Applying Bayesian data mining to measure the effect of vehicular defects on crash severity

Abstract

The National Motor Vehicle Crash Causation Survey (NMVCCS), conducted from 2005 to 2007, showed that an estimated 44,000 crashes occurred due to vehicular defects— 2% of the NMVCCS crashes. Vehicle defects have an adverse effect upon overall roadway safety as they can increase the likelihood of traffic crashes, thus increasing the frequency of crash-related injuries and fatalities. Even though Louisiana requires a biennial vehicular safety inspection, recent traffic crash statistics have shown a higher than average percentage of vehicle defect–related crash fatalities in Louisiana (3% of all traffic fatalities). This fact called for an in-depth analysis of the vehicle defect–related crashes in Louisiana. The current study used 7 years (2010–2016) of traffic crash data from Louisiana to investigate the association between crash severity and vehicle-defect types by applying a Bayesian data mining approach. The findings showed that vehicle age is associated with severe injury crashes. Worn tires and defective brakes are the over-represented vehicle-defect categories. The significant association patterns can be used by different stakeholders to enhance roadway safety and reduce vehicular defect associated crashes.

Keywords:

• vehicle defect
• vehicle age
• vehicle type
• crash severity
• empirical Bayes data mining

Acknowledgements

The authors appreciate the help of Dr. Xiaoduan Sun for the data and suggestions. The authors would like to thank two anonymous reviewers for their valuable comments and suggestions.

Funding

This project has not been funded by any research grant.