PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS

Rana Bedir Urfalı; Ersin Kaya

doi:10.17780/ksujes.1817646

TR EN

ELEKTROMOBİLİTEDE TAHMİNE DAYALI MODELLEME: BİR ZAMAN SERİSİ ANALİZİ

Öz

Bu çalışma, gelecekteki eğilimleri tahmin etmek amacıyla, bir ildeki elektrikli bisiklet kullanımını zaman serisi modelleri ile analiz etmektedir. Sürdürülebilir ulaşımın kritik bir parçası olan elektrikli bisikletler, kentsel hareketliliği ve talep tahminini iyileştirir; bu da elektrikli bisiklet paylaşım sistemleri yönetimi için hayati önem taşır. Tahminler ilk olarak yalnızca bisiklet kullanım verileriyle tek değişkenli modeller (ARIMA, Prophet, SSA) kullanılarak yapıldı. Ardından, sıcaklık, yağış ve rüzgar hızı gibi dış etkenleri dahil eden çok değişkenli yaklaşımlara (SARIMAX, Multivariate Prophet, MSSA) geçildi. Bu sayede veri seti, çevresel etkileri yakalamak için meteorolojik değişkenlerle birleştirildi. Model performansı RMSE ve MAE metrikleriyle değerlendirildi. Sonuçlar, doğrulukta bölgesel farklılıklar olduğunu gösterdi: ARIMA, tek değişkenli modeller arasında dört bölgede en iyisiydi, ancak SARIMAX ve Multivariate Prophet çoğu bölgede üstün tahminler üretti. Ayrıca MSSA, dış etkileri dahil etmenin faydasını vurgulayarak 13 bölgede SSA'yı sürekli olarak geride bıraktı. Genel bulgular, hava durumu verilerini entegre etmenin tahmin hassasiyetini artırdığını ve daha iyi operasyonel stratejileri desteklediğini ortaya koymaktadır. Bu yaklaşım, elektrikli bisiklet paylaşım hizmetlerini ve şehir içi ulaşım planlamasını optimize ederek daha verimli kaynak kullanımına ve yüksek kullanıcı memnuniyetine katkıda bulunmaktadır.

Anahtar Kelimeler

PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS

Öz

This study analyzes electric bicycle usage in a province with time series models to forecast future trends. Electric bicycles, as an essential element of sustainable transportation, enhance urban mobility and reduce traffic congestion. Accurate demand forecasting is crucial for the operational planning and administration of e-bike sharing systems. In this study, forecasts were first made using only bicycle usage data with univariate time series models (ARIMA, Prophet, SSA). Next, multivariate approaches (SARIMAX, Multivariate Prophet, MSSA) incorporated external factors such as temperature, precipitation, and wind speed. The dataset, therefore, combined bicycle usage records with meteorological variables to capture environmental impacts on demand. Model performance was assessed using RMSE and MAE metrics. Results showed regional variations in accuracy: ARIMA performed best among univariate models in four regions, while SARIMAX and Multivariate Prophet produced superior forecasts in most regions. Furthermore, MSSA consistently outperformed SSA in 13 regions, highlighting the benefit of including external influences. Overall, the findings demonstrate that integrating weather data improves forecast precision and supports better operational strategies. This contributes to optimizing e-bike sharing services and urban transport planning, enabling more efficient resource use and greater user satisfaction.

Anahtar Kelimeler

Kaynakça

Afriyie, J. K., Twumasi-Ankrah, S., Gyamfi, K. B., Arthur, D., & Pels, W. A. (2020). Evaluating the performance of unit root tests in single time series processes. Mathematics and Statistics, 8(6), 656-664. https://doi.org/10.13189/ms.2020.080605
Agarwal, A., Alomar, A., & Shah, D. (2022). On multivariate singular spectrum analysis and its variants. ACM SIGMETRICS Performance Evaluation Review, 50(1), 79-80. https://doi.org/10.1145/3547353.3526952
Alagade, A., & Sahu, M. (2025). Satellite-based assessment and forecasting of greenhouse gas (GHG) concentrations in Indian megacities using advanced statistical methods. Environmental Science and Pollution Research, 32(24), 15006-15024. https://doi.org/10.1007/s11356-025-36583-1
Alencar, V. A., Pessamilio, L. R., Rooke, F., Bernardino, H. S., & Borges Vieira, A. (2021). Forecasting the carsharing service demand using uni and multivariable models. Journal of Internet Services and Applications, 12(1), 4. https://doi.org/10.1186/s13174-021-00137-8
Alharbi, F. R., & Csala, D. (2022). A seasonal autoregressive integrated moving average with exogenous factors (SARIMAX) forecasting model-based time series approach. Inventions, 7(4), 94. https://doi.org/10.3390/inventions7040094
Alhussan, A. A., Khafaga, D. S., Abotaleb, M., Mishra, P., & El-Kenawy, E. S. M. (2024). Global potato production forecasting based on time series analysis and advanced waterwheel plant optimization algorithm. Potato Research, 67(4), 1965-2000. https://doi.org/10.1007/s11540-024-09728-x
Almazrouee, A. I., Almeshal, A. M., Almutairi, A. S., Alenezi, M. R., Alhajeri, S. N., & Alshammari, F. M. (2020). Forecasting of electrical generation using prophet and multiple seasonality of holt–winters models: A case study of Kuwait. Applied Sciences, 10(23), 8412. https://doi.org/10.3390/app10238412
Chowdhury, M. S., & Hafsa, B. (2022). Multi-decadal land cover change analysis over Sundarbans Mangrove Forest of Bangladesh: A GIS and remote sensing based approach. Global Ecology and Conservation, 37, e02151. https://doi.org/10.1016/j.gecco.2022.e02151

Cocca, M., Teixeira, D., Vassio, L., Mellia, M., Almeida, J. M., & Couto da Silva, A. P. (2020). On car-sharing usage prediction with open socio-demographic data. Electronics, 9(1), 72. https://doi.org/10.3390/electronics9010072
Fattah, J., Ezzine, L., Aman, Z., El Moussami, H., & Lachhab, A. (2018). Forecasting of demand using ARIMA model. International journal of engineering business management, 10, 1847979018808673. https://doi.org/10.1177/1847979018808673
Iftikhar, H., Khan, F., Rodrigues, P. C., Alharbi, A. A., & Allohibi, J. (2025). Forecasting of inflation based on univariate and multivariate time series models: an empirical application. Mathematics, 13(7), 1121. https://doi.org/10.3390/math13071121
İnaç, H. (2024). PREDICTION OF DRIVING TIME OF ELECTRIC SCOOTER (E-SCOOTER) DRIVERS BY MACHINE LEARNING. Elektronik Sosyal Bilimler Dergisi, 23(91), 1041-1057. https://doi.org/10.17755/esosder.1432527
Jaber, A., Csonka, B., & Juhász, J. (2022). Long term time series prediction of bike sharing trips: A cast study of Budapest city. world, 1(2), 3. https://doi.org/10.1109/SCSP54748.2022.9792540
Jagannathan, J., & Divya, C. (2021). Time series analyzation and prediction of climate using enhanced multivariate prophet. International Journal of Engineering Trends and Technology, 69(10), 89-96. https://doi.org/10.14445/22315381/IJETT-V69I10P212
Komarica, J., Glavić, D., & Kaplanović, S. (2024). Predicting and analyzing electric bicycle adoption to enhance urban mobility in belgrade using ANN models. Applied Sciences, 14(19), 8965. https://doi.org/10.3390/app14198965
Kwarteng, S., & Andreevich, P. (2024). Comparative analysis of ARIMA, SARIMA and Prophet model in forecasting. Research & Development, 5(4), 110-120. https://doi.org/10.11648/j.rd.20240504.13
Patterson, K., Hassani, H., Heravi, S., & Zhigljavsky, A. (2011). Multivariate singular spectrum analysis for forecasting revisions to real-time data. Journal of Applied Statistics, 38(10), 2183-2211. https://doi.org/10.1080/02664763.2010.545371
Sanami, S., Mosalli, H., Yang, Y., Yeh, H. G., & Aghdam, A. G. (2025, July). Demand forecasting for electric vehicle charging stations using multivariate time-series analysis. In 2025 American Control Conference (ACC) (pp. 3461-3466). IEEE. https://doi.org/10.48550/arXiv.2502.16365
Subramanian, M., Cho, J., Veerappampalayam Easwaramoorthy, S., Murugesan, A., & Chinnasamy, R. (2023). Enhancing sustainable transportation: AI-driven bike demand forecasting in smart cities. Sustainability, 15(18), 13840. https://doi.org/10.3390/su151813840
Swari, M. H. P., Irawan, H. A., Muliawati, A., Aliansyah, Z., & Diyasa, I. G. S. M. (2025, December). Enhancing Time Series Forecasting Accuracy through Hybrid ARIMA–MLP Integration: A Case Study on E-Bicycle Sales. In 2025 International Conference on Informatics, Multimedia, Cyber and Information System (ICIMCIS) (pp. 967-972). IEEE. https://doi.org/10.1109/ICIMCIS68501.2025.11327149
Taoussi, B., Boudia, S. M., & Mazouni, F. S. (2025). Wind speed forecasting using univariate and multivariate time series models. Stochastic Environmental Research and Risk Assessment, 39(2), 547-579. https://doi.org/10.1007/s00477-024-02881-2
Vishnu, G., Kaliyaperumal, D., Pati, P. B., Karthick, A., Subbanna, N., & Ghosh, A. (2023). Short-term forecasting of electric vehicle load using time series, machine learning, and deep learning techniques. World Electric Vehicle Journal, 14(9), 266. https://doi.org/10.3390/wevj14090266
Wang, J., Peng, X., Wu, J., Ding, Y., Ali, B., Luo, Y., ... & Zhang, K. (2024). Singular spectrum analysis (SSA) based hybrid models for emergency ambulance demand (EAD) time series forecasting. IMA Journal of Management Mathematics, 35(1), 45-64. https://doi.org/10.1093/imaman/dpad019
Zhang, Y., Zhong, M., Geng, N., & Jiang, Y. (2017). Forecasting electric vehicles sales with univariate and multivariate time series models: The case of China. PloS one, 12(5), e0176729. https://doi.org/10.1371/journal.pone.0176729
Zunic, E., Korjenic, K., Hodzic, K., & Donko, D. (2020). Application of facebook's prophet algorithm for successful sales forecasting based on real-world data. arXiv preprint arXiv:2005.07575. https://doi.org/10.5121/ijcsit.2020.12203

Ayrıntılar

Birincil Dil

İngilizce

Konular

Yapay Zeka (Diğer)

Bölüm

Araştırma Makalesi

Yazarlar

Rana Bedir Urfalı ^*
0009-0003-9495-2482
Türkiye

Ersin Kaya
0000-0001-5668-5078
Türkiye

Yayımlanma Tarihi

3 Mart 2026

Gönderilme Tarihi

4 Kasım 2025

Kabul Tarihi

5 Şubat 2026

Yayımlandığı Sayı

Yıl 2026 Cilt: 29 Sayı: 1

DOI

https://doi.org/10.17780/ksujes.1817646

IZ

https://izlik.org/JA29UL67FG

Kaynak Göster

RIS / Bibtex

APA

Bedir Urfalı, R., & Kaya, E. (2026). PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 29(1), 295-314. https://doi.org/10.17780/ksujes.1817646

AMA

1.Bedir Urfalı R, Kaya E. PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi. 2026;29(1):295-314. doi:10.17780/ksujes.1817646

Chicago

Bedir Urfalı, Rana, ve Ersin Kaya. 2026. “PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS”. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi 29 (1): 295-314. https://doi.org/10.17780/ksujes.1817646.

EndNote

Bedir Urfalı R, Kaya E (01 Mart 2026) PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi 29 1 295–314.

IEEE

[1]R. Bedir Urfalı ve E. Kaya, “PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS”, Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, c. 29, sy 1, ss. 295–314, Mar. 2026, doi: 10.17780/ksujes.1817646.

ISNAD

Bedir Urfalı, Rana - Kaya, Ersin. “PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS”. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi 29/1 (01 Mart 2026): 295-314. https://doi.org/10.17780/ksujes.1817646.

JAMA

1.Bedir Urfalı R, Kaya E. PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi. 2026;29:295–314.

MLA

Bedir Urfalı, Rana, ve Ersin Kaya. “PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS”. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, c. 29, sy 1, Mart 2026, ss. 295-14, doi:10.17780/ksujes.1817646.

Vancouver

1.Rana Bedir Urfalı, Ersin Kaya. PREDICTIVE MODELING IN ELECTROMOBILITY: A TIME SERIES ANALYSIS. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi. 01 Mart 2026;29(1):295-314. doi:10.17780/ksujes.1817646