01025699

Component

https://doi.org/10.3141/1950-01

http://scholar.google.com...sche&publication_year=2006

The general purpose of this research is to improve insight into road safety on Belgian highways by means of a layered model. The monthly number of persons killed on highways in Belgium is decomposed into three parts: exposure, accident risk, and fatality risk. The evolution in each of these dimensions is investigated separately. More specifically, for each dimension a descriptive and explanatory analysis reveals the optimal unobserved components model. The separate analysis of each dimension may reveal different underlying developments. The impact of meteorological, socioeconomic, legislative, and calendar factors on exposure, accident risk, and fatality risk is investigated. The analysis indicates that, although for each dimension the same basic components are available, the optimal model of each dimension has its unique structure of descriptive components and significant variables. Precipitation and snow enhance accident risk, while temperature plays a significant role for exposure. Fatality risk decreases in case of an extra day with precipitation and was significantly affected by the child restraint law. The economic indicators mainly affect accident risk. When the three models are brought back together, the number of highway deaths between 1993 and 2001 is well reconstructed.

01030706

English

pp 1-8

2006

Transportation Research Record: Journal of the Transportation Research Board

0309099595

Print

Figures (3) ; References (37) ; Tables (1)

Crash exposure; Crash risk forecasting; Fatalities; Highway safety; Legislation; Mathematical models; Risk analysis; Socioeconomic factors; Time of crashes; Traffic crashes; Weather conditions

Decomposition (Mathematics)

Belgium

Economics; Highways; Law; Safety and Human Factors; Society; I81: Accident Statistics

TRIS, TRB, ATRI

Mar 3 2006 10:34AM

• Application of "Highway Safety Manual" Draft Chapter: Louisiana Experience
• Bayesian Multivariate Poisson Regression for Models of Injury Count, by Severity
• Calibration of Safety Prediction Models for Planning Transportation Networks
• Comparison of Two Negative Binomial Regression Techniques in Developing Accident Prediction Models
• Effects of Sample Size on Goodness-of-Fit Statistic and Confidence Intervals of Crash Prediction Models Subjected to Low Sample Mean Values
• Refinement of Accident Prediction Models for Spanish National Network
• Use of Propensity Score Matching Method and Hybrid Bayesian Method to Estimate Crash Modification Factors of Signal Installation
• Using Macrolevel Collision Prediction Models in Road Safety Planning Applications
• Using the Rural Two-Lane Highway Draft Prototype Chapter