Construction on deep soft clayey ground often experiences significant consolidation settlement. Preloading accelerates this process and is typically monitored using settlement plates (SPs). However, traditional methods lack spatial coverage, failing to capture localized settlement variations. This study proposes a novel settlement prediction method for soft ground construction sites using Drone-LiDAR technology. Data were collected over 18 months from a site in Busan Newport, comprising 33 data points. After removing outliers and extracting ground elevations, an optimal grid size of 50 cm × 50 cm was used to construct time-series ground elevation models. The method predicts settlements, distinguishing between increases due to filling and decreases caused by settlement or excavation. Validation against 21 SPs yields a root mean squared error and mean absolute error of less than 0.22 m and 0.19 m, respectively. The method is also applied to estimate fill and excavation volumes and assess the degree of consolidation, providing spatially detailed insights into construction progress. This study demonstrates the potential of Drone-LiDAR for accurate settlement prediction and effective construction management.
Lee et al. (Sun,) studied this question.