사회경제지표를 반영한 거시적 교통사고 분석모형 개발

Abstract

The establishment and enforcement of local level safety policy, of which autonomous communities take charge, is playing the most essential role in the establishment and enforcement of national level traffic safety policy programs. However, in case of the analysis of traffic accidents made by the nation and local autonomous entities, it is too much concentrated on the analysis of regional aspects, such as black spots or dangerous zones, so there remains a difficulty with the establishment and assessment of autonomous - community - centered traffic safety planning with such a microscopic analysis. For the establishment and enforcement of local level safety measures, there needs to be the development of macroscopic accident analysis models that reflect the conditions and nature of the whole society rather than microscopic analysis through the existing primary factors. Thus, this research aims to identify the effects of regional characteristics on fatal crashes by collecting the indicators that represent the regional characteristics of 226 local governments across the nation, and selecting the indicators that seem to have association with traffic accidents. From the results of analysis, the truncated negative binomial model was selected as the most suitable one, and among the total of ten independent variables, sales volumes of oil, investment costs related to traffic accidents,foreign population, total length of roads, the proportion of highways, internal traffic volumes, and the ratio of women drivers were found to have correlation with the occurrence of fatal accidents. As a result, it is judged that there might be a difficulty in developing a model using the currently established statistical data; thus, for more developed research later, it is necessary to consider the research with panel analysis that makes it possible to make a simultaneous analysis of cross-section data and time - series data together with the collection of new indicators.