Identification of Combination Effects of Severe Crashes Using Data Mining Techniques

Kyeongjoo Kwon; Juneyoung Park; Hyunsuk Lee

doi:10.7470/jkst.2021.39.3.369

All Issue

2021 Vol.39, Issue 3 Preview Page Next Page

Article

Identification of Combination Effects of Severe Crashes Using Data Mining Techniques 데이터 마이닝 기법을 활용한 복합 사고영향요인 선정에 관한 연구

30 June 2021. pp. 369-382

PDF XML

Abstract

Recently, the number of fatalities that occurred on domestic expressway has continuously decreased, but there are still much more fatalities compared to major OECD countries. In order to achieve the OECD level of fatalities, it is necessary to accurately identify the cause of the crash and establish systematic traffic safety measures reflecting the characteristics of the crash. In Korea, safety plan for expressway has been established based on macroscopic perspective with simple basic crash statistics. However, expressway is environment that is highly probable for severe crashes due to high traffic speed. Therefore, in order to reduce crash severity, safety policies need to be applied efficiently based on more detailed crash analysis. In the existing expressway safety policy introduction system, there is a problem in that only fragmentary and temporary improvement measures have been applied based on a single cause of the crash analysis. This method is effective from a macroscopic point of view, but it is difficult to reduce severe crashes such as serious injuries and massive crashes. Because crashes are caused by a combination of multiple factors, and the mechanism of occurrence is also complex. Thus, the main objective of this study is to evaluate traffic safety combination factors to supplement the problems of the existing system. Various big data mining techniques were applied to identify interaction effects between variables. Traditional statistical methods were also adopted to reflect the latest research trends. Furthermore, this paper suggested directions for establishing expressway traffic safety policy from a more comprehensive perspective.

Keywords

combination factors

countermeasures

data mining

highway

severe crash

국가 차원에서는 거시적 관점의 정책 수립을 위해 단순 기초통계 수준 내 교통사고 동향을 파악하고 분석을 통해 전반적인 정책을 수립하고 있다. 하지만 고속도로는 통행속도가 높아 심각한 사고 발생 개연성이 높은 환경으로 교통안전 증진을 위해 기존보다 세분화된 사고분석이 필요하다. 최근 국내 고속도로에서 발생한 사망자 수는 지속적으로 감소하는 추세를 나타내고 있으나 주요 OECD 국가에 비해 여전히 많은 사망자가 발생하고 있다. OECD 수준의 사망자 수 달성을 위해서는 정확한 사고 원인 파악 및 사고 특성을 반영한 체계적인 교통안전대책 수립이 필요한 실정이다. 그러나 기존 고속도로 안전정책 도입체계는 단일 사고원인에 기반하여 단편적이고 일시적인 개선대책만을 적용해 온 문제점이 있다. 또한 교통사고는 한 가지 이상의 요인이 복합적으로 결합되어 발생하며 발생 메커니즘 또한 복잡하다. 기존 사고분석 방식은 사고 빈도를 감소시키기 위해 인적 요인, 도로 환경요인 등 특정 원인에 초점을 맞춰 주요 대책을 적용해왔으나 이러한 방식은 거시적 관점에서는 효과가 있으나 중상 및 대형사고 등 심각한 사고를 감소시키기에는 실효성이 저하된다. 고속도로 교통안전정책 수립을 위해 복합 사고요인을 파악하고 보다 종합적 관점을 통한 사고 분석이 필요한 실정이다. 따라서 본 연구에서는 기존 체계의 문제점을 보완하기 위해 교통안전 복합 영향요인 도출을 통한 정책 수립 방향성에 대해 제시하였다. 빅데이터에 대한 이슈가 증대됨에 따라 최신 연구 동향을 반영하여 다양한 데이터 마이닝 및 기초통계 분석 등의 기법 적용을 통해 변수 간 상호작용 효과를 분석하였다. 이와 더불어 도출된 방법론을 토대로 타당성 검토를 통해 보다 종합적 관점에서 고속도로 교통안전정책 수립을 위한 방향성을 제시하는 연구를 수행하였다.

키워드

복합 영향요인

교통안전대책

데이터 마이닝

고속도로

심각 사고

References

Abdel-Aty M., Haleem K. (2011), Analyzing Angle Crashes at Unsignalized Intersections Using Machine Learning Techniques, Accident Analysis and Prevention, 43, 461-470. 10.1016/j.aap.2010.10.00221094345

Ahmed M., Abdel-Aty M. (2011), The Viability of Using Automatic Vehicle Identification Data for Real-time Crash Prediction, IEEE Transactions on Intelligent Transportation Systems, 13, 459-468. 10.1109/TITS.2011.2171052

Breiman L (2001), Random Forests, Machine Learning, 45(1), 5-32. 10.1023/A:1010933404324

Castro Y., Kim Y. J. (2015), Data Mining on Road Safety: Factor Assessment on Vehicle Accidents Using Classification Models, International Journal of Crashworthiness, 21(2), 104-111. 10.1080/13588265.2015.1122278

Choi J. H., Han S. T., Kang H. C., Kim E. S. (1998), Data Mining Analysis Using Answer Tree, SPSS Academy.

Chung Y.-S. (2013), Factor Complexity of Crash Occurrence: An Empirical Demonstration Using Boosted Regression Trees, Accident Analysis & Prevention, 61, 107-118. 10.1016/j.aap.2012.08.01522975365

Friedman J. (1991), Multivariate Adaptive Regression Splines, Ann. Stat. 19, 1-141. 10.1214/aos/1176347973

Han S. J., Jo W. B., Chang S. (2014), Analysis on Truck Accidents Using Classification and Regression Trees, Journal of Transport Research, 21(4), 87-103. 10.34143/jtr.2014.21.4.87

Hothorn T., Hornik K., Strobl C., Zeileis A., Hothorn M. T. (2015), Package ‘party’, Package Reference Manual for Party Version 0.9-998, 16, 37.

Kang B. S., Kim K. S. (2013), Statistical Analysis of Social Science, Hannarae Academy, 542-566.

Kang Y., Son S., Cho N. (2017), Analysis of Traffic Accidents Injury Severity in Seoul using Decision Trees and Spatiotemporal Data Visualization, Journal of cadastre & land informatix, 47(2), 233-254.

Kim J. H., Yoon S. Y. (2019), Analysis of Policy Changes and User Satisfaction of Road Transportation Services using Opinion Mining Techniques, International Journal of Highway Engineering, 21(5), 65-74. 10.7855/IJHE.2019.21.5.065

Kim Y. J., Kim D. Y. (2018), Methodology of Local Government Policy Issues Through Big Data Analysis, The Journal of the Korea Contents Association, 18(10), 229-235.

Lee J. M., Kim T. H., Lee Y. T., Won J. M. (2008), Developing the Traffic Accident Prediction Model using Classification And Regression Tree Analysis, International journal of highway engineering, 10(1), 31-39.

Lim J.-S. (2018), A Study on the Traffic Safety Management Plan of Commercial Truck: Focusing on Regional Traffic Safety Policy. Korea Logistics Review, 28(3), 101-121. 10.17825/klr.2018.28.3.101

Nam D., Lee S., Kim N. (2016), Issue-Tree and QFD Analysis of Transportation Safety Policy with Autonomous Vehicle, The Journal of The Korea Institute of Intelligent Transport Systems, 15(4), 26-32. 10.12815/kits.2016.15.4.026

Nam G. S., Kim J. H., Ahn B. J. (2004), Factor Extraction of the Traffic Accident Using Data Mining Techniques, Journal of The Korean Data Analysis Society, 6(2), 519-525.

Oh C. S., Lee Y., Ko M. (2016), Establishment of ITS Policy Issues Investigation Method in the Road Section applied Textmining, The Journal of The Korea Institute of Intelligent Transport Systems, 15(6), 10-23. 10.12815/kits.2016.15.6.010

Oh J. S. (2015), Identifying Research Opportunities in the Convergence of Transportation and ICT Using Text Mining Technique, The Korea Transport Institute, 22(4), 93-110. 10.34143/jtr.2015.22.4.93

Park Y. W., Back S., Park S. H., Kwon O. H. (2016), A Study on the Crash Severity of Expressway Work Zones Using Decision Tree, J. Korean Soc. Transp., 34(6), Korean Society of Transportation, 535-547. 10.7470/jkst.2016.34.6.535

Prati G., Pietrantoni L., Fraboni F. (2017), Using Data Mining Techniques to Predict the Severity of Bicycle Crashes, Accident Analysis & Prevention, 101, 44-54. 10.1016/j.aap.2017.01.00828189058

Shin H. J. (2016), An Empirical Comparison for Several Classification Tree Methods in R package, Master dissertation, Ewha Womans University, Korea.

Therneau T. M., Atkinson B., Ripley B. (2015), rpart: Recursive Partitioning, R Package Version 4. 1-10.

Yu Rongjie, Abdel-Aty Mohamed (2014), Analyzing Crash Injury Severity for a Mountainous Freeway Incorporating Real-time Traffic and Weather Data, Safety Science. 63. 50-56. 10.1016/j.ssci.2013.10.012

Zheng Z., Lu P., Lantz B. (2018), Commercial Truck Crash Injury Severity Analysis Using Gradient Boosting Data Mining Model, Journal of Safety Research, 65, 115-124. 10.1016/j.jsr.2018.03.00229776520

Information

Publisher :Korean Society of Transportation
Publisher(Ko) :대한교통학회
Journal Title :Journal of Korean Society of Transportation
Journal Title(Ko) :대한교통학회지
Volume : 39
No :3
Pages :369-382
Received Date : 2021-01-15
Revised Date : 2021-02-15
Accepted Date : 2021-03-16
DOI :https://doi.org/10.7470/jkst.2021.39.3.369

Journal of Korean Society of Transportation ISSN:1229-1366(Print) 2234-4217(Online) 대한교통학회지

All Issue