논문 상세보기
pp.61-70
PURPOSES : This study aimed to derive the factors that contribute to crash severity in mixed traffic situations and suggest policy implications for enhancing traffic safety related to these contributing factors. METHODS : California autonomous vehicle (AV) accident reports and Google Maps based on accident location were used to identify potential accident severity-contributing factors. A decision tree analysis was adopted to derive the crash severity analyses. The 24 candidate variables that affected crash severity were used as the decision tree input variables, with the output being the crash severity categorized as high, medium, and low. RESULTS : The crash severity contributing factor results showed that the number of lanes, speed limit, bus stop, AV traveling straight, AV turning left, rightmost dedicated lane, and nighttime conditions are variables that affect crash severity. In particular, the speed limit was found to be a factor that caused serious crashes, suggesting that the AV driving speed is closely related to crash severity. Therefore, a speed management strategy for mixed traffic situations is proposed to decrease crash severity and enhance traffic safety. CONCLUSIONS : This paper presents policy implications for reducing accidents caused by autonomous and manual vehicle interactions in terms of engineering, education, enforcement, and governance. The findings of this study are expected to serve as a basis for preparing preventive measures against AV-related accidents.
1. 서론
2. 데이터셋 구축
2.1. 데이터 수집
2.2. 구글맵 활용한 도로교통특성자료 수집
2.3. 데이터 전처리
3. 방법론 및 분석결과
3.1. 사고 심각도 영향요인 도출 방법
3.2. 의사결정나무 분석결과
4. 시사점
5. 결론
Acknowledgement
REFERENCES
