Turner-Fairbank Highway Research Center

 
 
 
 
« back to the list

Predicting motor vehicle collisions using Bayesian neural network models: An empirical analysis

Product Type

Journal Article

Author

Xie, Y., D. Lord, and Y. Zhang

Date

Sep-07

Full citation

Xie, Y., D. Lord, and Y. Zhang. Predicting motor vehicle collisions using Bayesian neural network models: An empirical analysis. In Accident Analysis and Prevention, Vol. 39, No. 5, Pergamon Press, Oxford, NY, September 2007, pp. 922-933.

Abstract

Statistical models have frequently been used in highway safety studies. They can be utilized for various purposes, including establishing relationships between variables, screening covariates and predicting values. Generalized linear models (GLM) and hierarchical Bayes models (HBM) have been the most common types of model favored by transportation safety analysts. Over the last few years, researchers have proposed the back-propagation neural network (BPNN) model for modeling the phenomenon under study. Compared to GLMs and HBMs, BPNNs have received much less attention in highway safety modeling. The reasons are attributed to the complexity for estimating this kind of model as well as the problem related to over-fitting the data. To circumvent the latter problem, some statisticians have proposed the use of Bayesian neural network (BNN) models. These models have been shown to perform better than BPNN models while at the same time reducing the difficulty associated with over-fitting the data. The objective of this study is to evaluate the application of BNN models for predicting motor vehicle crashes. To accomplish this objective, a series of models was estimated using data collected on rural frontage roads in Texas. Three types of models were compared: BPNN, BNN and the negative binomial (NB) regression models. The results of this study show that in general both types of neural network models perform better than the NB regression model in terms of data prediction. Although the BPNN model can occasionally provide better or approximately equivalent prediction performance compared to the BNN model, in most cases its prediction performance is worse than the BNN model. In addition, the data fitting performance of the BPNN model is consistently worse than the BNN model, which suggests that the BNN model has better generalization abilities than the BPNNmodel and can effectively alleviate the over-fitting problem without significantly compromising the nonlinear approximation ability. The results also show that BNNs could be used for other useful analyses in highway safety, including the development of accident modification factors and for improving the prediction capabilities for evaluating different highway design alternatives.

Available From

Accident Analysis and Prevention

Link To Journal Article

Predicting motor vehicle collisions using Bayesian neural network models: An empirical analysis

Keywords

artificial neural networks
Bayesian methods
negative binomial
statistical model
frontage roads


« Back to the List

HSIS Summaries

HSIS Summary Reports are two to eight pages in length and include a brief description of the issue addressed, data used, methodology applied, significant results, and practical implications.

Read More

Research Reports

A variety of research studies have been performed using data from HSIS. Many of the final reports prepared are now available electronically.

Read More

Technical Summaries

Research reports are often summarized in executive summaries, technical briefs, or other abbreviated formats. Included here are those road safety summaries that involved research using HSIS data.

Read More

Safety Analysis Tools

In addition to conducting research, HSIS resources are also used to develop products that can be used by practitioners in the analysis of safety problems.

Read More

Other Projects

HSIS data are sometimes used in research studies that result in other types of finished products, such as dissertations, theses, and conference proceedings.

Read More