All Issue

2023 Vol.41, Issue 6 Preview Page

Article

Identification of Crash-Prone Road Segments Based on an Ensemble Machine Learning Strategy : Use of Driving Behavior Data

개별 차량 주행행태 자료를 이용한 앙상블 학습전략 기반 도로 위험 구간 식별 기법

Jeonghoon JEE1
,
Yeseo GU2
,
Young JO3
,
Cheol OH4*
지 정훈1
,
구 예서2
,
조 영3
,
오 철4*
1Master’s Degree, Department of Transportation and Logistics Engineering, Hanyang University, Gyeonggi 15588, Korea
2Master’s Degree, Department of Transportation and Logistics Engineering, Hanyang University, Gyeonggi 15588, Korea
3Senior Researcher, Department of Road and Traffic Research, Korea Institute of Civil Engineering and Building Technology, Gyeonggi 10223, Korea
4Professor, Department of Transportation and Logistics Engineering, Hanyang University, Gyeonggi 15588, Korea
1한양대학교 교통·물류공학과 석사과정
2한양대학교 교통 · 물류공학과 석사과정
3한국건설기술연구원 도로교통연구본부 수석연구원
4한양대학교 교통·물류공학과 교수
*Corresponding Author
31 December 2023. pp. 739-752
PDF XML Citation
Abstract

Various road safety evaluation studies have been conducted for traffic safety management. To date, traffic safety measures have been primarily focused on reactive safety management using macroscopically aggregated traffic data collected by infrastructure-based detection systems. However, the need for proactive safety management has been raised as individual vehicle driving behavior data is now readily available. Detection of crash prone road segments allows for realizing more proactive safety management to prevent traffic crashes. This study applied statistical techniques and machine learning models based on individual vehicle data capable of analyzing driving behavior. It also aims to implement a more accurate road hazard identification model using an ensemble learning strategy. Twenty-two driving safety indicators were derived to characterize vehicle driving behavior from inter-vehicular interaction, longitudinal, and lateral perspectives, based on the data collected from PVD, which provides individual vehicle driving behavior data.Ten of these indicators, which showed significant correlation with traffic accident data, were selected through statistical correlation analysis. The average number of accidents in the analysis section of traffic accident data was used to categorize road sections into hazardous and normal categories, serving as the target variable. The ten statistically significant driving safety indicators were used as explanatory variables. Classification models such as SVM, ANN, and KNN were performed. Ensemble learning, known for enhancing robustness compared to single models, was employed by combining these three models to derive the final model. The accuracy of the model by ensemble learning strategy was 90.8%, showing an improvement of up to 4.8% and at least 2.7% compared to single models. This indicates that ensemble learning strategies can enable more accurate implementation of road hazard identification models for proactive traffic safety management. The methodology of this study is expected to serve as foundational data for implementing models for real-time crash prone road segments identification in environments where real-time data collection and utilization are feasible.

Keywords
crash prone road segments detection model
driving safety indicator
machine learning
ensemble learner
proactive traffic safety management

교통안전 관리를 위한 다양한 도로 안전성 평가 연구가 진행되고 있다. 지금까지 교통사고 안전 대책은 인프라 기반 검지 체계에서 수집되는 거시적으로 집계된 교통 데이터를 활용한 사후 교통안전 관리가 주를 이루었다. 그러나 현재 개별 차량 주행행태 자료의 수집이 가능해짐에 따라 선제적 교통안전 관리의 필요성이 제기되고 있다. 사전에 도로 위험 구간 식별이 가능하다면 교통사고 예방을 위한 보다 능동적인 교통안전 관리가 가능하다. 본 연구는 주행행태 분석이 가능한 개별 차량 자료를 기반으로 통계적 기법과 머신러닝 모델을 통합적으로 적용하고, 앙상블 학습전략을 활용하여 보다 정확한 도로 위험 구간 식별 모델을 구현하고자 한다. 개별 차량 주행행태 자료인 PVD에서 수집되는 항목을 기반으로 차량의 주행행태를 차량간 상호작용, 종방향 및 횡방향 관점에서 특성화할 수 있는 22가지 주행 안전성 평가지표를 산출하였다. 이를 통계적 상관분석을 통해 교통사고 건수 자료와 유의미한 상관성을 가지는 10가지 주행 안전성 평가지표를 선정하였다. 교통사고 건수 자료의 분석 구간 내 평균 사고 건수를 기준으로 집계 단위별 위험 도로구간과 일반 도로구간으로 범주화하여 목표변수로 활용하였고, 상관분석 결과 통계적으로 유의미한 10가지 주행 안전성 평가지표를 설명변수로 활용하여 SVM, ANN, KNN의 분류 모델을 수행하였다. 앙상블 학습은 다중 모델 조합으로, 단일 모델 대비 견고성을 향상시킬 수 있다고 알려져 있다. 앞서 수행한 3가지 모델을 앙상블 학습하여 최종 모델을 도출하였다. 앙상블 학습전략에 의한 모델의 정확도는 90.8%로 단일 모델 대비 최대 4.8%, 최소 2.7% 향상된 결과가 나타났다. 이는 선제적 교통안전 관리를 위한 도로 위험 구간 식별 모델 구현 시 앙상블 학습전략은 보다 정확한 구현이 가능함을 의미한다. 본 연구의 방법론은 향후 실시간 데이터 수집 및 활용이 가능한 환경에서 실시간 도로 위험 구간 식별을 위한 모델 구현 시 기초 자료로 활용이 가능할 것으로 기대된다.

키워드
도로 위험 구간 식별 모델
주행 안전성 평가지표
머신러닝
앙상블 학습전략
선제적 교통안전 관리
References
1
AlMamlook R. E., Kwayu K. M., Alkasisbeh M. R., Frefer A. A. (2019), Comparison of Machine Learning Algorithms for Predicting Traffic Accident Severity, In 2019 IEEE Jordan International Joint Conference on Electrical Engineering and Information Technology (JEEIT) (272-276), IEEE. 10.1109/JEEIT.2019.8717393
2
Ara Z., Hashemi M. (2021), Identifying the Severity of Road Accident Impact on Traffic Flow by Ensemble Model, In 2021 IEEE 22nd International Conference on Information Reuse and Integration for Data Science (IRI) (115-122), IEEE. 10.1109/IRI51335.2021.00022
3
Archer J. (2005), Indicators for Traffic Safety Assessment and Prediction and Their Application in Micro-simulation Modelling: A Study of Urban and Suburban Intersections, KTH Royal Institute of Technology, Doctoral dissertation.
4
Arvin R., Kamrani M., Khattak A. J. (2019), How Instantaneous Driving Behavior Contributes to Crashes at Intersections: Extracting Useful Information from Connected Vehicle Message Data, Accident Analysis & Prevention, 127, 118-133. 10.1016/j.aap.2019.01.01430851563
5
Augustine T., Shukla S. (2022, April), Road Accident Prediction Using Machine Learning Approaches, In 2022 2nd International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE) (808-811), IEEE. 10.1109/ICACITE53722.2022.9823499
6
Bagdadi O., Varhelyi A. (2013), Development Of A Method For Detecting Jerks In Safety Critical Events, Accident Analysis & Prevention, 50, 83-91. 10.1016/j.aap.2012.03.03223200443
7
Boo Y., Choi Y. (2022), Comparison of Mortality Prediction Models for Road Traffic Accidents: An Ensemble Technique for Imbalanced Data, BMC public health, 22(1), 1476. 10.1186/s12889-022-13719-335918672PMC9344638
8
Cherkassky V., Ma Y. (2004), Practical Selection of SVM Parameters and Noise Estimation for SVM Regression, Neural Networks, 17(1), 113-126. 10.1016/S0893-6080(03)00169-214690712
9
Cho E., Yun Y., Oh C., Lee G. (2023), Derivation of Riding Risk Precursors Using 100 Delivery Motor Scooter Naturalistic Riding Study, Accident Analysis & Prevention, 190, 107186. 10.1016/j.aap.2023.10718637369163
10
Cooper D. F., Ferguson N. (1976), Traffic Studies at T-Junctions. 2. A Conflict Simulation Record, Traffic Engineering & Control, 17(Analytic).
11
Cunto F., Saccomanno F. F. (2008), Calibration and Validation of Simulated Vehicle Safety Performance at Signalized Intersections, Accident analysis & prevention, 40(3), 1171-1179. 10.1016/j.aap.2008.01.00318460386
12
Dietterich T. G. (1997), Machine-learning Research, AI Magazine, 18(4), 97-97.
13
Drew P. J., Monson J. R. (2000), Artificial Neural Networks, Surgery, 127(1), 3-11. 10.1067/msy.2000.10217310660751
14
FHWA (2003), Surrogate Safety Measures from Traffic Simulation Models, Federal Highway Administration Report FHWA-RD-03-050. McLean, VA.
15
Hansen L. K., Salamon P. (1990), Neural network ensembles, IEEE Transactions on Pattern Analysis & Machine Intelligence, (10), 993-1001. 10.1109/34.58871
16
Haywar J. (1971), Near Misses as A Measure of Safety At Urban Intersections, Pennsylvania Transportation and Traffic Safety Center.
17
HE Haibo et al. (2008), ADASYN: Adaptive Synthetic Sampling Approach for Imbalanced Learning, 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), IEEE, 1322-1328. 10.1109/IJCNN.2008.4633969
18
Jang J., Ko J., Park J., Oh C., Kim S. (2020), Identification of Safety Benefits by Inter-vehicle Crash Risk Analysis Using Connected Vehicle Systems Data on Korean Freeways, Accident Analysis & Prevention, 144, 105675. 10.1016/j.aap.2020.10567532634761
19
Jo Y., Jang J., Park S., Oh C. (2021), Connected Vehicle-based Road Safety Information System (CROSS): Framework and Evaluation, Accident Analysis & Prevention, 151, 105972. 10.1016/j.aap.2021.10597233465744
20
Jo Y., Jung A., Park H., Park J., Oh C. (2022), Prioritizing Driving Safety Indicators Using Real-world C-ITS Data to Identify Hazardous Freeway Sections, J. Korean Soc. Transp., 40(6), Korean Society of Transportation, 863-878. 10.7470/jkst.2022.40.6.863
21
Kim Y., Kang K., Park J., Oh C. (2023), A Methodology for Prioritizing Safety Indicators Using Individual Vehicle Trajectory Data, Journal of Transportation Safety & Security, 1-25. 10.1080/19439962.2023.2178567
22
Ko J., Jang J., Oh C. (2021), Assessing the Safety Benefits of In-vehicle Warning Information by Vehicle Interaction Analysis in C-ITS Environments, J. Korean Soc. Transp., 39(1), Korean Society of Transportation, 1-13. 10.7470/jkst.2021.39.1.001
23
Ko J., Jang J., Oh C. (2022), A Multi-agent Driving Simulation Approach for Evaluating The Safety Benefits of Connected Vehicles, IEEE Transactions on Intelligent Transportation Systems, 23(5), 4512-4524. 10.1109/TITS.2020.3045675
24
Kuncheva L. I. (2004), Combining Pattern Classifiers: Methods and Algorithms, John Wiley & Sons. 10.1002/0471660264
25
Mahdinia I., Mohammadnazar A., Arvin R., Khattak A. J. (2021), Integration of Automated Vehicles in Mixed Traffic: Evaluating Changes in Performance of Following Human-Driven Vehicles, Accident Analysis & Prevention, 152, 106006. 10.1016/j.aap.2021.10600633556655
26
Njobelo G., Sando T., Sajjadi S., Mtoi E., Ozguven E. E., Sobanjo J. (2018), Safety Evaluation of The Advanced Stop Assist System in Connected Vehicle Environment, Transportation Research Record, 2672(22), 47-57. 10.1177/0361198118797831
27
Oh C., Park S., Ritchie S. G. (2006), A Method for Identifying Rear-end Collision Risks Using Inductive Loop Detectors, Accident Analysis & Prevention, 38(2), 295-301. 10.1016/j.aap.2005.09.00916246286
28
Opits D., Maclin R. (1999), Popular ensemble methods: An empirical study. Journal of artificial intelligence research, 11, 169-198. 10.1613/jair.614
29
Sakhare R. S., Desai J. C., Mahlberg J., Mathew J. K., Kim W., Li H., Bullock D. M. (2021), Evaluation of The Impact of Queue Trucks with Navigation Alerts Using Connected Vehicle Data, Journal of Transportation Technologies, 11(4), 561-576. 10.4236/jtts.2021.114035
30
Wang X., Wang T., Tarko A., Tremont P. J. (2015), The Influence of Combined Alignments on Lateral Acceleration on Mountainous Freeways: A Driving Simulator Study, Accident Analysis & Prevention, 76, 110-117. 10.1016/j.aap.2015.01.00325626165
31
Xiao, J. (2019), SVM and KNN Ensemble Learning for Traffic Incident Detection, Physica A: Statistical Mechanics and its Applications, 517, 29-35. 10.1016/j.physa.2018.10.060
32
Yuan R., Li Z., Guan X., Xu L. (2010), An SVM-based Machine Learning Method for Accurate Internet Traffic Classification, Information Systems Frontiers, 12, 149-156. 10.1007/s10796-008-9131-2
33
Zhang S., Li X., Zong M., Zhu X., Wang R. (2017), Efficient kNN Classification with Different Numbers of Nearest Neighbors, IEEE Transactions on Neural Networks and Learning Systems, 29(5), 1774-1785. 10.1109/TNNLS.2017.267324128422666
Information
  • Publisher :Korean Society of Transportation
  • Publisher(Ko) :대한교통학회
  • Journal Title :Journal of Korean Society of Transportation
  • Journal Title(Ko) :대한교통학회지
  • Volume : 41
  • No :6
  • Pages :739-752
  • Received Date : 2023-07-12
  • Revised Date : 2023-08-21
  • Accepted Date : 2023-12-16
  • DOI :https://doi.org/10.7470/jkst.2023.41.6.739
Journal Informaiton Journal of Korean Society of Transportation Journal of Korean Society of Transportation
Archives
: 2015 Vol.34~
  • scopus
  • NRF
  • KOFST
  • crossref cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close