차량 화재로 인한 재산피해에 영향을 미치는 화재인자 분석에 관한 연구
A Study on Analysis of Fire Factors Affecting Property Damage on Vehicle Fires
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
최근 10년간 차량 화재 건수는 연평균 14%의 감소세를 보이고 있으나 재산피해는 14%, 인명피해는 사망 15.8%, 부상 6.8% 증가세를 나타내고 있다. 이에 차량 화재로 인한 재산피해나 인명피해를 최소화하기 위해 차량 화재의 요인에 관한 분석의 요구에 따라 특정 유형의 차량, 발생 장소, 발화요인에 관한 차량 화재 위험성 연구가 수행되었다. 그러나 기존 연구에서는 특정 요인에 관한 제한적인 차량 화재 분석 결과를 도출하고 있어 다양한 차량 화재 요인에 대한 분석에는 한계를 보이고 있다. 본 논문은 소방청 국가화재정보시스템의 차량 화재 데이터를 대상으로 재산피해를 중심으로 다양한 차량 화재의 발생 원인과 영향요인을 빅데이터 분석 기법을 적용하여 분석하였다. 이를 위해 실험에서는 상관관계 및 유의확률 분석을 통해 재산피해에 관한 화재 요인 분석을 위한 인자를 선정하고, 랜덤 포레스트 알고리즘을 이용하여 차량 화재에서 재산피해에 영향을 미치는 화재 인자의 중요도를 도출하였다. 실험 결과에서 화재출동소요시간(13.4%), 소방서거리(13.1%), 화재진압시간(12.7%), 119안전센터거리’(12.4%), ‘동원인력소계’(11.6%) 등의 순으로 재산피해에 가장 높은 비중으로 영향을 미치는 것으로 분석되었다. 이러한 분석 결과로부터 고찰을 통해 차량 화재로 인한 피해의 최소화를 위해서는 화재 현장의 접근성 개선, 화재진압을 위한 적절한 소방력 편제, 화재 소화 장비 구비 등에 관한 제도적 개선방안이 필요함을 제시하였다.
Abstract
Over the past 10 years, the number of vehicle fires has been decreasing by an average of 14% per year, but property damage is increasing by 14%, and human casualties are increasing by 15.8% for deaths and 6.8% for injuries. Accordingly, in order to minimize property damage or human casualties caused by vehicle fires, vehicle fire risk researches have been conducted on specific types of vehicles, occurrence locations, and ignition factors in accordance with the need for analysis of the factors of vehicle fires. However, previous researches have derived limited vehicle fire analysis results regarding specific factors, showing limitations in analyzing various vehicle fire factors. This paper analyzed the causes and influencing factors of various vehicle fires, focusing on property damage, by applying big data analysis techniques to vehicle fire data from the National Fire Data System. To this end, in the experiment, factors for fire factor analysis of property damage were selected through correlation and significance probability analysis, and the importance of fire factors affecting property damage in vehicle fires was derived using the random forest algorithm. In the experimental results, fire factors that affect property damage are dispatch time(13.4%), fire station distance(13.1%), fire suppression time(12.7%), 119 safety center distance(12.4%), and participation personnel(11.6%) were analyzed in the following order. Through consideration of these analysis results, it was suggested that in order to minimize damage caused by vehicle fires, institutional improvement measures such as improving accessibility to fire sites, organizing appropriate firefighting forces for fire suppression, and providing fire extinguishing equipment are necessary.
참고문헌 (19)
- "National Fire Data System" (2023)[URL]
- J. Ryu, "The analysis of the risk of vehicle fires in Korea and the effectiveness of initial fire extinguishing", Journal of the Korea Society of Disaster Information, 15(1), pp.20~27 (2019)
- D. L. Zhang, "Study on vehicle fire safety:Statistic, investigation methods and experimental analysis", Safety Science, 17, pp.194~204 (2019)
- Y-H. Kim, "A study on the case analysis and investigation techniques of vehicle fires", Master, Dept. of Fire Safety Engineering, pp.2015
- A. Dorsz, "Analysis of fire hazards associated with the operation of electric vehicles in enclosed structures", Energies, 15(1) (2021)
- J-I Lee, "A study on the fire risk and the fire investigation of hybrid cars", Transactions of KSAE, 27(2), pp.117~124 (2019)
- S. Cui, "Analysis of the fire hazard and leakage explosion simulation of hydrogen fuel cell vehicles", Journal of Thermal Science and Engineering Progress, 41 (2023)
- T. Navestad, "A survey of vehicle fires in Norwegian road tunnels 2008–2011", Tunnelling and Underground Space Technology, 41, pp.104~112 (2014)
- H. R. Boehmer, "Fire hazard analysis of modern vehicles in parking facilities", Fire Technology, 57(5), pp.2097~2127 (2021)
- B. Truchot, "An experimental evaluation of toxic gas emissions from vehicle fires", Fire Safety Journal, 97, pp.111~118 (2018)
- H-P. Lee, "A study on the fire risk of car interior materials", Journal of Korean Institute of Fire Sci. & Eng, 24(2) (2010)
- "National Fire Agency_Address information of outbreak of fire" (2020)[URL]
- K. Pearson, "Mathematical contributions to the theory of evolution. III. Regression, heredity and panmixia", Philosophical transactions of the Royal Society of London. Series A, Containing papers of a mathematical or physical character, 187, pp.253~318 (1896)
- A. J. Bishara, "Testing the significance of a correlation with nonnormal data: Comparison of Pearson, Spearman, transformation, and resampling approaches", Psychological methods, 17(3), pp.399~417 (2012)
- R. L. Wasserstein, "Moving to a world beyond “p<0.05”", The American Statistician, 73, pp.1~19 (2019)
- T. Y. Kim, "A different perspective on the p-value problem with big data", Quantitative Bio-Science, 39(1), pp.71~74 (2020)
- L. Breiman, "Random forests", Machine Learning, 45, pp.5~32 (2001)
- V. Svetnik, "Lecture Notes in Computer Science, Vol. 3077", pp.334~343 (2004)
- W. Jeong, "Analysis of fire impact factors and operational effectiveness of firefighting facilities in multiplex available premises", JKITS, 17(3), pp.445~453 (2022)