Abstract
Spatial Autoregressive Modeling of Predicting Property Damaged in Fire Events
Track: Emergency Medical Services/Fire
Author(s): Alireza Ghaffari, Ali Asgary
Fire incidents are still a major risk in urban areas especially large cities. Understanding the spatial patterns of fire risks and factors that contribute to their impacts could improve fire response and therefore reduce human injuries, fatalities and property losses. This is even more important and critical in large urban areas such as Toronto where density of population and activities are very high.
This study aims to develop models that can predict property damaged in Toronto. The results of such models would improve fire response management systems.
More than 20,000 fire incidents in Toronto for the period of 2001 and 2006 provided by Ontario Office of Fire Marshall have been geocoded and used in this study.
Several spatial autoregressive models examine the relation between dependent variable (property damaged) and independent variables or predictors (e.g. distance to fire stations, responding personnel, type of building,…) and the results show that spatial autoregressive models are able to better predict fire risks in Toronto.
Alireza Ghaffari
York University
83 Formosa Drive
Richmond Hill , Ontario L4S 1T1
Canada
Phone: 905 737 6800
E-mail: aghaffar@yorku.ca
Ali Asgary
York University
4700 Keele Street
Toronto , Ontario M3J 1P3
Canada
Phone: 416 736 2100
E-mail: asgary@yorku.ca