Probabilistic Evaluation of Retrofitting Priority of HousesUsing a Model of Evacuation Route Selection under Road Blockade

Abstract

We propose a new method to evaluate quantitatively the effectiveness of retrofitting of houses for evacuation. A model of evacuation route selection under road blockade is constructed based on a subject experiment and built it into an agent model for evacuation simulation. The priority of houses that should be retrofitted is probabilistically evaluated. Because evacuation behavior is considered, this method can appropriately identify houses that should be retrofitted. An index is newly introduced to evaluate retrofitting priority of houses quantitatively using improvement ratio of number of evacuees in a unit time. The proposed method can show retrofitting priority of houses effectively when the ground motion intensity is assumed as JMA Seismic Intensity 6 upper.