「家」是人們最熟悉且最常待的地方，但是從103年後全國人員死亡之火災案件中發現，住宅火災造成的死亡人數歷年來位居首位，因此本研究旨在了解人與環境互動對住宅火災風險之影響，以期降低住宅火災所造成的生命財產損失。

本研究的研究區共361個村里，彙整103年至107年發生之一般（集合）住宅火災案件類別資料，依發生地點以及時段區分，輔以相關文獻建立影響住宅火災風險的因子，利用多元線性迴歸(MLR)及地理加權迴歸(GWR)分析與製圖，並比較模型間的差異。

研究結果顯示，不同時段(全時段、依每日三時段區分及非假日、假日時段)的MLR模型中，住宅火災風險與影響因素間正負向關係相同，係數值及解釋能力差異不大，後續以全時段建構GWR模型，調整後的R2從63.6%提升至95.2%，AICc值從668.535下降至44.435，表示GWR模型的解釋力及配適度較佳，另外區域間的住宅火災風險與影響因素存有差異性，說明相同影響因子因不同的自然、人文環境，增加或降低住宅火災風險，呈現空間異質性(spatial heterogeneity)。透過各個村里特性，以及具顯著性影響因素數量和重要性，進行比對後劃分為6個區域，其中4個區域平均降水日數是影響住宅火災風險最重要的因子：新竹舊市區、新豐-湖口地區呈正向影響，說明下雨天待在家中，用電量上升使老舊線路容易因電線過負載造成火災；竹市-竹北地區及竹科-住商重劃區呈負向影響，可能原因與晴天容易外出，屋內未關閉的爐火以及未熄滅的微小火源引起火災。竹南-頭份地區最重要變項為平均兒童人口比，呈正向影響，顯示小孩玩火的潛在危險。新竹城郊地區最重要變項是道路面積比，同時該變項是其他5個區域均有的次重要變項，說明道路密度最能反應實際居住使用情形。整體而言，研究區內人的活動對住宅火災風險具明顯的地域、城鄉差異，最後針對不同區域特性如何降低住宅火災風險，提出消防上的防火策略。

"Home” is the familiar place where we spend a lot of time to stay at. However, after 2014, it is found that the majority of national deadly fires are residential fire cases. The purpose of the study is to understand how the interaction of human activities and the environment affect residential fire to expect to decrease the loss of life and property.

The study is based on three hundred and sixty-one villages in Study Area to residential fire cases from 2014 to 2018 that are distinguished between occurred locations and time periods. With the current data and reference papers, the factors which have the effects on residential fire are established to analyze and draw to compare the differences between two models of Multiple linear regression (MLR) and Geographically Weighted Regression (GWR).

In the MLR model for different cases (all cases, day/evening/night cases, calendar events), the result shows the positive/negative correlation of residential fire risk is the same as factors that have no major difference of coefficient value and explanation. Therefore, the thesis will adopt the GWR model to analyze all cases. According to the results, adjusted R2 is from 63.6% to 95.2% and AICc is from 668.535 to 44.435. The GWR model has better explanation and goodness of fit. What is more, the GWR model indicates there are distinct positive/negative correlation between factors and residential fire risk of the different districts that illustrates the number of residential fire cases are related to the environment and human activities with presenting spatial heterogeneity. By making use of each village characteristic and the number of significant factors, it can make comparisons and overlapping to divide into six districts where the average rainy days of the four districts have the great impact on residential fire risk：In former urban areas of Hsinchu, Xinfeng-Hukou Area, the data are positive correlation. The results say staying at home in the rainy days might raise usage of electrical power to cause the over current load in the cables, especially crucial for old cables, that will have higher risk of the fire. In Hsinchu-Zhubei Area and Residential & Commercial Mixed Readjustment Area in Hsinchu Science Park, the data are negative correlation. The cause might be the human habit. People usually would like to go outside in the good weather. However, they sometimes forget to switch off stoves and do not notice the un-extinguished light flame that will lead to the risk of fire.

The Zhunan-Toufen Area presents the positive correlation and the important factor is proportion of population of children that shows the potential danger of children playing fire. For the suburban areas of Hsinchu, the important factor is the proportion of road areas and the sub-important factor for another five districts as well that mean the density of the roads is rather relevant to real residential status.
In conclusion, based on urban-rural differences and districts, human activities play a significant role in the residential fire. Finally, the thesis provides the fire opinions on how to diminish residential fire risk according to the identities of different districts.

一、中文部分
王翎雅 (2015)。桃園縣建築物火警空間特性分析。國立中央大學土木工程學系
在職專班，碩士論文。
吳榮平 (2007)。都市火災空間分布及發生因素之研究。國立臺北大學都市計劃
研究所，博士論文。
李仁傑 (2017)。家庭保全服務與住宅火災保險之商品整合下最適決策模式研究。
國立東華大學企業管理學系，博士論文。
林炎欣 (2008)。房價特徵模型之空間自我相關問題分析。國立成功大學都市計
劃學系，碩士論文。
邱榮振 (2006)。高雄市火災發生潛勢分析之研究。國立高雄應用科技大學土木
工程與防災科技研究所，碩士論文。
倪如霖 (2016)。公共自行車租借量之影響因素分析-地理加權迴歸和函數資料分
析方法之應用。國立交通大學運輸與物流管理學系，碩士論文。
許智宏 (2006)。都市混合土地使用形態及其影響因素之研究—以台南市為例。
國立成功大學都市計劃學系，碩士論文。
張文菘 (2013)。桃園地區土地利用變遷與影響因素之空間分析。國立臺灣師範
大學地理學系，碩士論文。
張國楨、張文菘、曾露儀 (2012)。都市土地利用與人口老化對基層醫療資源分
布之影響:以台北市為例。地理研究，56，25-40。
陳章瑞 (2013)。以地理加權迴歸模型之空間分析探討都市公園之寧適效益。造
園景觀學報，19，17-46。
梁月、郭红霞 (2014)。火災態勢與社會經濟因素關係研究的新方法。消防科與
技術，33(12)，1443-1446。
彭欣婷 (2015)。都市代謝作用之空間分析與情境評估─以臺北都會區為例。國
立臺北大學都市計劃研究所，碩士論文。
黃崑鏜 (1993)。火災特性與土地使用，空間結構之關連性探討:以台北市78年
至80年之建築物火災案例為對象。國立臺灣大學建築與城鄉研究所，碩士論文。
蔡宛容 (2013)。運用地理加權迴歸方法探討都市土地混合使用空間分佈之影響
因素。國立成功大學都市計劃學系，碩士論文。
蔣得心、林峰田 (2008)。都市地區火災風險分區劃設方法之研究。危機管理學
刊，5(2)，67-74。
錢擴仁 (2006)。住宅火災風險評估之研究。國立臺灣海洋大學河海工程學系，
碩士論文。

二、英文部分
Anselin, L. (1998). "Exploratory spatial data analysis in a geocomputational environment." Geocomputation: A Primer. Chichester New York, Wiley.
Anselin, L. and S. Rey (1991). "Properties of tests for spatial dependence in linear regression models." Geographical analysis, 23(2), 112-131.
Asgary, A., Ghaffari, A., & Levy, J. (2010). "Spatial and temporal analyses of structural fire incidents and their causes: A case of Toronto, Canada." Fire Safety Journal, 45(1), 44-57.
Association, F. E. M. (1997). "Socioeconomic Factors and the Incidence of Fire." United States Fire Administration, National FireData Center.
Bühler, M. D., de Torres, M., & Garibaldi, L. A. (2013). "Demography and Socioeconomic vulnerability influence fire occurrence in Bariloche (Argentina)."Landscape and Urban Planning, 110, 64-73.
Brunsdon, C., Fotheringham, A. S., & Charlton, M. E. (1996). "Geographically weighted regression: a method for exploringspatial nonstationarity." Geographical analysis, 28(4), 281-298.
Budescu, D. V. (1993). "Dominance analysis: a new approach to the problem of relative importance of predictors in multiple regression." Psychological bulletin, 114(3), 542.
Corcoran, J., Higgs, G., & Anderson, T. (2013). "Examining the use of a geodemographic classification in an exploratory analysis of variations in fire incidence in South Wales, UK." Fire Safety Journal, 62, 37-48.
Corcoran, J., Higgs, G., Rohde, D., & Chhetrl, P. (2011). "Investigating the association between weather conditions, calendar events and socio-economic patterns with trends in fire incidence: an Australian case study." Journal of Geographical Systems, 13(2), 193-226.
Fotheringham, A. S., Brunsdon, C., & Charlton, M. (2003). Geographically weighted regression: the analysis of spatially varying relationships, John Wiley & Sons.Jennings, C. R. (2013). "Social and economic characteristics as determinants of residential fire risk in urban neighborhoods: A review of the literature." Fire Safety Journal, 62, 13-19.
Martínez-Fernández, J., Chuvieco, E., & Koutsias, N. (2013). "Modelling long-term fire occurrence factors in Spain by accounting for local variations with geographically weighted regression." Natural Hazards and Earth System Sciences, 13(2), 311-327.
Matthews, S. A. and T.-C. Yang (2012) "Mapping the results of local statistics: Using geographically weighted regression." Demographic research, 26:, 151.
Mercer, D. E. and J. P. Prestemon (2005). "Comparing production function models for wildfire risk analysis in the wildland–urban interface." Forest policy and economics, 7(5), 782-795.
Nisanci, R. (2010). "GIS based fire analysis and production of fire-risk maps: The Trabzon experience." Scientific Research and Essays, 5(9), 970-977.
Oliveira, S., Pereira, J. M. C., San-Miguel-Ayanz, J., & Lourenço, L. (2014). "Exploring the spatial patterns of fire density in Southern Europe using Geographically Weighted Regression." Applied Geography, 51, 143-157.
Preisler, H. K., Brillinger, D. R., Burgan, R. E., & Benoit, J. W. (2004). "Probability based models for estimation of wildfire risk." International Journal of wildland fire, 13(2), 133-142.
Rodrigues, M., Jiménez-Ruano, A., Peña-Angulo, D., & Riva, J. D. L. (2018). "A comprehensive spatial-temporal analysis of driving factors of human-caused wildfires in Spain using geographically weighted logistic regression." Journal of environmental management, 225, 177-192.
Schaenman, P., Hall, S. R., Swartz, A., & Karter, E.(1977)." Procedures for improving the measurement of local fire protection effectiveness." Boston: National Fire Protection Association, 53-71.
Schoennagel, T., Balch, J. K., Brenkert-Smith, H., Dennison, P. E., Harvey, B. J., Krawchuk, M. A., …Whitlock, C. (2017). "Adapt to more wildfire in western North American forests as climate changes." Proceedings of the National Academy of Sciences, 114(18), 4582-4590.
Song, C., Kwan, M. P., & Zhu, J. (2017). "Modeling fire occurrence at the city scale: A comparison between geographically weighted regression and global linear regression." International journal of environmental research and public health, 14(4), 396.
Špatenková, O. and Virrantaus, K. (2013). "Discovering spatio-temporal relationships in the distribution of building fires." Fire Safety Journal, 62, 49-63.
Yamashita, K. (2008). Understanding urban fire: Modeling fire incidence using classical and geographically weighted regression, University of Colorado at Boulder.