Bayesian Network for Motorcycle Crash Severity Analysis

Accurate predictive modeling is often hindered by the prevalent issue of class imbalance within weather-related crash datasets. To address this critical challenge, this study introduces a novel and tailored synthetic data generation technique aimed at effectively handling nominal predictors specific to weather-related crash cases in North Carolina. Data treatment techniques such as the synthetic minority over-sampling technique-nominal (SMOTE-N) and adaptive synthetic-nominal (ADASYN-N) were investigated in this study. A comprehensive comparison of these data treatment techniques is conducted using two prominent machine learning models: the bagging algorithm (random forest [RF]) and the boosting algorithm (extreme gradient boosting [XGBoost]). The findings indicate that the effectiveness of data treatment varies with the severity level and the algorithm used. The ADASYN-N technique was observed to be highly effective for severe and moderate injury crash prediction using both the RF and XGBoost models, while the control dataset and SMOTE-N demonstrated notable performance in property damage only crash prediction using both the RF and XGBoost models. The findings from evaluating the performance of these models with data treatment methods serve as a benchmark for practitioners in selecting appropriate synthetic sample generation techniques, consequently facilitating the development of more accurate crash severity prediction models and contributing to enhanced traffic safety strategies.

show abstract

Association Patterns of Work Zone Crashes using Bayesian Network

Das,

Rahman,

Liu

et al. 2024

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

Ensuring the safety of work zones is a top priority for transportation agencies because of the dangers posed by vehicles changing lanes and paths within these areas. Recent statistics highlight the seriousness of this issue, showing a shocking 46% increase in fatal collisions within work zones in 2019 compared with 2011. Therefore, this study examined crashes related to intrusions or encroachments in work zones to uncover the underlying mechanisms. Analyzing four years of crash data (2016–2019) from the Texas Department of Transportation, this research utilized Bayesian network to identify crucial factors, their relationships, and potential alternative scenarios. The severity of injuries in work zone intrusion accidents was significantly influenced by male drivers, curved roads, and specific patterns of driver distraction and condition. The study revealed three distinct scenarios with complete probability of specific attributes: (1) crashes on rural non-principal arterial roads; (2) collisions with non-barrier fixed objects; and (3) non-injury crashes involving non-barrier fixed objects and driving violations. The detailed findings from this study can provide valuable insights for safety engineers, enabling them to reduce work zone crashes caused by encroachments. By comprehending the key factors and their effects, transportation agencies can implement effective measures to lessen the risks associated with work zone encroachments, ultimately creating a safer environment for both drivers and road workers.

show abstract

Bayesian Network for Motorcycle Crash Severity Analysis

Cited by 3 publications

References 29 publications

Unraveling Crash Causation: A Deep Dive into Non-Motorists on Personal Conveyance

Unraveling Crash Causation: A Deep Dive into Non-Motorists on Personal Conveyance

Effectiveness of Data Imbalance Treatment in Weather-Related Crash Severity Analysis

Association Patterns of Work Zone Crashes using Bayesian Network

Contact Info

Product

Resources

About