Application of Association Rules in Freeway Accident Data Analysis

摘要

The statistical models, such as Poisson or negative binomial regression models, have been employed to analyze vehicle accident frequency for many years. However, these models have their own model assumptions and pre-defined underlying relationship between dependent and independent variables. If these assumptions are violated, the model could lead to erroneous estimation of accident likelihood. Association rules, one of the most widely applied data mining techniques, have been commonly employed in business administration, industry, and engineering. Association rules do not require any pre-defined underlying relationship between target (dependent) variable and predictors (independent variables) and has been shown to be a powerful tool, particularly for discovering unknown relationships and patterns among the data. This study collected the 2001-2002 accident data of National Freeway 1 in Taiwan. Association rule techniques were applied to identify the empirical relationship between traffic accidents and highway geometric variables, traffic characteristics and environmental factors. The analysis results of association rules indicated that the horizontal curve, non-fog zone, number of lanes, peak hour factor, average daily tractor-trailer volume and precipitation variables associate with freeway accidents.