Optimization and Performance Evaluation of Hybrid Deep Learning Models for Traffic Flow Prediction

Goparaju, Sai Usha; Biju, Rahul; M, Pravalika; MC, Bhavana; Gangadharan, Deepak; Mandal, Bappaditya; C, Pradeep

Abstract

Traffic flow prediction has been regarded as a critical problem in intelligent transportation systems. An accurate prediction can help mitigate congestion and other societal problems while facilitating safer, cost and time-efficient travel. However, this requires the prediction algorithm to consider several complex characteristics of traffic flow data. These complex characteristics are an amalgamation of the spatial, temporal and periodic features exhibited by traffic flow data. To extract and leverage these features for traffic flow prediction, several hybrid deep learning models have been developed recently; however, there are still some challenges to determine the optimal architecture considering both spatial and temporal features. In this work, we perform an extensive comparison of hybrid deep learning models with and without periodicity to understand the prediction accuracy of these popular models. We propose an optimization framework that unifies genetic algorithm (GA) embedded optimization with prediction models in order to derive optimized deep learning architectures for hybrid traffic flow prediction exploring 2D spatial and temporal information. The framework enables the improvement of prediction performance and eliminates the hand-tuning process. An improved temporal convolutional network (TCN) architecture is derived using the GA driven optimization, which achieves superior traffic flow prediction accuracy compared to all other existing hybrid deep learning models on the freeway and urban traffic data from the PeMS traffic data set. We also evaluate the performance of the derived hybrid deep learning algorithms on the Raspberry PI embedded platform.

Citation

Goparaju, S. U., Biju, R., M, P., MC, B., Gangadharan, D., Mandal, B., & C, P. (2023, June). Optimization and Performance Evaluation of Hybrid Deep Learning Models for Traffic Flow Prediction. Presented at 2023 IEEE 97th Vehicular Technology Conference (VTC2023-Spring), Florence, Italy

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 2023 IEEE 97th Vehicular Technology Conference (VTC2023-Spring)
• Start Date: Jun 20, 2023
• End Date: Jun 23, 2023
• Acceptance Date: Aug 14, 2023
• Online Publication Date: Aug 14, 2023
• Publication Date: Aug 14, 2023
• Deposit Date: Aug 31, 2023
• Publicly Available Date: Jun 23, 2025
• Journal: IEEE Conference on Vehicular Technology (VTC)
• Print ISSN: 1550-2252
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• Book Title: IEEE Conference on Vehicular Technology (VTC)
• ISBN: 979-8-3503-1115-0
• DOI: https://doi.org/10.1109/vtc2023-spring57618.2023.10200600
• Keywords: Traffic flow prediction; Hybrid deep learning models; Temporal convolutional networks
• Public URL: https://keele-repository.worktribe.com/output/546389
• Publisher URL: https://ieeexplore.ieee.org/document/10200600
• Additional Information: Published in: 2023 IEEE 97th Vehicular Technology Conference (VTC2023-Spring); ; ISSN Information:; Electronic ISSN: 2577-2465; Print on Demand(PoD) ISSN: 1090-3038