Application of deep learning techniques in predicting motorcycle crash severity

Abstract

Machine learning (ML) techniques play a crucial role in today's modern world. Over the last years, road traffic safety is one of the applications where ML-methods have been successfully employed to prevent road users from being killed or seriously injured. A reliable data-driven predictive model is essential for this purpose. This could be achieved by successfully applying an intelligent transportation system to identify a driver at a higher risk of crashes. This study investigates the capabilities of different deep learning techniques to predict motorcycle crash severity. This study is based on 2,430 motorcycle crashes in a mountainous area in the United States over a 10-year period. Different deep networks (DNNs), including deep belief network, standard recurrent neural network (RNN), multilayer neural network, and single-layer neural network, were considered and compared in terms of prediction accuracy of motorcycle crash severity. Before conducting any analysis, feature reduction was performed to identify the optimal number of variables to include in the models by minimizing the error rate. Different metrics including the area under the curve and confusion matrix were used to compare the different models. Although the analyses were conducted on a relatively small dataset, the results indicate that almost all the DNN models better perform in predicting the severity of motorcycle crashes, compared with the single layer neural network. Finally, the RNN outperforms the other three neural network models. A comprehensive discussion has been made about the methodological approach implemented in this study.

Evaluation and improvement of computationally intelligent techniques for prediction of Road traffic crashes RTC severity on transport and logistics sector

“…Ebrahim Shaik et al [3] did the summarizes and an overview of many neural network models for RTC and includes an overview of next future works planification and perspective such as RBF (radial basis function), MPL (multilayer perceptron layer ), and SPL ( single perceptron layer). Rezapour et al [4] in 2020 implement many neural network models for RTC using SPL ( single perceptron layer) with MNN (multilayer neural network ) , and RNN ( recurrent neural network) for the prediction of the intensity, and frequency of RTC to evaluate their results. Sameen et al [5] in 2016, estimated with a recent study that traffic accidents will be the fifth cause of death worldwide in 2030.…”

“…Nearly 1.3 million humans die every year as a result of RTC, according to WHO on 20 June 2022. There are a number of researches that have been accomplished to classify and predict and sound the alarm RTC by employing ML techniques [4]. These techniques have been employed for prediction and classification in the RTC field [5].…”

Human injury-based safety decision of automated vehicles

“…The following references appear in the Supplemental information: Alkheder et al., 2017 , Assi, 2020 , Chen et al., 2015 , Chen et al., 2016 , He et al., 2018 , Ji and Levinson, 2020 , Liu et al., 2020 , Lubbe and Kiuchi, 2015 , Mansoor et al., 2020 , Rezapour et al., 2020 , Rezapour and Ksaibati, 2020 , Sameen and Pradhan, 2017 , Tang et al., 2019 , Wahab and Jiang, 2020 , Wang and Kim, 2019 , Zheng et al., 2019 .…”

Presentation of Machine Learning Approaches for Predicting the Severity of Accidents to Propose the Safety Solutions on Rural Roads

“…Rezapour et al developed and then compared a single-layer perceptron neural network (SLPNN) with a recurrent neural network to predict the frequency and severity of motorbike accidents. e SLPNN was employed to investigate accident research as a distance-based pattern matching technique to detect the correct road segment [20]. Najafi Moghaddam Gilani et al ( 2021) provided a study to identify the influential variables on vehicle accident occurrence.…”