Use of machine learning algorithms for damage estimation of reinforced concrete buildings

Identifying the vulnerabilities in a building is a crucial step towards earthquake risk mitigation. Rapid visual screening is a quick and popular method for seismic vulnerability assessment. It helps identify buildings that require detailed investigation, which is done by modelling using seismic analysis software. This is a time-consuming and resource-intensive task. This article proposes the use of machine learning to bypass the seismic analysis of buildings. A case study using 1296 building models and maximum inter-storey drift ratio as the measure of damage has been presented. Random forest gives the best prediction accuracy in the study.