본문 바로가기
KERD
논문KCI 등재

보행자 사고 위험의 공간적 격차 식별을 위한 베이지안 모형 개발 - 대전광역시 소규모 행정 구역별 상대위험도 변화에 중점을 두고 -

Developing a Bayesian Model to Identify Spatial Disparities in Pedestrian Crash Risk - With a Specific Focus on Changes in Relative Risk across Small Administrative Entities in Daejeon, South Korea -

채수복|임영빈
(사)위기관리이론과실천 (Crisis and Emergency Management: Theory and Praxis)|2024.07|Crisisonomy|제20권 제7호|pp. 49-64|DOI 10.14251/crisisonomy.2024.20.7.49

1

조회

0

다운로드

1

인용

0

참고문헌

키워드

지도mapping보행자 사고pedestrian accident베이지안 접근법크로네커 곱공간적 격차Bayesian approachKronecker productspatial disparity

초록

이 연구는 대전광역시 내 소규모 행정구역(즉, 동)에서의 보행자 사고 위험의 공간적 격차를 파악하기위해 베이지안 기반의 시공간 모형을 개발하고자 한다. 연구의 분석 단위로 행정동을 선택하고 그에따라 보행자 사고를 집계하였다. 그 후 일반화선형혼합모형에 바탕을 둔 일련의 베이지안 모형을 개발하고, 고정효과와 확률효과를 모두 포함하는 최적의 모형을 선정하였다. 연구의 결과에 따르면, 고용률, 노령화 지수, 교차로 밀도, 평균 차량 속도 및 차량 주행거리가 높은 행정동에서 보행자 사고의 상대위험도가 증가하는 것으로 나타났다. 아울러 본 연구에서 도출된 최적 모형은 복잡한 시공간의 상호작용을크로네커 곱을 통해 설명하고 있다. 분석 결과를 종합하여, 보행자 사고의 상대위험도가 높은 지역을지도를 통해 나타내고, 연구 기간 동안 이 위험의 행정동 간 공간적 격차 및 시간에 따른 변화에 대해파악할 수 있었다. 본 연구의 분석기법을 활용하여 우리는 소지역 수준에서의 보행자 안전 정책을 수립하기 위한 객관적 근거를 제시하고, 근거 기반의 의사결정 지원에 기여할 수 있을 것이다.

Abstract

This study attempts to develop a spatio-temporal Bayesian model to identify disparities in pedestrian crash risk across small administrative entities (i.e., dongs) in Daejeon, South Korea. We employed a dong as the unit of analysis and aggregated crashes involving pedestrians for each administrative dong accordingly. We then developed a series of Bayesian models based on the generalized linear mixed model and selected an optimal one that entails both fixed and random effects. The results show that administrative dongs with higher employment, aging index, intersection density, average vehicle speed, and vehicle kilometers traveled tend to increase the relative risk of pedestrian crashes. In addition, our model accounts for sophisticated space-time interaction effects characterized by the Kronecker product. Based on these findings, we present a map of regions with higher relative risk of pedestrian crashes, revealing spatial disparities across dongs and relevant changes over the study period. We argue that our analytic operational framework supports informed decision-making for devising pedestrian safety policy at the small area level.

참고문헌 (42)

  1. Besag, Julian, "Bayesian Image Restoration, with Two Applications in Spatial Statistics", Annals of the Institute of Statistical Mathematics, 43(1), pp.1~20 (1991)
  2. "Data Open Market" (2024)[URL]
  3. Ewing, Reid, "Measuring the Unmeasurable : Urban Design Qualities Related to Walkability", Journal of Urban Design, 14(1), pp.65~84 (2009)
  4. Gim, Tae-Hyoung Tommy, "Maximum Likelihood and Firth Logistic Regression of the Pedestrian Route Choice", International Regional Science Review, 40(6), pp.616~637 (2016)
  5. Guo, Zhan, "Pedestrian Environment and Route Choice : Evidence from New York City and Hong Kong", Journal of Transport Geography, 28, pp.124~136 (2013)
  6. Hong, Kiman, "Research on Selection of Vulnerable Areas to Walking Traffic Accidents for the Elderly Considering Jaywalking Accidents", Journal of the Society of Disaster Information, 17(2), pp.341~350 (2021)
  7. Jeon, Young Eun, "Spatio-temporal Analysis with Risk Factors for Five Major Violent Crimes", The Korean Journal of Applied Statistics, 35(5), pp.619~629 (2022)
  8. Jeong, Bo Mi, "A Comparative Analysis of Applying Traffic Accident Models with Spatial Effects", Crisis and Emergency Management : Theory and Praxis, 14(7), pp.131~140 (2018)
  9. Jun, Hee-Jung, "A Deep Learning Approach to Predict Elderly and Child Pedestrian Crash Hot Spots in the Seoul Metropolitan Area", Journal of Korea Planners Association, 56(7), pp.115~127 (2021)
  10. Kang, Huengsik, "Classifying the Severity of Pedestrian Accidents using Ensemble Machine Learning Algorithms : A Case Study of Daejeon City", Journal of Digital Convergence, 20(5), pp.39~46 (2022)
  11. Kang, Ilsuk, "A Study on the Method for Determining Statistical Risk Boundaries for Spatial Data with Application to Traffic Accident Data", The Korean Data Analysis Society, 25(2), pp.563~572 (2023)
  12. Kelly, C. E., "A Comparison of Three Methods for Assessing the Walkability of the Pedestrian Environment", Journal of Transport Geography, 19(6), pp.1500~1508 (2011)
  13. Kim, Eun Jung, "Effects of Built Environment and Individual Characteristics on Health Condition", Journal of the Korean Regional Science Association, 27(3), pp.27~42 (2011)
  14. Kim, Seunghoon, "Classifying Severity of Senior Driver Accidents In Capital Regions Based on Machine Learning Algorithms", Journal of Digital Convergence, 19(4), pp.25~31 (2021)
  15. Kim, Taewan, "Factors Affecting Spatial Clustering in Pedestrian Collisions Involving the Elderly", Journal of Korea Planners Association, 57(2), pp.61~79 (2022)
  16. Kwon, Cheol Woo, "Comparative Analysis of Traffic Accident Severity of Two-Wheeled Vehicles Using XGBoost", The Journal of the Korea Institute of Intelligent Transport Systems, 20(4), pp.1~12 (2021)
  17. Lawson, Andrew B., "Using R for Bayesian Spatial and Spatio-Temporal Health Modeling" (2021), CRC Press
  18. Lee, Hasik, "Estimation of Pedestrian Crash Exposure Considering Floating Population", Journal of Korean Society of Transportation, 37(4), pp.302~321 (2019)
  19. Lee, Hojun, "An Analysis of Associations between Neighborhood Road Networks and Pedestrian Traffic Accidents: Focused on the TAAS (2012-2014) in Seoul, Korea", Journal of Korea Planning Association, 52(1), pp.141~161 (2017)
  20. Lee, Sang Hyuk, "Comparative Analysis of Elderly’s and Non-elderly’s Human Traffic Accident Severity", The Journal of the Korea Institute of Intelligent Transport Systems, 11(6), pp.133~144 (2012)
  21. Lee, So-Min, "A Study on the Influence Factors of Pedestrian Environment Satisfaction on of the Socially Vulnerable Groups by Life Cycle Stages : The Walkability and the Pedestrian Network in the Neighborhoods", Urban Design, 22(4), pp.17~34 (2021)
  22. Lym, Youngbin, "Bayesian Multilevel Models for Identifying Vehicle Crashes Related to Senior Driver: A Case Study on Daejeon Metropolitan City", Crisisonomy, 20(4), pp.81~95 (2024)
  23. Lym, Youngbin, "Identifying Regions of Excess Injury Risks Associated with Distracted Driving: A Case Study in Central Ohio, USA", SSM - Population Health, 20, pp.101293~ (2022)
  24. Lym, Youngbin, "Examining Neighborhood-Level Vehicle Crash Risks in Daejeon Metropolitan City using Spatiotemporal Bayesian Models", Journal of the Korea Academia-Industrial Cooperation Society, 25(5), pp.176~184 (2024)
  25. Mehta, Vikas, "Look Closely and You Will See, Listen Carefully and You Will Hear: Urban Design and Social Interaction on Streets", Journal of Urban Design, 14(1), pp.29~64 (2009)
  26. Min, Kyungil, "Priority Analysis of Traffic Accident Reduction Factors for Elderly Pedestrian Using AHP : Focused on Daejeon Metropolitan City", The Korean Journal of Local Government Studies, 24(4), pp.349~368 (2021)
  27. Moraga, Paula, "Spatial Statistics for Data Science" (2023), CRC Press
  28. Park, Eunchong, "Interaction Effect Analysis of Determining Factors for Older Pedestrian Crash : Focused on Walking Environment Improvement Projects and Pedestrian Safety Facilities", Journal of Korea Planning Association, 58(5), pp.108~124 (2023)
  29. Park, Joonbum, "A Study on the Environmental Factors Influencing Elderly Pedestrian Traffic Accidents in Busan", Journal of the Association of Korean Geographers, 8(2), pp.289~303 (2019)
  30. Park, Sehyun, "Effects of Road Networks on Vehicle-Pedestrian Crashes in Seoul", The Journal of the Korea Institute of Intelligent Transportation Systems, 19(2), pp.18~35 (2020)
  31. Riebler, Andrea, "An Intuitive Bayesian Spatial Model for Disease Mapping That Accounts for Scaling", Statistical Methods in Medical Research, 25(4), pp.1145~1165 (2016)
  32. Rue, Håvard, "Bayesian Computing with INLA : A Review", Annual Review of Statistics and Its Application, 4(1), pp.395~421 (2017)
  33. Rue, Håvard, "Approximate Bayesian Inference for Latent Gaussian Models by Using Integrated Nested Laplace Approximations", Journal of the Royal Statistical Society Series B : Statistical Methodology, 71(2), pp.319~392 (2009)
  34. Sahu, Sujit K., "Bayesian Modeling of Spatio-Temporal Data with R" (2022), CRC Press
  35. Seo, Jeemin, "A Study on the Physical Environmental Factors Influencing Pedestrian Traffic Accidents in Seoul, Korea : Focused on the 2014 TAAS Data", Journal of Korea Planning Association, 51(3), pp.197~216 (2016)
  36. Simpson, Daniel, "Penalising Model Component Complexity : A Principled, Practical Approach to Constructing Priors", Statistical Science, 32(1), pp.1~28 (2017)
  37. Spiegelhalter, David J., "Bayesian Measures of Model Complexity and Fit", Journal of the Royal Statistical Society Series B: Statistical Methodology, 64(4), pp.583~639 (2002)
  38. Stoker, Philip, "Pedestrian Safety and the Built Environment", Journal of Planning Literature, 30(4), pp.377~392 (2015)
  39. "Traffic Accident Analysis System (TAAS)" (2024)[URL]
  40. Vicente, G., "Crime Against Women in India: Unveiling Spatial Patterns and Temporal Trends of Dowry Deaths in the Districts of Uttar Pradesh", Journal of the Royal Statistical Society. Series a, Statistics in Society, 183(2), pp.655~679 (2019)
  41. Wang, Xiaofeng, "Bayesian Regression Modeling with Inla", pp.2018, Chapman and Hall
  42. Yang, Kwang Sik, "An Analysis of the Determinants of Elderly Pedestrian Fatal Crash", Journal of the Korean Urban Management Association, 34(1), pp.21~33 (2021)

관련 논문AI추천

연령, 성별, 사고 심각도를 고려한 보행취약계층의 교통사고 위험 탐구 - 데이터베이스 설계 및 계층적 베이지안 접근법을 활용하여 -임영빈(사)위기관리이론과실천 · Crisisonomy2024 · 제20권 제8호베이지안 다수준 모형을 활용한 고령운전자 사고의 식별 - 대전광역시 사례를 중심으로 -김기정, 임영빈(사)위기관리이론과실천 · Crisisonomy2024 · 제20권 제4호도로네트워크 특성과 차대사람 사고발생 빈도간의 관련성 분석 : 서울시를 사례로고승영, 김동규, 박세현, 박호철한국ITS학회 · 한국ITS학회 논문지2020 · 제19권 제2호대구광역시 교통약자 보행자 교통사고 공간 군집 분석박성희, 윤상후, 최화빈, 황영은한국디지털정책학회 · 디지털융복합연구2022 · 제20권 제3호서울시 자치구 특성을 고려한 교통약자 교통안전 영향요인 분석김도경, 성중기(사)위기관리이론과실천 · Crisisonomy2016 · 제12권 제9호데이터마이닝과 베이지안 기법을 활용한 소규모 지역의 부주의 운전에 의한 교통사고 위험 모형 구축 - 미국 오하이오주 북부 지역을 대상으로 -임영빈(사)위기관리이론과실천 · Crisisonomy2024 · 제20권 제3호

이 간행물의 다른 논문 (11건)

세월호 참사 연구동향과 문헌고찰이수비, 전예빈pp. 1-13머신러닝을 이용한 삼척 오십천 유역의 홍수유출모의전원혁, 이병현, 김병식, 안성욱pp. 15-23수문지형학적 분석에 의한 세천 취약성 분석최창호, 이승수, 이슬기pp. 25-38시뮬레이션 기반 응급상황 대처에 대한 교육요구도 조사 학령전기 아동어머니를 대상으로박민향pp. 39-48국방위기에 대비한 군무원 인력관리 발전방향 - 육군 군무원 이직의도 분석을 중심으로 -신상범pp. 65-78
원문 보기