A decision framework to assess urban fire vulnerability in cities of developing nations: empirical evidence from Mumbai

Abstract The article aims to analyze the various causes and relationships of fire with the urban pattern. Various spatial analytics and geostatistical techniques are applied to reveal spatiotemporal variations in the datasets. The novel machine learning framework models important variables identified by random forest technique as predictors to develop GWR models. The framework is applied to establish the relationship between fire vulnerability and urban patterns for two periods (day and night) for the southern region of Mumbai city. We found that the urban pattern has a strong relationship with fire vulnerability, especially during the day time. High R-square values of 0.9086 and 0.7448 are achieved for the day and night periods, respectively. Further, significant differences in the influence of the predictors is observed during the periods. As cities are becoming more prone to fire, this study has the potential to help decision-makers with proactive measures over time and space.

[1]  Kirsi Virrantaus,et al.  Discovering spatio-temporal relationships in the distribution of building fires , 2013 .

[2]  Anka Lisec,et al.  Estimating the Performance of Random Forest versus Multiple Regression for Predicting Prices of the Apartments , 2018, ISPRS Int. J. Geo Inf..

[3]  P. Christenson.  Variable selection in multiple regression , 1982 .

[4]  Ananto Yudono,et al.  Expert System Development for Urban Fire Hazard Assessment. Study Case: Kendari City, Indonesia , 2017 .

[5]  Charles Jennings,et al.  Social and economic characteristics as determinants of residential fire risk in urban neighborhoods: A review of the literature , 2013 .

[6]  Hyung-Sup Jung,et al.  Spatial prediction of flood susceptibility using random-forest and boosted-tree models in Seoul metropolitan city, Korea , 2017 .

[7]  Marco Piras,et al.  An INSPIRE-compliant open-source GIS for fire-fighting management , 2017, Natural Hazards.

[8]  Weiguo Song,et al.  A Comparison between Spatial Econometric Models and Random Forest for Modeling Fire Occurrence , 2017 .

[9]  Nicklas Guldåker,et al.  Spatio-temporal patterns of intentional fires,social stress and socio-economic determinants: A case study of Malmö,Sweden , 2014 .

[10]  Akira Imada,et al.  A Literature Review: Forest Management with Neural Network and Artificial Intelligence , 2014 .

[11]  O. Viedma,et al.  Recent land-use and land-cover changes and its driving factors in a fire-prone area of southwestern Turkey. , 2017, Journal of environmental management.

[12]  Charles R. Jennings.  Socioeconomic Characteristics and Their Relationship to Fire Incidence: A Review of the Literature , 1999 .

[13]  Rajashree Kotharkar,et al.  Measuring Compact Urban Form: A Case of Nagpur City, India , 2014 .

[14]  Shiqiang Du,et al.  Analyzing explanatory factors of urban pluvial floods in Shanghai using geographically weighted regression , 2017, Stochastic Environmental Research and Risk Assessment.

[15]  Lara Wiesche,et al.  Time-dependent ambulance allocation considering data-driven empirically required coverage , 2015, Health care management science.

[16]  J. Pereira,et al.  Relationships between Human Population Density and Burned Area at Continental and Global Scales , 2013, PloS one.

[17]  Xiang Chen,et al.  Image Based Forest Fire Detection Using Dynamic Characteristics with Artificial Neural Networks , 2009, 2009 International Joint Conference on Artificial Intelligence.

[18]  M. Pereira,et al.  Negligent and intentional fires in Portugal: Spatial distribution characterization. , 2018, The Science of the total environment.

[19]  Jianguo Wu,et al.  Urban heat islands and landscape heterogeneity: linking spatiotemporal variations in surface temperatures to land-cover and socioeconomic patterns , 2009, Landscape Ecology.

[20]  Ronan A Lyons,et al.  Risk factors associated with unintentional house fire incidents, injuries and deaths in high-income countries: a systematic review , 2017, Injury Prevention.

[21]  J. Pereira,et al.  Modeling spatial patterns of fire occurrence in Mediterranean Europe using Multiple Regression and Random Forest , 2012 .

[22]  Arnab Jana,et al.  Resource allocation for handling emergencies considering dynamic variations and urban spaces: fire fighting in Mumbai , 2019, ICTD.

[23]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[24]  C. Purushothaman,et al.  Quantitative analysis of plastic debris on recreational beaches in Mumbai, India. , 2013, Marine pollution bulletin.

[25]  Onisimo Mutanga,et al.  High density biomass estimation for wetland vegetation using WorldView-2 imagery and random forest regression algorithm , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[26]  Catherine Linard,et al.  Geographical random forests: a spatial extension of the random forest algorithm to address spatial heterogeneity in remote sensing and population modelling , 2019, Geocarto International.

[27]  R. Searle,et al.  Socio-economic and demographic predictors of accidental dwelling fire rates , 2016 .

[28]  Jing Yao,et al.  Exploring Spatiotemporal Dynamics of Urban Fires: A Case of Nanjing, China , 2018, ISPRS Int. J. Geo Inf..