Harnessing the power of Large Language Models (LLMs), this study explores the use of three state-of-the-art LLMs, specifically GPT-3.5-turbo, LLaMA3-8B, and LLaMA3-70B, for crash severity analysis and inference, framing it as a classification task. We generate textual narratives from original traffic crash tabular data using a pre-built template infused with domain knowledge. Additionally, we incorporated Chain-of-Thought (CoT) reasoning to guide the LLMs in analyzing the crash causes and then inferring the severity. This study also examine the impact of prompt engineering specifically designed for crash severity inference. The LLMs were tasked with crash severity inference to: (1) evaluate the models’ capabilities in crash severity analysis, (2) assess the effectiveness of CoT and domain-informed prompt engineering, and (3) examine the reasoning abilities with the CoT framework. Our results showed that LLaMA3-70B consistently outperformed the other models, particularly in zero-shot settings. The CoT and Prompt Engineering techniques significantly enhanced performance, improving logical reasoning and addressing alignment issues. Notably, the CoT offers valuable insights into LLMs’ reasoning process, unleashing their capacity to consider diverse factors such as environmental conditions, driver behavior, and vehicle characteristics in severity analysis and inference.

Modern machine learning methods, such as tree ensembles, have recently become extremely popular due to their versatility and scalability in handling heterogeneous data and have been successfully applied across a wide range of domains. In this study, two widely applied tree ensemble methods, i.e., random forest (parallel ensemble) and gradient boosting (sequential ensemble), were investigated to predict resilient modulus, using routinely collected soil properties. Laboratory test data on sandy soils from nine borrow pits in Georgia were used for model training and testing. For comparison purposes, the two tree ensemble methods were evaluated against a regression tree model and a multiple linear regression model, demonstrating their superior performance. The results revealed that a single tree model generally suffers from high variance, while providing a similar performance to the traditional multiple linear regression model. By leveraging a collection of trees, both tree ensemble methods, Random Forest and eXtreme Gradient Boosting, significantly reduced variance and improved prediction accuracy, with the eXtreme Gradient Boosting being the best model, with an R² of 0.95 on the test dataset.

Generating meaningful inferences from crash data is vital to improving highway safety. Classic statistical methods are fundamental to crash data analysis and often regarded for their interpretability. However, given the complexity of crash mechanisms and associated heterogeneity, classic statistical methods, which lack versatility, might not be sufficient for granular crash analysis because of the high dimensional features involved in crash-related data. In contrast, machine learning approaches, which are more flexible in structure and capable of harnessing richer data sources available today, emerges as a suitable alternative. With the aid of new methods for model interpretation, the complex machine learning models, previously considered enigmatic, can be properly interpreted. In this study, two modern machine learning techniques, Linear Discriminate Analysis and eXtreme Gradient Boosting, were explored to classify three major types of multi-vehicle crashes (i.e., rear-end, same-direction sideswipe, and angle) occurred on Interstate 285 in Georgia. The study demonstrated the utility and versatility of modern machine learning methods in the context of crash analysis, particularly in understanding the potential features underlying different crash patterns on freeways.