A modern survey method for determining extraordinary loads based on multi-source online data

Abstract

Induced by unusual events such as people gatherings or furniture stacking, extraordinary loads rarely occur but can exhibit significant amplitudes. Though, sufficient statistical data are required to quantify the stochastic behavior of such loads, traditional surveys typically depend on manual on-site questionnaires, leading to low efficiency, occupant resistance and problematic weight estimation. In this study, multi-source online data are therefore used to propose a new survey method. Object weights are obtained from commodity information on e-commerce websites. Object quantities and occurrence intervals are acquired from questionnaires whose distribution employs an online user database. Using the proposed method, an online extraordinary load survey is conducted on residential buildings, covering 43,941 ${\mathrm{m}}^2$ of floor area and collecting 14,186 weight samples. Three types of extraordinary loads are considered and their statistical properties of amplitudes and occurrence intervals are determined. The survey results indicate that the variations of amplitude statistics with area are well fitted by a form of the power function. On average, furniture stacking events are found to occur 0.07 time per year, but are most likely to exhibit a considerable amplitude. Finally, the acquired data are utilized to determine the fractile values of the lifetime maximum extraordinary loads.