Abstract
Train safety has been a top priority in the railroad industry. Understanding accident risks is of paramount importance for prioritizing effective prevention strategies. Previous work has focused on estimating the severity of derailments and various statistical models based on structured data were used. However, unstructured data records which provide considerable information about train derailments have received minimal consideration due to a lack of procedures of processing and interpreting them. To narrow this knowledge gap, this study aims to quantitatively estimate derailment severity by considering unstructured data utilizing topic modeling. A statistical model that integrates both structured and unstructured data was established to analyze U.S. freight train derailments from 1996 to 2019. The comparative results of predictions revealed that the model with combined text information outperformed the one without the unstructured data.Quantile regression was also developed to assess various statistical distributions of derailment severity. Both models with unstructured data provide a deeper understanding of derailment severity and ultimately improve railroad safety performance.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.