Analyzing Important Elements for Improving the Safety of Motorways
Source abstract
Abstract
This study aims to identify the factors that influence the occurrence of traffic accidents to improve motorway traffic safety. Various data, including the frequency of traffic accidents, traffic volume, geometric structure, and congestion level, were collected from individual sections of motorways in South Korea. Using the collected data, a traffic accident frequency prediction model was developed by applying an explainable artificial intelligence (AI)-based approach. The developed deep neural network model was combined with Shapley Additive Explanations to identify the variables that significantly affect the frequency of traffic accidents. The analysis identified five significant factors: segment length, total traffic volume, the proportion of truck traffic, the number of dangerous driving behaviors, and the duration of congestion. The results demonstrate the potential of using explainable AI in predicting traffic accident frequency. By identifying the factors that influence traffic accidents using this model, we can pinpoint areas for improvement, which may ultimately help reduce highway traffic accidents.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Transportation Systems and Logistics · Aerospace, Electronics, Mathematical Modeling · Engineering Diagnostics and Reliability
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of D.N.A.+ Convergence, Ajou University, Suwon 16499, Republic of Korea
Department of Transportation Engineering, Ajou University, Suwon 16499, Republic of Korea
Department of Transportation System Engineering, Ajou University, Suwon 16499, Republic of Korea
Funding
Korea Agency for Infrastructure Technology Advancement (KAIA) grant funded by the Ministry of Land, Infrastructure, and Transport
References (34)
Show source-provided references
1. World Health Organization (2024, May 12). Road Traffic Injuries. Available online: https://www.who.int/news-room/fact-sheets/detail/road-traffic-injuries.
2. Aga, M.A., Woldeamanuel, B.T., and Tadesse, M. (2021). Statistical modeling of numbers of human deaths per road traffic accident in the Oromia region, Ethiopia. PLoS ONE, 16.10.1371/journal.pone.0251492 ↗
3. World Health Organization (2015). Global Status Report on Road Safety 2015, World Health Organization.
4. Chen, S., Shao, H., and Ji, X. (2021). Insights into factors affecting traffic accident severity of novice and experienced drivers: A machine learning approach. Int. J. Environ. Res. Public Health, 18.10.3390/ijerph182312725 ↗
5. Advances, challenges, and future research needs in machine learning-based crash prediction models: A systematic review10.1016/j.aap.2023.107378 ↗
6. Al-Mistarehi, B.W., Alomari, A.H., Imam, R., and Mashaqba, M. (2022). Using machine learning models to forecast severity level of traffic crashes by R Studio and ArcGIS. Front. Built Environ., 8.10.3389/fbuil.2022.860805 ↗
7. Lee, J., Yoon, T., Kwon, S., and Lee, J. (2019). Model evaluation for forecasting traffic accident severity in rainy seasons using machine learning algorithms: Seoul city study. Appl. Sci., 10.10.3390/app10010129 ↗
8. Srinivasan, R., and Carter, D. (2011). Development of Safety Performance Functions for North Carolina, No. FHWA/NC/2010-09.
9. Homoscedasticity: An overlooked critical assumption for linear regression10.1136/gpsych-2019-100148 ↗
10. Poisson, Poisson-gamma and zero-inflated regression models of motor vehicle crashes: Balancing statistical fit and theory10.1016/j.aap.2004.02.004 ↗
11. Bansal, Explainable AI: To reveal the logic of black-box models10.1007/s00354-022-00201-2 ↗
12. Toward explainable artificial intelligence for early anticipation of traffic accidents10.1177/03611981221076121 ↗
13. The statistical analysis of crash-frequency data: A review and assessment of methodological alternatives10.1016/j.tra.2010.02.001 ↗
14. Exploring the application of the linear Poisson autoregressive model for analyzing the dynamic impact of traffic laws on fatal traffic accident frequency10.1155/2020/8854068 ↗
15. Crash prediction based on random effect negative binomial model considering data heterogeneity10.1016/j.physa.2019.123858 ↗
16. Negative binomial regression model for road crash severity prediction10.5539/mas.v12n4p38 ↗
17. Multivariate random-parameters zero-inflated negative binomial regression model: An application to estimate crash frequencies at intersections10.1016/j.aap.2014.04.018 ↗
18. The relationship between truck accidents and geometric design of road sections: Poisson versus negative binomial regressions10.1016/0001-4575(94)90038-8 ↗
19. Assi, K. (2020, January 20–21). Prediction of traffic crash severity using deep neural networks: A comparative study. Proceedings of the 2020 International Conference on Innovation and Intelligence for Informatics, Computing and Technologies (3ICT), Sakheer, Bahrain.10.1109/3ICT51146.2020.9311974 ↗
20. A deep neural network inverse solution to recover pre-crash impact data of car collisions10.1016/j.trc.2021.103009 ↗
21. Application of various machine learning architectures for crash prediction, considering different depths and processing layers10.1002/eng2.12215 ↗
22. Development of a global road safety performance function using deep neural networks10.1016/j.ijtst.2017.07.004 ↗
23. Barraza, J.F., Droguett, E.L., and Martins, M.R. (2021). Towards interpretable deep learning: A feature selection framework for prognostics and health management using deep neural networks. Sensors, 21.10.3390/s21175888 ↗
24. Koh, P.W., and Liang, P. (2017, January 6–11). Understanding black-box predictions via influence functions. Proceedings of the 34th International Conference on Machine Learning, Sydney, NSW, Australia.
25. Lundberg, S.M., and Lee, S.I. (2017, January 4–9). A unified approach to interpreting model predictions. Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, CA, USA.
26. Deep learning model for crash injury severity analysis using shapley additive explanation values10.1177/03611981221095087 ↗
27. Applying explainable machine learning techniques in daily crash occurrence and severity modeling for rural interstates10.1177/03611981221134629 ↗
28. Jin, Z., and Noh, B. (2023). From prediction to prevention: Leveraging deep learning in traffic accident prediction systems. Electronics, 12.10.3390/electronics12204335 ↗
29. Gabr, M.I., Helmy, Y.M., and Elzanfaly, D.S. (2023). Effect of missing data types and imputation methods on supervised classifiers: An evaluation study. Big Data Cogn. Comput., 7.10.3390/bdcc7010055 ↗
30. Ahsan, M.M., Mahmud, M.P., Saha, P.K., Gupta, K.D., and Siddique, Z. (2021). Effect of data scaling methods on machine learning algorithms and model performance. Technologies, 9.10.3390/technologies9030052 ↗
31. Liashchynskyi, P., and Liashchynskyi, P. (2019). Grid search, random search, genetic algorithm: A big comparison for NAS. arXiv.
32. Nohara, Y., Matsumoto, K., Soejima, H., and Nakashima, N. (2022). Explanation of machine learning models using shapley additive explanation and application for real data in hospital. Comput. Meth. Prog. Biomed., 214.10.1016/j.cmpb.2021.106584 ↗
33. How can SHAP values help to shape metabolic stability of chemical compounds?10.1186/s13321-021-00542-y ↗
34. Chadaga, K., Sampathila, N., Prabhu, S., and Chadaga, R. (2023). Multiple Explainable Approaches to Predict the Risk of Stroke Using Artificial Intelligence. Information, 14.10.1038/s41598-024-52428-2 ↗