서울시 화재 취약지역 예측 및 소방력 공간 최적화 연구 = Predicting Fire Risk Areas and Optimizing Firefighting Resources in Seoul

다국어 초록 (Multilingual Abstract)

The number of casualties and property damage caused by fires has been increasing in Korea recently. Emphasis has been placed on the critical importance of securing a golden time of 5 minutes to effectively respond to such fire incidents and minimize losses. However, achieving timely arrival at fire scenes within this crucial time frame is challenging due to high residential density and increased traffic. Particularly in Seoul, where population density is notably high, the risk of both human casualties and property damage is elevated compared to other regions. Consequently, optimal route selection and efficient spatial deployment of firefighting resources are essential for securing the golden time. This study conducted a comprehensive assessment of fire risk by considering actual fire truck speeds, arrival times, road characteristics, and building features in Seoul using Random Forest and network analysis. We identified areas where achieving the golden time is difficult and fire risk is substantial. Additionally, using kernel density and hot spot analysis, improvement strategies were proposed by selecting optimal locations for fire truck proximity, and the resulting effects were quantitatively analyzed. This study revealed that among the road characteristics, turning frequency impede fire truck speeds during both peak and off-peak hours, and the number of lanes enhance dispatch speed, but the impact difference between the peak and non-peak times was about four times. At the Dong-level, Sa-dang and Jeongneung-dong showed high fire vulnerability based on both peak-hour golden time accessibility and fire risk assessment. At the Gu-level, Gangnam-gu, Gwanak-gu, and Songpa-gu demonstrated heightened vulnerability due to their history of frequent fire incidents. To mitigate fire vulnerability, 17 key improvement areas were identified and then analyzing the effect of fire truck proximity deployment to these locations during peak hours revealed significant enhancement when positioned near Nonhyeon Elementary School, Geumjeong Central Market, and the vicinity of Banghwa Gas Station. This study contributes to understanding how road characteristics impact fire truck response speeds during peak and off-peak times. Furthermore, it provides a quantitative evaluation of implementing measures to secure the golden time and reduce fire risk in vulnerable areas. These results are expected to inform optimal decision-making for emergency fire vehicle operations, considering time frames and road conditions.

참고문헌 (Reference)

1 이윤하 ; 김민석 ; 김화영, "화재대응 취약지역 해소를 위한 비상소화장치 위치선정" 36 (36): 106-113, 2022

2 임정훈 ; 김헌주, "포항시 화재 취약지역 예측 및이에 따른 행정구역별 화재 피해 등급 측정" 21 (21): 166-176, 2021

3 통계청, "토지 용도지역별 기본 현황"

4 우창균 ; 김윤호 ; 박기홍, "클러스터링 알고리즘을이용한 효과적인 그림자 영역 검출 방법" 12 (12): 251-257, 2020

5 성중기 ; 하동익, "출동현황자료 분석을 통한 재난대비 긴급차량 우선신호제어 시스템 도입지역선정방안 연구" 15 (15): 24-35, 2016

6 이경아 ; 이영인, "첨두 및 비첨두시 VMS 교통정보의 가치 변화 연구" 30 (30): 135-147, 2012

7 박병호, "시간대에 따른 교통사고특성 및 교통사고모형 비교분석: 청주시 4지 신호교차로를 중심으로" 57 : 339-347, 2007

8 황의홍 ; 최지훈 ; 최돈묵, "소방차 출동 시 효율적인 골든타임 확보 방안에 관한 연구" 18 : 119-126, 2018

9 서민송 ; 에베르 엔리케 카스티요 오소리오 ; 유환희, "머신러닝을 이용한 경기도 화재위험요인 예측분석" 39 (39): 351-361, 2021

10 국토교통부, "노후 건축지구의 소방시설을 고려한 공간정보 기반의 화재위험지수 평가 플랫폼개발" 2021

11 삼성교통안전문화연구소, "긴급차량 운영 실태및 개선 대책 발표"

12 장기훈 ; 조성범 ; 조용성 ; 손승녀, "골든타임 확보를위한 소방차 통행시간 예측모형 개발" 19 (19): 1-13, 2020

13 이인아 ; 오형록 ; 이준기, "건물별 화재 위험도 예측 및 분석: 재산 피해액과 화재 발생 여부를 바탕으로" 6 (6): 133-144, 2021

14 Tlili, T., "Swarm-based approach for solving the ambulance routing problem" 112 : 350-357, 2017

15 Pedregosa, F., "Scikit-learn : Machine learning in Python" 12 : 2825-2830, 2011

16 Barnett, C. R., "Replacing international temperature–time curves with BFD curve" 42 (42): 321-327, 2007

17 Liu, Y., "Prediction of road traffic congestion based on random for" IEEE 2 : 361-364, 2017

18 Wahlqvist, J., "Influence of the built environment on design fires" 5 : 20-33, 2016

19 김성재 ; 최갑용 ; 장은미 ; 송완영, "GIS를 이용한 화재진압 취약성 지도 제작-대구광역시 달서구를사례로" 18 (18): 11-20, 2015

20 Koreatech(Korea University of Technology & Education), "Cities and provinces 119 general situation room improving situation management research service report" 2015

21 Murray, A. T., "Applications of location analysis" 293-306, 2015

22 Kapoor, A., "A comparative study of K-Means, K-Means++ and Fuzzy C-Means clustering algorithms" IEEE 1-6, 2017

23 Srikanth, L., "A case study on kernel density estimation and hotspot analysis methods in traffic safety management" IEEE 99-104, 2020

24 Kolesar, P., "A Model for Predicting Average Fire Engine Travel Times,Operations Research" 1975

25 소방청, "2022년도 화재통계연감" 2022

26 소방청, "2022 소방청 통계연보" 2022