Work zones are a high priority issue in the field of road transportation because of their impacts on traffic safety. A better understanding of work zone crashes can help to identify the contributing factors and countermeasures to enhance roadway safety. This study investigates the prediction of work zone crash severity and the contributing factors by employing a parametric approach using the mixed logit modeling framework and a non-parametric machine learning approach using the support vector machine (SVM). The mixed logit model belongs to the class of random parameter models in which the effects of flexible variables across different observations are identified, that is, data heterogeneity is taken into account. The performance of the SVM model is enhanced by applying three metaheuristic algorithms: particle swarm optimization (PSO), harmony search (HS), and the whale optimization algorithm (WOA). Empirical findings indicate that SVM provides higher prediction accuracy and outperforms the mixed logit model. Estimation results reveal key factors that increase the likelihood of severe work zone crashes. Furthermore, the analysis illustrates the ability of the three metaheuristics to enhance the SVM and the superiority of the harmony search algorithm in improving the performance of the SVM model.

Previous studies have examined the relationships between run-off-road (ROR) crashes and the contributing factors; however, the impact of lighting conditions has been insufficiently addressed. As a result, the objective of this study was to research the effect of lighting conditions on the injury severity of ROR crashes that involve large trucks. Based on the crash data pertaining to large trucks in the state of Oregon from 2007 to 2013, two separate mixed logit models were developed to capture the contributing factors that affect injury severity in each lighting condition. The levels of injury severity sustained by truck drivers were categorized into three main categories: severe injury (fatal and incapacitating), minor injury (nonincapacitating and possible injury), and no injury. The mixed logit model was used to account for unobserved factors (i.e., unobserved heterogeneity). Estimation results indicated that there are significant differences between dark and lighted conditions, that the level of injury severity outcomes was highly influenced by several complex interactions between factors, and that the effects of some factors could vary across observations. The contributing factors include driver, traffic flow, roadway geometric features, land use, and time characteristics.