Journal of Transportation Research

Journal of Transportation Research

Presenting a hybrid deep learning approach (Conv-LSTM) with the aim of predicting short-term traffic volume on two-lane suburban roads (Case study: Karaj-Chalus road)

Document Type : Original Article

Authors
Mehrdad Khorrami 1
Ali Paydar 2
Soheil Rezashoar 3
1 M.Sc., Student, Department of Civil Engineering, Malard Branch, Islamic Azad University, Malard, Tehran, Iran.
2 Assistant Professor, Department of Civil Engineering, Malard Branch, Islamic Azad University, Malard, Tehran, Iran.
3 Ph.D., Candidate, Department of Transportation Planning, Faculty of Engineering, Imam Khomeini International University, Qazvin, Iran.
10.22034/tri.2025.496001.3308
Abstract
In recent years, the rapid increase in the number of vehicles has led to widespread traffic problems. Accurate prediction of traffic flow can help reduce congestion, optimize transportation infrastructure, enhance travel safety, and mitigate environmental impacts. This study aimed to predict the traffic volume for the next hour on the Karaj-Chalus axis, one of Iran’s busiest roads with complex weather and topographic conditions. To this end, traffic data from two traffic counters and synoptic weather data (with three-hour intervals) were collected. Given the capability of Convolutional Neural Networks (CNN) in extracting hidden features and spatial-temporal patterns, as well as the efficiency of Long Short-Term Memory (LSTM) networks in modeling time series, a hybrid Conv-LSTM model was proposed. This model, leveraging the strengths of both approaches, was designed for accurate traffic volume prediction. The proposed model was implemented in the Python programming language and compared with an Attention-based Long Short-Term Memory (A-LSTM) model. Evaluation results showed that the Conv-LSTM model achieved a coefficient of determination (R²) of 82.45% on the test dataset, outperforming the A-LSTM model with an R² of 81.93%. Additionally, the Mean Squared Error (MSE) and Root Mean Squared Error (RMSE), which indicate the discrepancy between predicted and actual values, were reported as 14.311 and 120 for Conv-LSTM, and 14.740 and 122 for A-LSTM, respectively. These values indicate lower dispersion and higher prediction accuracy for Conv-LSTM compared to A-LSTM. The application of the proposed model in predicting traffic volume on the Karaj-Chalus axis provides valuable insights for identifying congested routes, redistributing traffic, preventing accidents, dynamic route planning, and optimizing travel time management. Furthermore, this approach contributes to improved environmental sustainability and transportation efficiency by reducing fuel consumption and air pollution.
Keywords
Traffic Volume
Neural Network
Long Short Term Memory
Convolution
Conv-LSTM
Subjects
-Abdel-Rahim, A., Dixon, M. P., Bacon, C. J., & Gonzalez, A. (2015). Measures to Alleviate Congestion at Rural Intersections.
-Ahmed, M. S., & Cook, A. R. (1979). Analysis of freeway traffic time-series data by using Box-Jenkins techniques.
-Chai, C., Ren, C., Yin, C., Xu, H., Meng, Q., Teng, J., & Gao, G. (2022). A Multifeature Fusion Short‐Term Traffic Flow Prediction Model Based on Deep Learnings. Journal of Advanced Transportation, 2022(1), 1702766.
-Chen, X., Xie, X., & Teng, D. (2020). Short-term traffic flow prediction based on ConvLSTM model. 2020 IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC).
-Dong, W., & Sun, W. (2022). Traffic flow prediction based on Bi LSTM and attention. International Core Journal of Engineering, 8(3), 439-444.
-Gu, J., Wang, Z., Kuen, J., Ma, L., Shahroudy, A., Shuai, B., Liu, T., Wang, X., Wang, G., & Cai, J. (2018). Recent advances in convolutional neural networks. Pattern recognition, 77, 354-377.
-Harwood, D. W., Council, F., Hauer, E., Hughes, W., & Vogt, A. (2000). Prediction of the expected safety performance of rural two-lane highways.
-Hou, Y., Deng, Z., & Cui, H. (2021). Short‐Term Traffic Flow Prediction with Weather Conditions: Based on Deep Learning Algorithms and Data Fusion. Complexity, 2021(1), 6662959.
-Hu, X., Liu, T., Hao, X., & Lin, C. (2022). Attention-based Conv-LSTM and Bi-LSTM networks for large-scale traffic speed prediction. The Journal of Supercomputing, 78(10), 12686-12709.
-IRIMO. I.R. of Iran Meteorological Organization (IRIMO), Iran Road Meteorological Data Website. https://irimo.ir/ (2024). Accessed: 2024-03-20.
-Kashyap, A. A., Raviraj, S., Devarakonda, A., Nayak K, S. R., KV, S., & Bhat, S. J. (2022). Traffic flow prediction models–A review of deep learning techniques. Cogent Engineering, 9(1), 2010510.
-Khorshidi, N., Zargari, S. A., Rezashoar, S., & Mirzahossein, H. (2025). Optimizing Travel Time Reliability with XAI: A Virginia Interstate Network Case Using Machine Learning and Meta-Heuristics. Machine Learning with Applications, 100709.
-Kim, S., Hori, T., & Watanabe, S. (2017). Joint CTC-attention based end-to-end speech recognition using multi-task learning. 2017 IEEE international conference on acoustics, speech and signal processing (ICASSP).
-Li, T., Ni, A., Zhang, C., Xiao, G., & Gao, L. (2020). Short‐term traffic congestion prediction with Conv–BiLSTM considering spatio‐temporal features. IET Intelligent Transport Systems, 14(14), 1978-1986.
-Lu, J., Xiong, C., Parikh, D., & Socher, R. (2017). Knowing when to look: Adaptive attention via a visual sentinel for image captioning. Proceedings of the IEEE conference on computer vision and pattern recognition.
-Ma, C., Dai, G., & Zhou, J. (2021). Short-term traffic flow prediction for urban road sections based on time series analysis and LSTM_BILSTM method. IEEE Transactions on Intelligent Transportation Systems, 23(6), 5615-5624.
-Ramakrishnan, N., & Soni, T. (2018). Network traffic prediction using recurrent neural networks. 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA).
-Rasaizadi, A., Seyedabrishami, S., & Saniee Abadeh, M. (2021). Short‐Term Prediction of Traffic State for a Rural Road Applying Ensemble Learning Process. Journal of Advanced Transportation, 2021(1), 3334810.
-Rasaizadi, A., Sherafat, E., & Seyedabrishami, S. (2022). Short-term prediction of traffic state: Statistical approach versus machine learning approach. Scientia Iranica, 29(3), 1095-1106.
-Ren, C., Chai, C., Yin, C., Ji, H., Cheng, X., Gao, G., & Zhang, H. (2021). Short‐Term Traffic Flow Prediction: A Method of Combined Deep Learnings. Journal of Advanced Transportation, 2021(1), 9928073.
-Rezashoar, S., Kashi, E., & Saeidi, S. (2024). A hybrid algorithm based on machine learning (LightGBM-Optuna) for road accident severity classification (case study: United States from 2016 to 2020). Innovative Infrastructure Solutions, 9(8), 319.
-Rezashoar, S., & Rassafi, A. A. (2025). Analyzing the Performance of the Red Deer Optimization Algorithm in Comparison to Other Metaheuristic Algorithms. Journal of AI and Data Mining, 13(1), 53-61.
-RMTO. Road Maintenance and Transportation Organization (RMTO), Iran Road Traffic Data Website. https://141.ir/ (2024). Accessed: 2024-03-20.
 -Sherafat, E., Farooq, B., Karbasi, A. H., & Seyedabrishami, S. (2023). Attention-LSTM for Multivariate Traffic State Prediction on Rural Roads. arXiv preprint arXiv:2301.02731.
-Van Der Voort, M., Dougherty, M., & Watson, S. (1996). Combining Kohonen maps with ARIMA time series models to forecast traffic flow. Transportation Research Part C: Emerging Technologies, 4(5), 307-318.
-Wang, Y., Xiao, Z., & Cao, G. (2022).
A convolutional neural network method based on Adam optimizer with power-exponential learning rate for bearing fault diagnosis. Journal of Vibroengineering, 24(4), 666-678.
-Wu, P., Huang, Z., Pian, Y., Xu, L., Li, J., & Chen, K. (2020). A combined deep learning method with attention‐based LSTM model for short‐term traffic speed forecasting. Journal of Advanced Transportation, 2020(1), 8863724.
-Yu, Z., Yu, J., Fan, J., & Tao, D. (2017). Multi-modal factorized bilinear pooling with co-attention learning for visual question answering. Proceedings of the IEEE International Conference on Computer Vision.
-Zhang, K., Wu, L., Zhu, Z., & Deng, J. (2020). A multitask learning model for traffic flow and speed forecasting. IEEE Access, 8, 80707-80715.
-Zhang, Y., Cheng, T., Ren, Y., & Xie, K. (2020). A novel residual graph convolution deep learning model for short-term network-based traffic forecasting. International Journal of Geographical Information Science, 34(5), 969-995.
-Zhao, Z., Chen, W., Wu, X., Chen, P. C., & Liu, J. (2017). LSTM network: a deep learning approach for short‐term traffic forecast. IET Intelligent Transport Systems, 11(2), 68-75.
-Zheng, H., Lin, F., Feng, X., & Chen, Y. (2020). A hybrid deep learning model with attention-based conv-LSTM networks for short-term traffic flow prediction. IEEE Transactions on Intelligent Transportation Systems, 22(11), 6910-6920.
 
Journal of Transportation Research
Volume 23, Issue 3
Summer 2026
Pages 439-458