Mr Ratiranjan Jena1, Prof Biswajeet Pradhan1, Prof Ghassan Beudoun1
1Centre for Advanced Modelling and Geospatial Information Systems (CAMGIS), Faculty of Engineering and IT, University of Technology Sydney, Sydney, Australia
The vulnerability assessment of earthquake in urban or regional scale comprises of a principal component in mitigating the structural vulnerability due to earthquakes. Challenges exist in the delineation of earthquake fragilities of buildings exposed to the hazard. However, vulnerability or fragility for structural elements can be derived using analytical, expert opinion based methods or empirical methods provide the loss ratio probability, conditionally on a particular level of measured intensity. Therefore, this study proposes a neural network approach for the assessment of structural vulnerability in Banda Aceh city by considering and analysing the factors of building density, size of buildings, building area and building materials. There are some uncertainties that are involved in vulnerability assessment and especially the accurate assessment of earthquake vulnerability is important to obtain the bounds of the expected losses, which has been proved through the neural network methodologies. Our results show that the proposed method is able to estimate the vulnerability of earthquake in Banda Aceh city accurately with an accuracy of 85%. The validation of the resulted structural vulnerability map was conducted using the earthquake events and previously published inventory map.
Biography:
Mr. Ratirankan Jena is a {PhD student at the Centre for Advanced Modelling and Geospatial Information Systems (CAMGIS), Faculty of Engineering and IT, University of Technology Sydney, Australia. His research focuses on the uses of geospatial techniques for mapping and monitoring environmental changes and natural disasters.