based models produce lower prediction errors than ARIMA and that SVR and Prophet perform better. The monthly forecasts for 2026 reveal that electric vehicle (EV) registrations in Turkey will continue to show a strong and accelerating growth trend in the coming period, while the growth rate of hybrid vehicle (HV) registrations will slow down significantly, showing a more limited increase or a stagnant trend.

Keywords

Electric Vehicle , Hybrid Vehicle , ARIMA , Time-Series Forecasting , Machine Learning

References

• Türkiye İstatistik Kurumu (TÜİK). (2025). Motorlu Kara Taşıtları İstatistikleri (2020–2025). Türkiye İstatistik Kurumu.
• Kontopoulou, V. I., Panagopoulos, A. D., Kakkos, I., & Matsopoulos, G. K. (2023). A review of ARIMA vs. machine learning approaches for time series forecasting in data-driven networks. Future Internet, 15(8), 255. https://doi.org/10.3390/fi15080255
• Hossain, M. L., Shams, S. M., & Ullah, S. M. (2025). Time-series and deep learning approaches for renewable energy forecasting in Dhaka: a comparative study of ARIMA, SARIMA, and LSTM models. Discover Sustainability, 6(1), 1-25. https://doi.org/10.1007/s43621-025-02060-5
• Szostek, K., Mazur, D., Drałus, G., & Kusznier, J. (2024). Analysis of the Effectiveness of ARIMA, SARIMA, and SVR Models in Time Series Forecasting: A Case Study of Wind Farm Energy Production. Energies (19961073), 17(19). https://doi.org/10.3390/en17194803
• Zaino, R., Ahmed, V., Alhammadi, A. M., & Alghoush, M. (2024). Electric vehicle adoption: A comprehensive systematic review of technological, environmental, organizational and policy impacts. World electric vehicle journal, 15(8), 375. https://doi.org/10.3390/wevj15080375
• Bas, J., Cirillo, C., & Cherchi, E. (2021). Classification of potential electric vehicle purchasers: A machine learning approach. Technological Forecasting and Social Change, 168, 120759. https://doi.org/10.1016/j.techfore.2021.120759
• Murat, C. (2025). AI-Driven Optimization of a 2.45 Ghz Microwave Drying System for Raw Corn: Enhancing Efficiency and Uniformity via Hybrid CNN-RNN Control. Black Sea Journal of Engineering and Science, 9(1), 11-12.
• Murat, C., Kaya, A., & Palandoken, M. (2025). Non-foster microwave circuit design for wide band active impedance matching operation. AEU-International Journal of Electronics and Communications, 155954. https://doi.org/10.1016/j.aeue.2025.155954
• Murat, C., Palandoken, M., Kaya, I., & Kaya, A. (2021). A novel ISM band reflector type applicator design for microwave ablation systems. Electromagnetic biology and medicine, 40(2), 286-300. https://doi.org/10.1080/15368378.2021.1885434
• Çakmak, T., Murat, C., & Ustabaş, İ. (2025). Performance Comparison of Carbon Footprint and Dielectric Properties of Geopolymer Mortars with Portland Cement Mortars. Silicon, 1-23. https://doi.org/10.1007/s12633-025-03538-7
• Çakmak, T., Murat, C., & Ustabaş, İ. (2025, July). A novel study on dielectric aspect of mortars about multi-binder structures consisting of divergent binders. In Structures (Vol. 77, p. 109066). Elsevier. https://doi.org/10.1016/j.istruc.2025.109066
• Saraç, M. S., & Ertürk, M. A. (2025). Forecasting the Number of Electric Vehicles in Turkey Towards 2030: SARIMA Approach. Energies, 18(18), 4808. https://doi.org/10.3390/en18184808
• Yuan, C. Z., & Ling, S. K. (2020). Long short-term memory model-based agriculture commodity price prediction application. In Proceedings of the 2020 2nd International Conference on Information Technology and Computer Communications (pp. 43-49). https://doi.org/10.1145/3417473.3417481
• Mahmud, H., Rahman, Z. T., Rahman, B., & Haque, S. N. (2025). Predicting Stock Prices in the Banking Sector of Bangladesh: A Machine Learning Approach. https://doi.org/10.21203/rs.3.rs-6791982/v1
• Dhankhar, S., Dhankhar, N., Sandhu, V., & Mehla, S. (2024). Forecasting electric vehicle sales with the ARIMA and exponential smoothing methods: The case of India. Transportation in Developing Economies, 10(2), 32. https://doi.org/10.1007/s40890-024-00216-y
• Cheng, N., Zheng, P., Ruan, X., & Zhu, Z. (2023). Electric vehicle charging load prediction based on variational mode decomposition and Prophet-LSTM. Frontiers in Energy Research, 11, 1297849. https://doi.org/10.3389/fenrg.2023.1297849
• Veysi, S., Moshfeghi, M., Sadrfaridpour, A., & Emamy, P. (2025). Electric Vehicle Sales Forecast for the UK: Integrating Machine Learning, Time Series Models, and Global Trends. Algorithms, 18(7), 430. https://doi.org/10.3390/a18070430
• Liu, J., Pan, H., Luo, R., Chen, H., Tao, Z., & Wu, Z. (2025). An electric vehicle sales hybrid forecasting method based on an improved sentiment analysis model and secondary decomposition. Engineering Applications of Artificial Intelligence, 150, 110561. https://doi.org/10.1016/j.engappai.2025.110561
• Box, G. E., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). Time series analysis: forecasting and control. John Wiley & Sons. https://doi.org/10.1111/jtsa.12194
• Taylor, S. J., & Letham, B. (2018). Forecasting at scale. The American Statistician, 72(1), 37-45. https://doi.org/10.7287/peerj.preprints.3190v2
• Arslan, S. (2022). A hybrid forecasting model using LSTM and Prophet for energy consumption with decomposition of time series data. PeerJ Computer Science, 8, e1001. https://doi.org/10.7717/peerj-cs.1001
• Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
• Siami-Namini, S., Tavakoli, N., & Namin, A. S. (2018, December). A comparison of ARIMA and LSTM in forecasting time series. In 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA) (pp. 1394-1401). Ieee. https://doi.org/10.1109/ICMLA.2018.00227
• Sezer, O. B., Gudelek, M. U., & Ozbayoglu, A. M. (2020). Financial time series forecasting with deep learning: A systematic literature review: 2005–2019. Applied Soft Computing, 90, 106181. https://doi.org/10.1016/j.asoc.2020.106181
• Drucker, H., Burges, C. J., Kaufman, L., Smola, A., & Vapnik, V. (1996). Support vector regression machines. Advances in neural information processing systems, 9.
• Smola, A. J., & Schölkopf, B. (2004). A tutorial on support vector regression. Statistics and Computing, 14(3), 199-222. https://doi.org/10.1023/B:STCO.0000035301.49549.88