Abstract

Traffic congestion has emerged as one of the major challenges faced by expanding urban environments, resulting in increased travel delays, excessive fuel consumption, and environmental pollution. Accurate traffic forecasting plays an essential role in urban planning and the development of efficient transportation systems. This study applies an experimental approach to evaluate the performance of various deep learning models under different traffic scenarios. The experiments were conducted using the Google Colab platform to assess multiple deep learning architectures, including Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and Graph Neural Networks (GNN). Real-world traffic data consisting of traffic flow, density, speed, and temporal variables were utilized for model training and evaluation. Model performance was measured using standard evaluation metrics including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and coefficient of determination (R²). Experimental findings indicate that LSTM and GNN models demonstrate superior predictive performance compared with conventional RNN and CNN approaches. The study contributes to the domains of deep learning, time-series forecasting, cloud-based experimentation, and Intelligent Transportation Systems (ITS).

References

• Chen, Y., Guo, J., Xu, H., Huang, J., & Su, L. (2023). Improved long short-term memory-based periodic traffic volume prediction method.IEEE Access, 11, 103502–103510. doi:10.1109/ACCESS.2023.3305398
• Naheliya, B., Redhu, P., & Kumar, K. (2023). A hybrid deep learning method for short-term traffic flow forecasting: GSA-LSTM. Indian Journal of Science and Technology, 16(46), 4358–4368. doi:10.17485/IJST/v16i46.2520
• Bharti, Redhu, P., & Kumar, K. (2023). Short-term traffic flow prediction based on optimized deep learning neural network: PSO-Bi-LSTM. Physica A: Statistical Mechanics and its Applications, 625, Article 129001. doi:10.1016/j.physa.2023.129001
• Jiang, W., Luo, J., He, M., & Gu, W. (2023). Graph neural network for traffic forecasting: The research progress. ISPRS International Journal of Geo-Information, 12(3), Article 100. Doi: 10.3390/ijgi12030100
• Afandizadeh, S., Abdolahi, S., & Mirzahossein, H. (2024). Deep learning algorithms for traffic forecasting: A comprehensive review and comparison with classical ones. Journal of Advanced Transportation, 2024, Article 9981657. doi:10.1155/2024/9981657
• Wang, Y., Ke, S., An, C., Lu, Z., & Xia, J. (2024). A hybrid framework combining LSTM NN and BNN for short-term traffic flow prediction and uncertainty quantification. KSCE Journal of Civil Engineering, 28(1), 363–374. Doi: 10.1007/s12205-023-2457-y
• Song, Y., Luo, R., Zhou, T., Zhou, C., & Su, R. (2024). Graph attention informer for long-term traffic flow prediction under the impact of sports events. Sensors, 24(15), Article 4796. Doi: 10.3390/s24154796
• Fu, F., Zheng, X., & Wu, Y. (2024). Urban traffic flow prediction based on Bayesian deep learning. Sustainability, 16(5), Article 1818. Doi: 10.3390/su16051818
• Guo, C., Zhang, Y., & Li, X. (2024). MVHS-LSTM: The comprehensive traffic flow prediction model based on multiple variables heuristic selection long short-term memory. Applied Sciences, 14(7), Article 2959. Doi: 10.3390/app14072959.
• Dai, G., Tang, J., & Luo, W. (2023). Short-term traffic flow prediction: An ensemble machine learning approach. Alexandria Engineering Journal, 74, 467–480. doi:10.1016/j.aej.2023.05.015