Predicting crash injury severity is a crucial constituent of reducing the consequences of traffic crashes. This study developed machine learning (ML) models to predict crash injury severity using 15 crash-related parameters. Separate ML models for each cluster were obtained using fuzzy c-means, which enhanced the predicting capability. Finally, four ML models were developed: feed-forward neural networks (FNN), support vector machine (SVM), fuzzy C-means clustering based feed-forward neural network (FNN-FCM), and fuzzy c-means based support vector machine (SVM-FCM). Features that were easily identified with little investigation on crash sites were used as an input so that the trauma center can predict the crash severity level based on the initial information provided from the crash site and prepare accordingly for the treatment of the victims. The input parameters mainly include vehicle attributes and road condition attributes. This study used the crash database of Great Britain for the years 2011–2016. A random sample of crashes representing each year was used considering the same share of severe and non-severe crashes. The models were compared based on injury severity prediction accuracy, sensitivity, precision, and harmonic mean of sensitivity and precision (i.e., F1 score). The SVM-FCM model outperformed the other developed models in terms of accuracy and F1 score in predicting the injury severity level of severe and non-severe crashes. This study concluded that the FCM clustering algorithm enhanced the prediction power of FNN and SVM models.

Many unfortunate victims in road traffic crashes do not receive ideal treatment because their injury severity is not understood at an early stage. Swift crash severity prediction enables trauma and emergency centers to estimate the potential damage resulting from a road traffic crash and accordingly dispatch the proper emergency units to provide appropriate emergency treatment. A two-layer ensemble machine learning model is proposed in this study to predict road traffic crash severity. The first layer integrates four base machine learning models: k-nearest neighbor, decision tree, adaptive boosting, and support vector machine; the second layer classifies the crash severity based on the feedforward neural network model. The models are developed using road traffic crash data of road intersections over 6 years (2011–2016) obtained from Great Britain’s Department of Transport online database. Only the crash features that can be instantaneously and easily obtained are used as an input. To simplify the two-layer ensemble model, principal component analysis technique is used for dimensionality reduction in the second layer of the model. The performance of the two-layer ensemble model is compared with five base models: k-nearest neighbor, decision tree, adaptive boosting, support vector machine, and feedforward neural network. The prediction results reveal that the two-layer ensemble model outperforms the five base classification models based on two performance indicators: testing accuracy and F1 score. The transferability of the developed model is tested using the 3-year crash dataset for Canada obtained from the National Crash Database Online. The outcome indicates that the two-layer ensemble model shows the best performance for the Canadian dataset also. The proposed two-layer ensemble model would be beneficial in predicting crash severity with high accuracy based on limited initial crash information obtained from the crash location. Using this information, trauma centers would be able to prepare for appropriate and prompt medical treatment.

The accurate prediction of road traffic crash (RTC) severity contributes to generating crucial information, which can be used to adopt appropriate measures to reduce the aftermath of crashes. This study aims to develop a hybrid system using principal component analysis (PCA) with multilayer perceptron neural networks (MLP-NN) and support vector machines (SVM) in predicting RTC severity. PCA shows that the first nine components have an eigenvalue greater than one. The cumulative variance percentage explained by these principal components was found to be 67%. The prediction accuracies of the models developed using the original attributes were compared with those of the models developed using principal components. It was found that the testing accuracies of MLP-NN and SVM increased from 64.50% and 62.70% to 82.70% and 80.70%, respectively, after using principal components. The proposed models would be beneficial to trauma centers in predicting crash severity with high accuracy so that they would be able to prepare for appropriate and prompt medical treatment.

Many techniques including logistic regression and artificial intelligence have been employed to explain school-goers mode choice behavior. This paper aims to compare the effectiveness, robustness, and convergence of three different machine learning tools (MLT), namely the extreme learning machine (ELM), support vector machine (SVM), and multi-layer perceptron neural network (MLP-NN) to predict school-goers mode choice behavior in Al-Khobar and Dhahran cities of the Kingdom of Saudi Arabia (KSA). It uses the students’ information, including the school grade, the distance between home and school, travel time, family income and size, number of students in the family and education level of parents as input variables to the MLT. However, their outputs were binary, that is, either to choose the passenger car or walking to the school. The study examined a promising performance of the ELM and MLP-NN suggesting their significance as alternatives for school-goers mode choice modeling. The performances of the SVM was satisfactory but not to the same level of significance in comparison with the other two. Moreover, the SVM technique is computationally more expensive over the ELM and MLP-NN. Further, this research develops a majority voting ensemble method based on the outputs of the employed MLT to enhance the overall prediction performance. The presented results confirm the efficacy and superiority of the ensemble method over the others. The study results are likely to guide the transport engineers, planners, and decision-makers by providing them with a reliable way to model and predict the traffic demand for transport infrastructures on the basis of the prevailing mode choice behavior.