Journal of Soft Computing and Information Technology

Journal of Soft Computing and Information Technology

Predicting Accident Severity in Vehicular Ad Hoc Networks (VANETs) Using AOA-Optimized LSTM

Document Type : Persian Original Article

Authors
Elahe Doostsadigh 1
Kambiz Majidzadeh 2
Mohammad Masdari 1
Amin Bababzadeh 1
Farhad Soleimanian Gharehchopogh 1
1 Department of Computer Engineering, Ur.C., Islamic Azad University, Urmia, Iran.
2 Department of IT and Computer Engineering, Urmia Branch, Islamic Azad University, Urmia, Iran
10.22034/jscit.2026.564785.2200
Abstract
Vehicular Ad Hoc Networks (VANETs), as a core component of intelligent transportation systems, play a crucial role in traffic safety analysis and traffic management. Accident severity prediction is one of the key applications of VANETs; however, due to the highly dynamic environment, diverse contributing factors, and complex patterns of road accidents, it remains a challenging problem. In this paper, an optimized Long Short-Term Memory (LSTM)–based approach is proposed for accident severity prediction, which can be employed as a decision-support tool in VANET-based environments and intelligent transportation systems. In the proposed method, the hyper-parameters of the LSTM model are automatically tuned using the Archimedes Optimization Algorithm (AOA) in order to enhance prediction accuracy, stability, and generalization capability. To evaluate the effectiveness of the proposed approach, the performance of the AOA-Optimized LSTM (LSTM+AOA) model is compared with baseline LSTM and RNN models, as well as a genetic algorithm–optimized model (GA+LSTM), using a comprehensive set of standard evaluation metrics. The experiments are conducted on the STATS19 UK road accident dataset from 2017, which contains information related to accident characteristics and traffic conditions. Experimental results demonstrate that the AOA-Optimized LSTM (LSTM+AOA) model outperforms the compared methods in accident severity prediction while exhibiting reduced overfitting and improved generalization performance. The findings indicate that employing the AOA for hyper-parameter optimization can significantly improve the performance and robustness of LSTM-based accident severity prediction models, highlighting the potential of the proposed approach as an effective decision-support tool for intelligent traffic management systems.
Keywords
Predicting accident severity
VANET
deep learning
optimization
hyperparameter
LSTM
AOA
[1] N. Bohra et al., “Intelligence-Based Strategies with Vehicle-to-Everything Network: A Review,” Future Internet, vol. 17, no. 2, 2025. https://doi.org/10.3390/fi17020079
[2] P. Sicard et al., “Trends in Urban Air Pollution over the Last Two Decades: A Global Perspective,” Science of the Total Environment, vol. 858, Art. no. 160064, 2023.
https://doi.org/10.1016/j.scitotenv.2022.160064
[3] K. N. Qureshi and A. H. Abdullah, “A Survey on Intelligent Transportation Systems,” Middle-East Journal of Scientific Research, vol. 15, no. 5, pp. 629–642, 2013.doi:10.5829/idosi.mejsr.2013.15.5.11215
[4] J. A. Guerrero-Ibáñez, C. Flores-Cortés, and S. Zeadally, “Vehicular Ad-hoc Networks (VANETs): Architecture, Protocols and Applications,” in Next-Generation Wireless Technologies: 4G and Beyond. London, U.K.: Springer, 2013, pp. 49–70.
[5] I. T. Abdel-Halim and H. M. A. Fahmy, “Prediction-based Protocols for Vehicular Ad Hoc Networks: Survey and Taxonomy,” Computer Networks, vol. 130, pp. 34–50, 2018.https://doi.org/10.1016/j.comnet.2017.10.009
[6] A. Mahmoud and A. Mohammed, “A Survey on Deep Learning for Time-series Forecasting,” in Machine Learning and Big Data Analytics Paradigms: Analysis, Applications and Challenges, 2021, pp. 365–392. https://doi.org/10.1007/978-3-030-59338-4_19
[7] X. Ma et al., “Learning Traffic as Images: A Deep Convolutional Neural Network for Large-scale Transportation Network Speed Prediction,” Sensors, vol. 17, no. 4, Art. no. 818, 2017.https://doi.org/10.3390/s17040818
[8] S. Hochreiter and J. Schmidhuber, “Long Short-term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997. https://doi.org/10.1007/978-3-642-24797-2_4
[9] J. Bergstra and Y. Bengio, “Random Search for Hyper-parameter Optimization,” Journal of Machine Learning Research, vol. 13, no. 1, pp. 281–305, 2012.
[10] K. G. Dhal et al., “A Comprehensive Survey on Arithmetic Optimization Algorithm,” Archives of Computational Methods in Engineering, vol. 30, no. 5, pp. 3379–3404, 2023.https://doi.org/10.1007/s11831-023-09902-3
[11] X.-S. Yang, “Nature-inspired Optimization Algorithms: Challenges and Open Problems,” Journal of Computational Science, vol. 46, Art. no. 101104, 2020.https://doi.org/10.1016/j.jocs.2020.101104
[12] S. S. Sepasgozar and S. Pierre, “Network Traffic Prediction Model Considering Road Traffic Parameters Using Artificial Intelligence Methods in VANET,” IEEE Access, vol. 10, pp. 8227–8242, 2022.
[13] Great Britain Department for Transport. Road Safety Data (STATS19). 2023. Available from:www.data.gov.uk/dataset/road-safety-data.
[14] L. Abualigah et al., “The Arithmetic Optimization Algorithm,” Computer Methods in Applied Mechanics and Engineering, vol. 376, Art. no. 113609, 2021.
https://doi.org/10.1016/j.cma.2020.113609
[15] Y.-J. Zhang et al., “AOAAO: The hybrid algorithm of arithmetic optimization algorithm with aquila optimizer,” IEEE Access, vol. 10, pp. 10907–10933, 2022.
https://doi.org/10.1109/ACCESS.2022.3144431
[16] M. Abualigah, “Recent Advances in Metaheuristic Algorithms: Theory and Applications,” Artificial Intelligence Review, vol. 55, no. 7, pp. 5445–5506, 2022.https://doi.org/10.1007/978-3-030-72284-5_18
[17] M. Q. Ibrahim et al., “Optimizing convolutional neural networks: A comprehensive review of hyperparameter tuning through metaheuristic algorithms,” Archives of Computational Methods in Engineering, vol. 32, no. 8, pp. 5123–5160, 2025.https://doi.org/10.1007/s11831-025-10292-x
[18] G. E. P. Box et al., Time Series Analysis: Forecasting and Control. Hoboken, NJ, USA: John Wiley & Sons, 2015.https://doi.org/10.2307/3008255
[19] V. N. Katambire et al., “Forecasting the Traffic Flow by Using ARIMA and LSTM Models: Case of Muhima Junction,” Forecasting, vol. 5, no. 4, pp. 616–628, 2023.https://doi.org/10.3390/forecast5040034
[20] J. Smith and T. Brown, “Leveraging STATS19 for Traffic Accident Prediction: A Case Study,” Transportation Research Procedia, vol. 78, pp. 123–130, 2024.
[21] S. Khatri et al., “Machine Learning Models and Techniques for VANET Based Traffic Management: Implementation Issues and Challenges,” Peer-to-Peer Networking and Applications, vol. 14, pp. 1778–1805, 2021.https://doi.org/10.1007/s12083-020-00993-4
[22] J. L. Elman, “Finding Structure in Time,” Cognitive Science, vol. 14, no. 2, pp. 179–211, 1990.
https://doi.org/10.1207/s15516709cog1402_1
[23] J. Kim et al., “A Comprehensive Survey of Deep Learning for Time Series Forecasting: Architectural Diversity and Open Challenges,” Artificial Intelligence Review, vol. 58, no. 7, pp. 1–95, 2025.https://doi.org/10.1007/s10462-025-11223-9
[24] A. Belhadi et al., “A Recurrent Neural Network for Urban Long-term Traffic Flow Forecasting,” Applied Intelligence, vol. 50, pp. 3252–3265, 2020.https://doi.org/10.1007/s10489-020-01716-1
[25] J. Zhang et al., “Improved Genetic Algorithm Optimized LSTM Model and Its Application in Short-term Traffic Flow Prediction,” PeerJ Computer Science, vol. 8, Art. no. e1048, 2022. https://doi.org/10.7717/peerj-cs.1048
[26] E. Fu et al., “Temporal Self-Attention-Based Conv-LSTM Network for Multivariate Time Series Prediction,” Neurocomputing, vol. 501, pp. 162–173, 2022.https://doi.org/10.1016/j.neucom.2022.06.014
[27] E. H. Alkhammash, “An Optimized Gradient Boosting Model by Genetic Algorithm for Forecasting Crude Oil Production,” Energies, vol. 15, no. 17, Art. no. 6416, 2022.https://doi.org/10.3390/en15176416
[28] Y. Li et al., “EA-LSTM: Evolutionary Attention-Based LSTM for Time Series Prediction,” Knowledge-Based Systems, vol. 181, Art. no. 104785, 2019.https://doi.org/10.1016/j.knosys.2019.05.028
[29] S. Atef, S. Zahran, and A. Karam, “Bayesian Hyperparameter Optimization of GRU and LSTM Models for Short-Term Traffic Flow Prediction: A Case Study of Globe Roundabout in Saudi Arabia,” Applied System Innovation, vol. 9, no. 3, Art. no. 57, 2026.https://doi.org/10.3390/asi9030057
[30] K. K. Chandriah and R. V. Naraganahalli, “RNN/LSTM with Modified Adam Optimizer in Deep Learning Approach for Automobile Spare Parts Demand Forecasting,” Multimedia Tools and Applications, vol. 80, no. 17, pp. 26145–26159, 2021.https://doi.org/10.1007/s11042-021-10913-0
[31] M. Zahid et al., “Short Term Traffic State Prediction via Hyperparameter Optimization Based Classifiers,” Sensors, vol. 20, no. 3, Art. no. 685, 2020.https://doi.org/10.3390/s20030685
[32] M. Aliyari et al., “Ant Colony Optimization of Hyper-Parameters in Multi-Head Attention Layer for Time Series Forecasting,” Applied Soft Computing, Art. no. 113562, 2025.https://doi.org/10.1016/j.asoc.2025.113562
[33] A. R. Abdellah and A. Koucheryavy, “VANET Traffic Prediction Using LSTM with Deep Neural Network Learning,” in International Conference on Next Generation Wired/Wireless Networking. Cham, Switzerland: Springer, 2020. https://doi.org/10.1007/978-3-030-65726-0_25
[34] C. Pan et al., “Vehicle Trajectory Prediction Based on Spatial-Temporal Graph Transformer Networks with LSTM,” Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, Art. no. 09544070251398023, 2025.https://doi.org/10.1177/09544070251398023
Journal of Soft Computing and Information Technology
Volume 15, Issue 2
Summer - August
Summer 2026
Pages 31-47