Fire Following Earthquake Modeling, Probabilistic Ignition Model for Building Stock

Abstract eng:
Fire following earthquake (FFE), as an indirect seismic hazard involves three main phases: ignition, spread and suppression. Records of historical FFE’s shows higher significance for ignitions taken place inside buildings. This paper therefore, focuses on ignition following earthquake (IFE) and introduces a probabilistic algorithm for intra-structure ignition modelling. The occurrence of intra-structure ignitions depends on several parameters such as structural damages, non-structural damages, ground and structural PGA, overturning critical acceleration for building components and equipments, buildings occupancy, building height, building area and earthquake time. A GIS-based computer program has been designed and developed in this work which models probability of ignitions for each building using synthetic earthquake scenarios. Logic tree and Monte Carlo simulation processes are used for combining uncertainties associated with different components controlling intra-structure ignition such as earthquake parameters and building behaviour. The proposed model is based on an analytical methodology which takes into account many effective parameters controlling IFE. The main objective of this study is to develop a methodology to probabilistically convolute different uncertainties associated with factors controlling intra-structure IFE. The approach is introduced as an alternative solution to the statistical ignition model currently being used in many FFE hazard models. However, detailed studies towards quantification and calibration of probability functions used in this study are beyond the scope of this paper and require further statistical data and investigations. The proposed model and the developed computer tool are used to model IFE for a city district in northern Tehran. Ignition probabilities obtained from this model are compared against those estimated by the HAZUS ignition model.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket