Development of Estimation Method for Building Structure using Open Data and Statistics

In recent years, the Ministry of Land, Infrastructure, Transport and Tourism of Japan has released 3D models of cities as open data in an initiative called “PLATEAU.” These data are expected to be used for disaster prevention planning and city planning. When conducting disaster prevention planning and urban planning in preparation for earthquakes, it is important to have information on the structure of buildings to determine the earthquake resistance of each building. However, it is often difficult to obtain information on the structure of each building, and there is no information about the building structure for almost all buildings in the PLATEAU initiative. Therefore, in this study, teacher data were created based on open data for the Tokyo metropolitan government, and the structure of each building was estimated through machine learning by creating features based on existing statistics, such as the national census.