DOI RECORD
Development of Artificial Neural Network Models to Predict Factors Influencing the Severity of Injuries in Truck Crashes
Abstract
Truck crashes are crucial for a nation's economic health and safety due to the significant safety risks associated with the trucking industry. Previous studies have primarily examined highway traffic safety, with limited safety investigations focusing specifically on trucks. This paper presents crash severity models using Artificial Neural Networks in Jordan from 2014 to 2023. Three models were developed to analyze factors affecting truck crash severity, including fatal, severe, moderate, and minor injuries. The data set included sixteen independent variables. Results showed that vehicle number, crash type, speed, and roadway design significantly impact crash severity, among others.,Artificial Neural Network models are suitable as a prediction tool, as the results were closely corresponding to the observed values with an accuracy of up to 98%. This paper recommends devising countermeasures to reduce the incidence of truck crashes and enhance truck safety.