All Issue

2026 Vol.22, Issue 1 Preview Page

Original Article

A Study on Deep Learning-Based Time-Series Forecasting of Highway Traffic Volume Using VDS Data

VDS 데이터 기반 딥러닝을 활용한 고속도로 교통량 시계열 예측에 관한 연구

윤병조1*ㆍ이선민2

Byoung-Jo Yoon1*, Sun-min Lee2

1Professor, College of Urban Science, Incheon National University, Incheon, Republic of Korea
2Researcher, College of Urban Science, Incheon National University, Incheon, Republic of Korea
31 March 2026. pp. 375-388
PDF Citation
Abstract
Purpose: This study aims to develop a deep learning-based time-series forecasting model for expressway traffic volume using data collected from Vehicle Detection System (VDS) installed in nationwide expressways, and to compared and analyze the predictive performance of representative deep learning models. Method: Nationwide expressway VDS data collected throughout 2024 were used to evaluate three deep learning models: LSTM, GRU, Transformer. A one-step-ahead time-series forecasting framework was adopted, in which traffic information from the previous 24 hours was used as input to predict volume for the subsequent hour. All models were trained and evaluated under identical experimental conditions. Result: The experimental results indicated that the GRU model exhibited relatively stable predictive performance compared to the other deep learning models in terms of overall error metrics and explanatory power. Despite its relatively simple architecture, the GRU model effectively captured the temporal variability of expressway traffic volume and achieved a level of prediction accuracy that can be considered suitable for practical applications when evaluated in actual traffic volume units. Conclusion: This study is meaningful in that it empirically examined the applicability of deep learning-based traffic volume forecasting using national expressway VDS time-series data. In particular, the efficiency and predictive stability of the GRU model were confirmed, suggesting its potential usefulness for the development of forecasting models to support real-time traffic management and operation in the future.
Keywords
Expressway Traffic Volume Forecasting
Vehicle Detection System (VDS)
Deep Learning
GRU
Time-Series Analysis
Intelligent Transportation System
연구목적: 본 연구는 전국의 고속도로에 설치된 차량 검지기(VDS) 데이터를 활용하여 딥러닝 기반 고속도로 교통량 시계열 예측 모델을 구축하고, 대표적인 딥러닝 모델 간 예측 성능을 비교ㆍ분석하는 것을 목적으로 한다. 연구방법: 2024년 한 해 동안 수집된 전국 고속도로 VDS 데이터를 대상으로 LSTM, GRU, Transformer 모델을 적용하였으며, 과거 24시간의 교통정보를 입력한 후 다음 1시간 교통량을 예측하는 1-step ahead 시계열 예측 구조를 동일한 학습 조건에서 비교ㆍ평가하였다. 연구결과: 실험 결과 GRU 모델이 전반적인 오차 지표와 설명력 측면에서 다른 딥러닝 모델과 비교하여 상대적으로 안정적인 예측 성능을 보였다. 특히, 비교적 간결한 구조임에도 불구하고 고속도로 교통량 시계열의 변동 특성을 효과적으로 반영하였고 실제 교통량 단위 기준에서도 실무적인 활용을 고려할 수 있는 수준의 예측 정확도를 확보한 것으로 분석되었다. 결론: 본 연구는 전국의 고속도로 VDS 시계열 데이터를 활용한 딥러닝 기반 교통량 예측을 실증적으로 검토하였다는 점에서 의의가 있다. 특히 GRU 모델의 효율성과 예측 안정성을 확인할 수 있었고 향후 실시간 교통 관리 및 운영을 위한 예측 모형 개발에 활용될 수 있을 것으로 기대된다.
키워드
고속도로 교통량 예측
차량 검지기(VDS)
딥러닝
GRU
시계열 분석
지능형 교통체계
References
  1. Abduljabbar, R.L., Dia, H., Tsai, P.-W., Liyanage, S. (2021). “Short-term traffic forecasting: An LSTM network for spatial-temporal speed prediction.” Future Transportation, Vol. 1, No. 1, pp. 21-37. 10.3390/futuretransp1010003
  2. Fan, X., Xiang, C., Gong, L., He, X., Qu, Y., Amirgholipour, S., Xi, Y., Nanda, P., He, X. (2020). “Deep learning for intelligent traffic sensing and predicting: Recent advances and future challenges.” CCF Transactions on Pervasive Computing and Intersection, Vol. 2, No. 4, pp. 240-260. 10.1007/s42486-020-00039-x
  3. Heidl, W., Eitzinger, C., Gyimesi, M., Breitnecker, F. (2011). “Learning over sets with recurrent neural networks: An empirical categorization of aggregation functions.” Mathematics and Computers in Simulation, Vol. 82, No. 3, pp. 442-449. 10.1016/j.matcom.2010.10.018
  4. Horcher, D., Tirachin, A. (2021). “A review of public transport economics.” Economics of Transportation, Vol. 25, 100196. 10.1016/j.ecotra.2021.100196
  5. Karim, F., Majumdar, S., Darabi, H., Chen, S. (2018). “LSTM fully convolutional networks for time series classification.” IEEE Access, Vol. 6, pp. 1662-1669. 10.1109/ACCESS.2017.2779939
  6. Kim, H.-Y., Lee, S.-S., Son, S.-H., Jeong, J-S. (2023). “Development of a speed prediction model for urban network based on gated recurrent unit.” Journal of the Korean Institute of Intelligent Transport Systems, Vol. 22, No. 1, pp. 103-114. 10.12815/kits.2023.22.1.103
  7. Korea Transportation Institute (2023). National Transportation Policy Evaluation Index Survey; Volume 3 - Estimation for Road Traffic Congestion Costs in 2021. Sejong.
  8. Li, S., Jin, X., Xuan, Y., Zhou, X., Chen, W., Wang, Y.-X., Yan, X. (2019). “Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting.” Advances in Neural Information Processing Systems (NeurIPS), Vol. 32, pp. 5243-5253.
  9. Lim, J., Lee, J., An, C., Park, E. (2024). “Enhancing real-time traffic volume prediction: A two-step approach of object detection and time series modeling.” IET Intelligent Transport System, Vol. 18, No. 12, pp. 2744-2758. 10.1049/itr2.12576
  10. Ma, X., Tao, Z., Wang, Y., Yu, H., Wang, Y., (2015). “Lomg short-term memory neural for traffic speed prediction using remote microwave sensor data.” Transportation Research Parc C: Emerging Technologies, Vol. 54, pp. 187-197. 10.1016/j.trc.2015.03.014
  11. Mahmoud, N., Abdel-Aty, M., Cai, Q., Yuan, J. (2021). “Predicting cycle-level traffic movements at signalized intersections using machine learning models.” Transportation Research Part C: Emerging Technologies, Vol. 124, 102930. 10.1016/j.trc.2020.102930
  12. Park, D., Kim, J., Park, J. (2024). “A methodology for real-time risk assessment of freeway freight routes using Transformer-LSTM time-series prediction with traffic safety big data.” Journal of the Korea Institute of Intelligent Transport Systems, Vol. 22, No. 1, pp. 44-58.
  13. Tang, P.T.P. (1992). “Table-driven implementation of the expm1 function in IEEE floating-point arithmetic.” ACM Transaction on Mathematical Software, Vol. 18, No. 2, pp. 211-222. 10.1145/146847.146928
  14. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., Polosukhin, I. (2017). “Attention is all you need.” Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA, pp. 5998-6008.
  15. Wu, H., Xu, J., Wang, J., Long, M. (2021). Autoformer: Decomposition Transformers with Auto-correlation for Long-term Series Forecasting. In Advances in Neural Information Processing Systems 34 (NeurIPS 2021).
  16. Yunita, A., Pratama, I.P., Almuzakki, M.Z., Ramadhan, H., Rizaldy, A.E., Pambudi, A.A., Budi, F.M.A., Hidayat, B. (2025). “Performance analysis of neural network architectures for time series forecasting: A comparative study of RNN, LSTM, GRU and hybrid models.” MethodsX, Vol. 15, 103462. 10.1016/j.mex.2025.103462 40777584 PMC12329085
  17. Zhou, H., Zhang, S., Peng, J., Zhang, S., Li, J., Xiong, H., Zhang, W. (2021). “Informer: Beyond efficient transformer for long sequence time-series forecasting.” Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 35, No. 12, pp. 11106-11115 10.1609/aaai.v35i12.17325
Information
  • Publisher :The Korean Society of Disaster Information
  • Publisher(Ko) :한국재난정보학회
  • Journal Title :Journal of the Society of Disaster Information
  • Journal Title(Ko) :한국재난정보학회논문집
  • Volume : 22
  • No :1
  • Pages :375-388
  • DOI :https://doi.org/10.15683/kosdi.2026.3.31.375
Journal Informaiton Journal of the Society of Disaster Information Journal of the Society of Disaster Information
  • NRF
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close