Journal Search Engine
The Journal of The Korea Institute of Intelligent Transport Systems Vol.24 No.6 pp.1-16
고속도로 2차 교통사고 심각도 분석 : 강건·분위수 회귀 모형 적용
Analysis of Secondary Highway Accident Severity : Robust and Quantile Regression Approach
Abstract
This study applied the Equivalent Property Damage Only (EPDO) index to quantify the traffic accident severity and examined the determinants of severity in secondary accidents on expressways. The analysis was based on a ten-year dataset (2014–2023) from the Korea Expressway Corporation. Descriptive statistics and correlation analysis were first conducted, followed by robust regression (RR) and quantile regression (QR) to identify the key influencing factors. The results showed that the accident scale (number of vehicles involved) and accident type (particularly pedestrian accidents) had the most significant effects on severity, with these effects being more pronounced in high-severity accidents. In contrast, environmental conditions such as weather and road surface played a relatively minor role. These findings address the limitations of conventional ordinal categorical approaches by introducing an EPDO-based framework for quantifying the severity of accidents in terms of social cost. From a policy perspective, this study highlights multi-vehicle and pedestrian accidents as priority targets and underscores the need for differentiated safety measures to prevent secondary accidents on expressways.
