The unobserved heterogeneity in traffic crash data hides certain relationships between the contributory factors and injury severity. The literature has been limited in exploring different types of clustering methods for the analysis of the injury severity in crashes involving large trucks. Additionally, the variability of data type in traffic crash data has rarely been addressed. This study explored the application of the k-prototypes clustering method to countermeasure the unobserved heterogeneity in large truck-involved crashes that had occurred in the United States between the period of 2016 to 2019. The study segmented the entire dataset (EDS) into three homogeneous clusters. Four gradient boosted decision trees (GBDT) models were developed on the EDS and individual clusters to predict the injury severity in crashes involving large trucks. The list of input features included crash characteristics, truck characteristics, roadway attributes, time and location of the crash, and environmental factors. Each cluster-based GBDT model was compared with the EDS-based model. Two of the three cluster-based models showed significant improvement in their predicting performances. Additionally, feature analysis using the SHAP (Shapley additive explanations) method identified few new important features in each cluster and showed that some features have a different degree of effects on severe injuries in the individual clusters. The current study concluded that the k-prototypes clustering-based GBDT model is a promising approach to reveal hidden insights, which can be used to improve safety measures, roadway conditions and policies for the prevention of severe injuries in crashes involving large trucks.

Severe and fatal crashes involving large trucks result in significant social and economic losses for human society. Unfortunately, the notably low proportion of severe and fatal injury crashes involving large trucks creates an imbalance in crash data. Models trained on imbalanced crash data are likely to produce erroneous results. Therefore, there is a need to explore novel sampling approaches for imbalanced crash data, and it is crucial to determine the appropriate combination of a machine learning model, sampling approach, and ratio. This study introduces a novel cluster-based under-sampling technique, utilizing the k-prototypes clustering algorithm. After initial cluster-based under-sampling, the consolidated cluster-based under-sampled data set was further resampled using three different sampling approaches (i.e., adaptive synthetic sampling (ADASYN), NearMiss-2, and the synthetic minority oversampling technique + Tomek links (SMOTETomek)). Later, four machine learning models (logistic regression (LR), random forest (RF), gradient-boosted decision trees (GBDT), and the multi-layer perceptron (MLP) neural network) were trained and evaluated using the geometric mean (G-Mean) and area under the receiver operating characteristic curve (AUC) scores. The findings suggest that cluster-based under-sampling coupled with the investigated sampling approaches improve the performance of the machine learning models developed on crash data significantly. In addition, the GBDT model combined with ADASYN or SMOTETomek is likely to yield better predictions than any model combined with NearMiss-2. Regarding changes in sampling ratios, increasing the sampling ratio with ADASYN and SMOTETomek is likely to improve the performance of models up to a certain level, whereas with NearMiss-2, performance is likely to drop significantly beyond a specific point. These findings provide valuable insights for selecting optimal strategies for treating the class imbalance issue in crash data.

The significance of large trucks for the expansion and well-being of the economy is a well-established fact. However, crashes involving large trucks significantly threaten the overall safety on the roads. Moreover, a significant proportion of fatal crashes involving large trucks occurs on interstate roadways in the United States. However, not many studies have focused on the heterogeneous effects of the contributory factors on injury outcomes of interstate crashes involving large trucks. The current study explores the application of a k-prototypes clustering-based mixed logit model to identify and analyze the heterogeneous effects of contributory factors on injury outcomes in different scenarios of interstate crashes involving large trucks. Data from six years of crashes involving large trucks that occurred on interstate roadways in the state of Pennsylvania, US, were used in this study. The list of contributory factors included the following: drivers’ demographics and behaviors; crash characteristics; vehicle-related factors; location and roadway attributes; and environmental factors. The results indicated that some of the contributory factors were significant for all scenarios of interstate crashes involving large trucks. However, the magnitude of those factors’ effects varied across scenarios. Moreover, some of the contributory factors were exclusive to certain scenarios of interstate crashes involving large trucks. Lastly, the identification of random parameters in the cluster-based models indicated that a cluster-based mixed logit model is a more effective approach for accurately estimating the effects of contributory factors on injury outcomes in large-truck interstate crashes. The empirical findings of this study can be used to develop more robust traffic laws and safety measures to reduce the frequency and severity of injury in different scenarios of interstate crashes involving large trucks.

Large trucks and interstate highways play critical roles for freight transportation in the US (United States). However, a significant proportion of fatal crashes involving large trucks occur on interstate roads. This impacts the smoother operation of freight transportation and national economy. A couple of studies have investigated the factors that contribute to the severity of injuries in crashes involving large trucks. However, until the differences in the effects of factors that contribute to the severity of injuries in large-truck-involved and non-large-truck-involved interstate crashes are revealed, the countermeasures are less likely to be effective. The current study has developed two separate mixed logit (or random parameter logit) models on interstate crashes that occurred in the south region of the United States between 2016 and 2019 for a comparative analysis between interstate crashes involving large trucks and those without large trucks. In the models, the effects of crash characteristics, roadway attributes, traffic condition, environmental factors, driver and vehicle related factors on injury outcomes (i.e., no-injury, non-severe & serious injuries) were explored. The empirical results indicated that certain contributory factors are unique to large-truck-involved interstate crashes, and the effects of most of the contributory factors on non-severe and serious injuries are higher in interstate crashes involving large trucks than those without large trucks. The study has recommended some preventive measures to avoid interstate crashes involving large trucks based on the findings.