Development of Residential Building Stock and Population Databases and Modeling the Residential Occupancy Rate for Iran

Abstract

Seismic risk mitigation planning relies on the assessment of possible losses because of destructive scenario earthquakes. In turn, realistic risk assessment procedures must be supported by an adequate element at risk data sets. In this research, two major investigations in Iran examine (1) the development and augmentation of fine resolution population and building databases through data fusion and modeling, and (2) the derivation of the residential occupancy rate for a 24-h period for a normal working day in large cities. The basic population and building data sources are obtained from the national census program, which exhibits very limited spatial details. A fine resolution global dataset is fused together with the census data to gain much detailed spatial information. The residential occupancy modeling is completed according to the distribution of the residential population during normal working days. The empirical modeling was completed using two sets of temporal data for both census and traffic for Tehran, Iran. The traffic data sets were the result of 60,000 survey forms, which were filled out by residents. To apply the process for different large cities in Iran, semiempirical population modeling is proposed. The findings are compared with a global model and show that for time ranges from early morning until 10:00 a.m. and also from 10:00 p.m. until midnight, the relative differences are noticeable and mainly above 22%. Also, from 10:00 a.m. until 10:00 p.m., all models match reasonably well.