Analyzing Spatial Non-Stationarity of Factors Affecting Expressway Traffic Crash Severity: A Geographically Weighted Ordered Logit Approach

Bumjoon Bae

doi:10.7470/jkst.2026.44.4.569

All Issue

2026 Vol.44, Issue 4 Preview Page Next Page

Article

Analyzing Spatial Non-Stationarity of Factors Affecting Expressway Traffic Crash Severity: A Geographically Weighted Ordered Logit Approach 고속도로 교통사고 심각도에 미치는 영향요인의 공간적 변동성 연구: 지리가중회귀 접근법을 중심으로

31 August 2026. pp. 569-581

PDF XML

Abstract

This study aims to apply a geographically weighted ordered logit model to empirically examine the spatial non-stationarity of factors affecting traffic crash severity on Korean expressways, thereby relaxing the spatial stationarity assumption inherent in conventional global models that presume identical effects nationwide. To this end, a dataset comprising 40,698 domestic expressway traffic crashes that occurred between 2015 and 2024 was analyzed using an adaptive bandwidth determined through cross-validation. The results indicate that the geographically weighted model provides a superior model fit compared to the corresponding global model. Moreover, the findings reveal substantial spatial variation in the local coefficient estimates of key variables–including nighttime, adverse weather (fog and rain), posted speed limits, vehicle-kilometer traveled, and roadway years of use. In addition, the threshold parameters associated with the latent severity scale exhibit notable spatial heterogeneity. By visualizing the spatial distribution of the estimated local parameters, this study provides empirical evidence supporting the development of geographically tailored road safety management and operational strategies. The findings highlight the need to move beyond uniform, top-down national policies toward region-specific safety interventions, including targeted facility improvements in high-risk areas, traffic volume redistribution strategies, and the institutional sharing of maintenance expertise among road agencies.

Keywords

expressway crash

geographically weighted model

injury severity

ordered logit model

spatial non-stationarity

본 연구는 고속도로 교통사고 심각도에 영향을 미치는 요인들이 전국적으로 동일하게 작용한다는 기존 전역적(global) 모형의 공간적 정상성 가정을 극복하고, 해당 요인들의 공간적 비정상성(spatial non-stationarity)을 실증적으로 규명하기 위해 지리가중 순서형 로짓 모형을 적용하는 데 목적이 있다. 이를 위해 2015년부터 2024년까지 발생한 40,698건의 국내 고속도로 교통사고 데이터를 구축하고 적응형 대역폭을 적용해 분석한 결과, 지리가중 모형이 기존 전역적 모형보다 통계적 적합도 측면에서 우수함이 확인되었다. 또한, 야간 시간대, 기상 조건(안개, 강우), 제한속도, 총통행거리, 고속도로 사용연수 등 주요 변수들의 국지적 계수 추정치가 지리적 위치에 따라 유의미하게 변동할 뿐만 아니라, 관측되지 않는 잠재적 사고심각도를 결정하는 구분 임계값 역시 공간적 이질성을 보인다는 점을 규명하였다. 본 연구는 추정된 계수의 공간적 분포를 시각화하여 제시함으로써, 전국 단위의 획일적인 하향식 정책에서 벗어나 지역별 고위험 구간 맞춤형 시설 개량 및 투자, 교통량 분산 유도, 도로 관리기관 간 유지관리 노하우 공유 등 지리적 맥락에 부합하는 세밀한 고속도로 안전 관리 및 운영 전략을 수립하는 데 필요한 실증적 기초 자료를 제공한다.

키워드

고속도로 교통사고

지리가중회귀

사고심각도

순서형 로짓 모형

공간적 비정상성

References

Adanu E. K., Dzinyela R., Okafor S., Jones S. (2024), Injury-Severity Analysis of Crashes Involving Defective Vehicles and Accounting for the Underlying Socioeconomic Mediators, Heliyon, 10(5), e26944.

10.1016/j.heliyon.2024.e2694438434351PMC10907794

Ahn S. Y., Lee J. D., Chang I. J. (2024), Analysis of Severity Factor of Secondary Incidents along Highway Segment, Journal of Korea Institute of Intelligent Transport Systems, 23(6), 28-42.

10.12815/kits.2024.23.6.28

Anastasopoulos P. C., Mannering, F. L. (2011), An Empirical Assessment of Fixed and Random Parameter Logit Models Using Crash- and Non-Crash-Specific Injury Data, Accident Analysis & Prevention, 43(3), 1140-1147.

10.1016/j.aap.2010.12.024

Bae B. J., Seo C. B. (2022), Do Public-Private Partnerships Help Improve Road Safety? Finding Empirical Evidence Using Panel Data Models, Transport Policy, 126, 336-342.

10.1016/j.tranpol.2022.08.006

Chen Y., Luo R., King M., Shi Q., He J., Hu Z. (2022), Spatiotemporal Analysis of Crash Severity on Rural Highway: A Case Study in Anhui, China, Accident Analysis & Prevention, 165, 106538.

10.1016/j.aap.2021.106538

Fotheringham A. S., Brunsdon C., Charlton M. (2002), Geographically Weighted Regression: The Analysis of Spatially Varying Relationships, Chichester, England, Wiley.

Gomes M. J. T., Cunto F., Silva A. R. (2017), Geographically Weighted Negative Binomial Regression Applied to Zonal Level Safety Performance Models, Accident Analysis & Prevention, 106, 254-261.

10.1016/j.aap.2017.06.011

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 J., Khattak A., Li X., Nie Q., Ling Z. (2020), Bicyclist Injury Severity in Traffic Crashes: A Spatial Approach for Geo-Referenced Crash Data to Uncover Non-Stationary Correlates, Journal of Safety Research, 73, 25-35.

10.1016/j.jsr.2020.02.006

Mun S. R., Lee Y. I. (2011), Analysis of Traffic Crash Severity on Freeway Using Hierarchical Binomial Logistic Model, International Journal of Highway Engineering, 13(4), 199-209.

10.7855/IJHE.2011.13.4.199

Park S. J., Kho S. Y., Park H. C. (2019), The Effects of Road Geometry on the Injury Severity of Expressway Traffic Accident Depending on Weather Conditions, Journal of Korea Institute of Intelligent Transport Systems, 18(2), 12-28.

10.12815/kits.2019.18.2.12

Rhee K. A., Kim J. K., Lee Y. I., Ulfarsson G. F. (2016), Spatial Regression Analysis of Traffic Crashes in Seoul, Accident Analysis & Prevention, 91, 190-199.

10.1016/j.aap.2016.02.023

Wang C., Li S., Shan J. (2021), Non-Stationary Modeling of Microlevel Road-Curve Crash Frequency with Geographically Weighted Regression, ISPRS International Journal of Geo-Information, 10(5), 286.

10.3390/ijgi10050286

Wang X., Liu J., Khattak A. J., Clarke D. B. (2016), Non-Crossing Rail-Trespassing Crashes in the Past Decade: A Spatial Approach to Analyzing Injury Severity, Safety Science, 82, 44-55.

10.1016/j.ssci.2015.08.017

Won M. S., Lee G. L., Oh C., Kang K. W. (2009), A Study on the Application of Accident Severity Prediction Model, J. Korean Soc. Transp., 27(4), Korean Society of Transportation, 167-173.

Xiong Y., Tobias J., Mannering F. L. (2014), The Analysis of Vehicle Crash Injury-Severity Data: A Markov Switching Approach with Road-Segment Heterogeneity, Transportation Research Part B: Methodological, 67, 109-128.

10.1016/j.trb.2014.04.007

Xu N., Liu J., Zhang Z., Jones S. (2025), Injury Severity of Police Officers Involved in Traffic Crashes: A Spatial Analysis of Alabama, Safety Science, 172, 106406.

10.1016/j.ssci.2023.106406

Yan X., He J., Zhang C., Liu Z., Wang C., Qiao B. (2021), Spatiotemporal Instability Analysis Considering Unobserved Heterogeneity of Crash-Injury Severities in Adverse Weather, Analytic Methods in Accident Research, 32, 100182.

10.1016/j.amar.2021.100182

Zafri N. M., Khan A. (2023), Using Geographically Weighted Logistic Regression (GWLR) for Pedestrian Crash Severity Modeling: Exploring Spatially Varying Relationships with Natural and Built Environment Factors, IATSS Research, 47(3), 325-334.

10.1016/j.iatssr.2023.07.004

Zeng Q., Hao W., Lee J., Chen F. (2020), Investigating the Impact of Real-Time Weather Conditions on Freeway Crash Severity: A Bayesian Spatial Analysis, International Journal of Environmental Resaerch and Public Health, 17, 2768.

10.3390/ijerph1708276832316427PMC7215785

Zhang Z., Xu N., Liu J., Jones S. (2024), Exploring Spatial Heterogeneity in Factors Associated with Injury Severity in Speeding-Related Crashes: An Integrated Machine Learning and Spatial Modeling Approach, Accident Analysis & Prevention, 206, 107697.

10.1016/j.aap.2024.107697

Information

Publisher :Korean Society of Transportation
Publisher(Ko) :대한교통학회
Journal Title :Journal of Korean Society of Transportation
Journal Title(Ko) :대한교통학회지
Volume : 44
No :4
Pages :569-581
Received Date : 2026-01-26
Revised Date : 2026-03-14
Accepted Date : 2026-03-21
DOI :https://doi.org/10.7470/jkst.2026.44.4.569

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

All Issue