A New Causality Model for Evaluating the Probability of Human Damage From Injury To Death in Collapsed Building

Abstract eng:
A recent challenge for seismic damage evaluation, which forms the basis of risk management for reducing the death toll in damaged buildings, is expanding the technical capabilities from rough estimation of regional damage in municipal units to minute damage estimation for individual buildings. With this in mind, we proposed the related numerical functions with structural parameter of the load-carrying capacity for buildings in 13th World Conference on Earthquake Engineering (WCEE) of 2004. In this paper, we discuss the method of estimating human casualties in damaged buildings by expanding these previous studies. As a method for estimating human casualties in various structural building types throughout the world, a universal model is set up with a causal scenario as follows: In the first stage of damage scenario, structural elements of buildings are damaged in accordance with the severity of ground motion. The degree of structural damage in an individual building can be quantitatively estimated on the damage scale from 0.0 (No Damage) to 1.0 (Total Collapse) by a Damage Index Function defined by the authors. In the second stage, the surviving interior space of building is reduced in proportion to the damage severity index and characterized by damage pattern of building. As for the deterioration of space, one of the authors defined the deterioration of surviving interior space as the volumetric reduction of a collapsed building in the 11th WCEE of 1996, and in the 16th WCEE an advanced numerical function for this volumetric reduction, defined by the Weibull probability distribution, is discussed in a different paper. In the third and final stage of the scenario, inhabitants trapped within the debris of the damaged building sustain serious injury corresponding to the volumetric reduction. In order to handle human casualties quantitatively in modeling, we introduce the Injury Severity Score (abbreviated as ISS), which is devised to classify earthquake victims with multiple trauma. Using various data on damage that occurred in the 1995 Kobe Earthquake, we compiled a statistical database of wooden frame building damage scaled by the Damage Index, and of human casualties ranked by severity according to the ISS and using GIS technology to match a map of Kobe with building lot numbers. By applying Bayesian methods to the database, we successfully functionalized the stochastic relation between the building damage index and the ISS for each of casualties. As a case study based on the Nankai Trough Quake, we calculated the assumed casualties at 0.5% of the population in Kochi Prefecture, Japan (about 731,000), only due to collapse of buildings and without taking into account tsunamis. These assumed victims break down as 2,386 people lightly injured (ISS scale range of 1 to 8), 603 people heavily injured (ISS scale range of 9 to 15), 535 people in critical condition, and 369 people killed instantly. The information obtained by this method is useful for indicating the amount of supply and demand on seismic countermeasures in each region.

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