Accident severity prediction in big data using auto-machine learning

Document Type : Research Article

Authors

1 Graduate School of Natural and Applied Sciences, Suleyman Demirel University, Isparta, Turkey.

2 Department of Civil Engineering, Faculty of Engineering, Suleyman Demirel University, Isparta, Turkey.

3 Department of Property Protection and Security, Suleyman Demirel University, Isparta, Turkey.

Abstract

Estimating the severity of a traffic accident is a problem in motor vehicle traffic because it affects saving human life. If the severity value can be predicted before the accident, all the emergency teams needed could be sent to the area to provide faster first aid. With this aim, we studied a big data set for accidents in the United States of America (USA) between 2016 and 2020, almost 2.25×106 rows long. First, the data is preprocessed by removing unnecessary variables. Then rows with blank cells are removed. Finally, about 1.7×106 rows length data are left for the prediction process. A Machine Learning (ML) algorithm has been used to classify the severity based on 16 input parameters. Binary-to-decimal count conversation has been used as a novel preprocessing method. As a result, the model has been built with a total accuracy of 0.816. The test results are also validated with precision, recall, and f1-score values. An Auto-Machine Learning (Auto-ML) model has been developed and trained to predict the severity of a possible traffic accident based on the weather and road conditions. Thus, it will be possible to direct emergency units to areas with high accident severity, and preventing a fatality.

Keywords

References

References:
1.Colling, D.A., Industrial Safety: Management andTechnology, Prentice Hall (1990).
2.International Traffic Safety Data and Analysis Group (Irtad)Road Safety Annual Report 2020, International TransportForum (2020).
3.Kashani, A.T. and Mohaymany, A.S. “Analysis of the trafficinjury severity on two-lane, two-way rural roads based onclassification tree models”, Safety Sci. 49(10), pp. 1314–1320(2011). https://doi.org/10.1016/j.ssci.2011.04.019.
4.AlMamlook, R.E., Kwayu, K.M., Alkasisbeh, M.R., et al.“Comparison of machine learning algorithms for predictingtraffic accident severity”, In 2019 IEEE Jordan InternationalJoint Conference on Electrical Engineering and InformationTechnology (JEEIT), pp. 272-276 (2019). https://doi.org/10.1109/JEEIT.2019.8717393.
5.Hasheminejad, S.H.A. and Hasheminejad, S.M.H. “Trafficaccident severity prediction using a novel multi-objectivegenetic algorithm”, International Journal of Crashworthiness,22(4), pp. 425-440 (2017). https://doi.org/10.1080/13588265.2016.1275431.
6.Alkheder, S., Taamneh, M., and Taamneh, S. “Severityprediction of traffic accident using an artificial neuralnetwork”, Journal of Forecasting, 36(1), pp. 100-108 (2017).https://doi.org/10.1002/for.2425.
7.Rezaie Moghaddam, F., Afandizadeh, S. and Ziyadi, M.“Prediction of accident severity using artificial neural networks” International Journal of Civil Engineering, 9(1), pp. 41-48 (2011).
8.Sameen, M.I. and Pradhan, B. “Severity prediction of trafficaccidents with recurrent neural networks”, Applied Sciences,7(6), p. 476 (2017). https://doi.org/10.3390/app7060476.
9.Zong, F., Xu, H. and Zhang, H. “Prediction for traffic accident severity: comparing the Bayesian network and regressionmodels”, Mathematical Problems in Engineering, 1, 475194(2013). https://doi.org/10.1155/2013/475194.
10.Rasaizadi, A., Sherafat, E. and Seyedabrishami, S.E. “Short-term prediction of traffic state: Statistical approach versusmachine learning approach”, Scientia Iranica, 29(3), pp.1095-1106 (2022).https://doi.org/10.24200/sci.2021.57906.5469.
11.Wen, X., Xie, Y., Jiang, L., et al. “Applications of machinelearning methods in traffic crash severity modelling: currentstatus and future directions”, Transport Reviews, 41(6), pp.855-879 (2021).https://doi.org/10.1080/01441647.2021.1954108.
12.Labib, M.F., Rifat, A.S., Hossain, M.M., et al. “Road accidentanalysis and prediction of accident severity by using machinelearning in Bangladesh”, In 2019 7th International Conference on Smart Computing and Communications (ICSCC), pp. 1-5(2019). https://doi.org/10.1109/ICSCC.2019.8843640.
13.Iranitalab, A. and Khattak, A. “Comparison of four statisticaland machine learning methods for crash severity prediction”,Accident Analysis and Prevention, 108, pp. 27-36(2017).https://doi.org/10.1016/j.aap.2017.08.008.
14.Nassiri, H. and Mohamadian Amiri, A. “Prediction of roadway accident frequencies: Count regressions versus machinelearning models”, Scientia Iranica, 21(2), pp. 263-275 (2014).
15.Nassiri, H. and Edrissi, A. “Modeling truck accident severityon two-lane rural highways”, Scientia Iranica, 13(2), pp. 193-200 (2006).
16.Pei, Y., Wen, Y., and Pan, S. “Traffic accident severityprediction based on interpretable deep learning model”,Transportation Letters, 17(5), pp. 895-909 (2025).https://doi.org/10.1080/19427867.2024.2398336.
17.Yang, Z., Zhang, W., and Feng, J. “Predicting multiple typesof traffic accident severity with explanations: A multi-taskdeep learning framework” Safety Science, 146, 105522 (2022).https://doi.org/10.1016/j.ssci.2021.105522.
18.Angarita-Zapataa, J.S., Maestre-Gongorab, G., and Calderínc,J.F. “A case study of Auto-ML for supervised crash severityprediction” In 19th World Congress of the International FuzzySystems Association (IFSA), 12th Conference of the EuropeanSociety for Fuzzy Logic and Technology (EUSFLAT), and 11th International Summer School on Aggregation Operators(AGOP), pp. 187-194, Atlantis Press (2021).https://doi.org/10.2991/asum.k.210827.026.
19.Alnami, H.M., Mahgoub, I., and Al–Najada, H. “Highwayaccident severity prediction for optimal resource allocation ofemergency vehicles and personnel”, In 2021 IEEE 11th Annual Computing and Communication Workshop and Conference(CCWC), pp. 1231-1238 (2021).https://doi.org/10.1109/CCWC51732.2021.9376155.
20.Theofilatos, A., Chen, C., and Antoniou, C. “Comparingmachine learning and deep learning methods for real-timecrash prediction”, Transportation Research Record, 2673(8),pp. 169-178 (2019). https://doi.org/10.1177/0361198119841571.
21.Zhang, M., Kujala, P., and Hirdaris, S. “A machine learningmethod for the evaluation of ship grounding risk in realoperational conditions”, Reliability Engineering and SystemSafety, 226, 108697 (2022). https://doi.org/10.1016/j.ress.2022.108697.
22.Zhang, M., Montewka, J., Manderbacka, T., et al. “A big dataanalytics method for the evaluation of ship-ship collision riskreflecting hydrometeorological conditions”, ReliabilityEngineering and System Safety, 213, 107674 (2021).https://doi.org/10.1016/j.ress.2021.107674.
23.Kaffash, S., Nguyen, A.T., and Zhu, J. “Big data algorithmsand applications in intelligent transportation system: A reviewand bibliometric analysis”, International Journal ofProduction Economics, 231, 107868 (2021).https://doi.org/10.1016/j.ijpe.2020.107868.
24.Terzi, S. and Erten, K.M. “The effect of big data analysis forsustainable transportation”, Journal of InnovativeTransportation, 1(1), 1102 (2020).
25.Mitchell, T.M., Machine Learning WCB, McGraw-HillBoston, MA (1997).
26.Zhang, X.D. “Machine learning”, In A Matrix AlgebraApproach to Artificial Intelligence, pp. 223-440, Springer,Singapore (2020).
27.Feurer, M., Eggensperger, K., Falkner, S., et al. “Auto-sklearn2.0: The next generation”, arXiv Preprint arXiv:2007, 04074(2020).
28.Mahat, M.S.S. “Number System Conversion for Beginners(Decimal to Binary, Octal and Hexadecimal Conversion)”,Turkish Journal of Computer and Mathematics Education(TURCOMAT), 12(14), pp. 1445-1458 (2021).
29.Moosavi, S., Samavatian, M.H., Parthasarathy, S., et al.“Accident risk prediction based on heterogeneous sparse data:New dataset and insights”, In Proceedings of the 27th ACMSIGSPATIAL International Conference on Advances inGeographic Information Systems, pp. 33-42 (2019).https://doi.org/10.48550/arXiv.1909.09638.
Scientia Iranica
Volume 32, Issue 7
Transactions on Civil Engineering
March and April 2025 Article ID:6626

Files

History

  • Receive Date: 12 April 2022
  • Revise Date: 17 November 2022
  • Accept Date: 22 February 2023

Share

How to cite

Statistics