Assessment of Crash Prediction Models for Intersections with Severity Weight Parameters Using Data Science Approaches

Seung-oh Son; Juneyoung Park

doi:10.7470/jkst.2022.40.2.190

All Issue

2022 Vol.40, Issue 2 Preview Page Next Page

Article

Assessment of Crash Prediction Models for Intersections with Severity Weight Parameters Using Data Science Approaches 데이터마이닝 기반의 사고심각도 가중치 적용 예측변수를 활용한 교차로 사고예측모형 개발

30 April 2022. pp. 190-204

PDF XML

Abstract

In this study, crash prediction models for urban intersections were developed using an index as dependent variables reflecting the crash severity weight from data mining technique. In general, the crash prediction model is also called Safety performance function (SPF), and is a regression model with the number of crashes aggregated in the sections or intersections to be analyzed as dependent variables. However, the number of crashes is simply the counted frequency of crashes, and the data does not reflect the characteristics of the crash severity factors. In this study, the crash severity analysis were conducted on crashes that occurred at urban intersections. In addition, the crash prediction models were developed using the crash score index reflecting the results of the severity analysis as a dependent variable. Random forest (RF) and Extreme boosting (XGB) were used for the analysis of intersection crash severity, and NB, Com-poisson, and XGB regression tree model were developed for crash prediction models. Finally, the index reflecting the RF and XGB weight results showed the best predictive performance. Since the proposed dependent variable reflects the results of crash severity analysis as well as excellent predictive performance, important implications based on severity factors can be presented. The model presented in this study can be used for safety evaluation and policy design of individual intersections.

Keywords

crash prediction model

crash severity analysis

extreme gradient boosting (XGB)

random forest

safety performance function (SPF)

본 연구에서는 데이터마이닝 기법을 통해 도출한 사고심각도 가중치를 적용한 환산 사고건수를 종속변수로 하는 도심부 교차로 사고예측모형을 개발하였다. 일반적으로 사고예측모형(Crash prediction model)은 안전성능함수(Safety performance functions)로도 불리며, 분석 대상인 구간 또는 교차로에서 집계된 사고건수를 종속변수로 하는 회귀모형이다. 그러나 여기서 사고건수는 단순히 집계된 사고의 빈도이며 사고심각도 및 사고의 특성변수가 반영되지 않은 데이터이다. 본 연구는 국내 교차로에서 발생한 사고를 대상으로 사고심각도 분석을 수행하여 심각한 사고 발생에 유의한 영향을 미치는 변수를 정량화하였으며, 이를 사고건수에 반영한 가중치 적용 사고건수를 종속변수로 설정하여 모형을 개발하였다. 교차로 사고심각도 분석에는 Random forest(RF)와 Extreme gradient boosting(XGB) 방법론이 활용되었으며, 사고예측모형은 NB, Com-poisson, 그리고 XGB 회귀트리가 활용되었다. 최종적으로 RF와 XGB 가중치 결과가 반영된 지표가 예측성능이 가장 우수한 것으로 나타났다. 제안된 종속변수는 우수한 예측성능 뿐만 아니라 사고심각도 분석 결과를 반영하고 있기 때문에, 심각도 요인 기반의 중요한 시사점을 제시할 수 있다. 본 연구에서 제시한 모형은 개별 교차로의 안전성 평가 및 정책 설계에 유효한 자료로 활용될 수 있다.

키워드

사고예측모형

사고심각도 분석

익스트림 그래디언트 부스팅

랜덤포레스트

안전성능함수

References

AASHTO (2010), Highway Safety Manual, First ed. Washington, D.C.

Abdel-Aty M. (2003), Analysis of Driver Injury Severity Levels at Multiple Locations Using Ordered Probit Models, Journal of Safety Research, 34(5), 597-603. 10.1016/j.jsr.2003.05.00914733994

Abdel-Aty M., Keller J., Brady P. A. (2005), Analysis of Types of Crashes at Signalized Intersections by Using Complete Crash Data and Tree-based Regression, Transportation Research Record, 1908(1), 37-45. 10.1177/0361198105190800105

Ahmadi A., Jahangiri A., Berardi V., Machiani S. G. (2020), Crash Severity Analysis of Rear-end Crashes in California Using Statistical and Machine Learning Classification Methods, Journal of Transportation Safety & Security, 12(4), 522-546. 10.1080/19439962.2018.1505793

Breiman L. J. M. l. (2001), Random Forests, 45(1), 5-32. 10.1023/A:1010933404324

Chang L. Y., Chen W. C. (2005), Data Mining of Tree-based Models to Analyze Freeway Accident Frequency, Journal of Safety Research, 36(4), 365-375. 10.1016/j.jsr.2005.06.01316253276

Chen, T., He, T., Benesty, M., Khotilovich, V., Tang, Y., Cho, H., Chen, K. (2015), Xgboost: Extreme Gradient Boosting, R Package Version 0.4-2, 1(4), 1-4.

Conway R. W., Maxwell W. L. (1962), A Queuing Model with State Dependent Service Rates, Journal of Industrial Engineering, 12(2), 132-136.

Franklin J. (2005), The Elements of Statistical Learning: Data Mining, Inference and Prediction, The Mathematical Intelligencer, 27(2), 83-85. 10.1007/BF02985802

Ijaz M., Zahid M., Jamal A. (2021), A Comparative Study of Machine Learning Classifiers for Injury Severity Prediction of Crashes Involving Three-wheeled Motorized Rickshaw, Accident Analysis & Prevention, 154, 106094. 10.1016/j.aap.2021.10609433756425

Kang K., Park J., Lee K., Park J., Song C. (2021), Development of Time-based Safety Performance Function for Freeways, The Journal of The Korea Institute of Intelligent Transport Systems, 20(6), 203-213. 10.12815/kits.2021.20.6.203

Kim H., Jeon G., Jang J., Yoon I. (2020), Analysis of Factors Affecting Buses and Trucks Crash Severity Using Meta Analysis. 10.7470/jkst.2020.38.6.520

Kuhnert P. M., Do K. A., McClure R. (2000), Combining Non-parametric Models with Logistic Regression: An Application to Motor Vehicle Injury Data, Computational Statistics & Data Analysis, 34(3), 371-386. 10.1016/S0167-9473(99)00099-7

Lee J. Y., Chung J. H., Son B. S. (2008), Analysis of Traffic Accident Severity for Korean Highway Using Structural Equations Model, J. Korean Soc. Transp., 26(2), Korean Society of Transportation, 17-24.

Liu C., Sharma A. (2018), Using the Multivariate Spatio-temporal Bayesian Model to Analyze Traffic Crashes by Severity, Analytic Methods inAaccident Research, 17, 14-31. 10.1016/j.amar.2018.02.001

Lord D., Guikema S. D. (2012), The Conway-Maxwell Poisson Model for Analyzing Crash Data, Applied Stochastic Models in Business and Industry, 28(2), 122-127. 10.1002/asmb.937

Lord D., Mannering F. (2010), The Statistical Analysis of Crash-frequency Data: A Review and Assessment of Methodological Alternatives, Transportation Research Part A: Policy and Practice, 44(5), 291-305. 10.1016/j.tra.2010.02.001

National Highway Traffic Safety Administration (NHTSA) (2015), Traffic Safety Factors, A Compilation of Motor Vehicle Crash Data From the Fatality Analysis Reporting System and the General Estimates System, United States of America.

Park J., Abdel-Aty M. (2015), Assessing the Safety Effects of Multiple Roadside Treatments Using Parametric and Nonparametric Approaches, Accident Analysis & Prevention, 83, 203-213. 10.1016/j.aap.2015.07.00826291920

Prato C. G., Kaplan S., Patrier A., Rasmussen T. K. (2018), Considering Built Environment and Spatial Correlation in Modeling Pedestrian Injury Severity, Traffic Injury Prevention, 19(1), 88-93. 10.1080/15389588.2017.132953528534647

Sellers K. F., Borle S., Shmueli G. (2012), The COM-Poisson Model for Count Data: A Survey of Methods and Applications, Applied Stochastic Models in Business and Industry, 28(2), 104-116. 10.1002/asmb.918

Sellers K. F., Shmueli G. (2010), A Flexible Regression Model for Count Data, The Annals of Applied Statistics, 943-961. 10.1214/09-AOAS306

Shirani-Bidabadi N., Mallipaddi N., Haleem K., Anderson M. (2020), Developing Bicycle-vehicle Crash-specific Safety Performance Functions in Alabama Using Different Techniques, Accident Analysis & Prevention, 146, 105735. 10.1016/j.aap.2020.10573532835954

Son S., Park J., Kim M., Choe B. (2019), Assessing the Safety Effects of Reduction of Speed Limit on Urban Roads, J. Korean Soc. Transp., 37(6), Korean Society of Transportation, 514-524. 10.7470/jkst.2019.37.6.514

TAAS (Traffic Accident Analysis System) (2021), Statistical Analysis of Traffic Accidents (2020).

Tang J., Liang J., Han C., Li Z., Huang H. (2019), Crash Injury Severity Analysis Using a Two-layer Stacking Framework, Accident Analysis & Prevention, 122, 226-238. 10.1016/j.aap.2018.10.01630390518

Wang K., Simandl J. K., Porter M. D., Graettinger A. J., Smith R. K. (2016), How the Choice of Safety Performance Function Affects the Identification of Important Crash Prediction Variables, Accident Analysis & Prevention, 88, 1-8. 10.1016/j.aap.2015.12.00526710265

Wang K., Zhao S., Jackson E. (2019), Functional Forms of the Negative Binomial Models in Safety Performance Functions for Rural Two-lane Intersections, Accident Analysis & Prevention, 124, 193-201. 10.1016/j.aap.2019.01.01530665054

Won M. S., Lee G. R., Gang G. U. (2009), A Study on the Application of Accident Severity Prediction Model, J. Korean Soc. Transp., 27(4), Korean Society of Transportation, 167-173.

Yan X., He J., Zhang C., Liu Z., Qiao B., Zhang H. (2021), Single-vehicle Crash Severity Outcome Prediction and Determinant Extraction Using Tree-based and Other Non-parametric Models, Accident Analysis & Prevention, 153, 106034. 10.1016/j.aap.2021.10603433647597

Zhao M., Liu C., Li W., Sharma A. (2018), Multivariate Poisson-lognormal Model for Analysis of Crashes on Urban Signalized Intersections Approach, Journal of Transportation Safety & Security, 10(3), 251-265.

Information

Publisher :Korean Society of Transportation
Publisher(Ko) :대한교통학회
Journal Title :Journal of Korean Society of Transportation
Journal Title(Ko) :대한교통학회지
Volume : 40
No :2
Pages :190-204
Received Date : 2022-01-28
Revised Date : 2022-02-21
Accepted Date : 2022-02-24
DOI :https://doi.org/10.7470/jkst.2022.40.2.190

Journal of Korean Society of TransportationISSN:1229-1366(Print) 2234-4217(Online)대한교통학회

All Issue