This paper presents a case study of a bridge failure during construction in an infrastructure development project in Nakhon Si Thammarat, Thailand. This highlights the critical role of thorough inspection and early detection of structural failures. The failure, related to foundationrelated issues, shows how even minor errors in the construction sequence can lead to significant structural problems. Key steps for evaluating foundation failures are assessing the structure's movement and utilizing an inclinometer to determine ongoing failure. Resistivity surveys, in conjunction with screw driving sounding tests (SDS), were performed to assess soil properties, while the finite element method (FEM) was applied to validate the observed failure behavior. The results show that an inclinometer effectively monitored these structures' movement. The resistivity surveys proved useful in identifying soil layers in the wide area. Meanwhile, the SDS tests were able to determine the soil's undrained shear strength. FEM simulations provided valuable insights into the behavior of underground structures. Consequently, comprehensive inspections are essential to mitigate foundation failure risks and ensure critical structures' safety and longevity.

The pile-retaining wall at Nonthaburi rural road no. 5036 was constructed using reinforced concrete piles or driven piles combined with a concrete retaining wall. The purpose of this structure was to enhance the slope stability of the canal-side road (road embankment along the canal). The damage to the driven piles occurred during the pile construction at 18 m depth below the ground surface. The resistivity survey and screw driving sounding test were employed to investigate the thickness of soft clay layers and unexpected stiff soil layers at the failure area. The field vane shear test was employed to investigate the sensitivity of the soft clay layer. Furthermore, the finite element model was analyzed to verify the failure behaviour of the road embankment during the driven pile's construction. Consequently, the investigation revealed that the subsoil in the failure area exhibited sensitivity values. The subsoil consisted of a layer of soft clay to medium stiff clay, ranging from 2-10 m below the ground surface, while the subsoil consisted of stiff clay below a depth of 10 m. The installation of the 18-m driven pile caused a disturbance in the soft sensitive clay layer above the stiff soil layer, resulting in a reduction in the strength of the soft clay and affecting the displacement of the driven pile during construction. Furthermore, the occurrence of rapid drawdown causes water seepage to continue to flow toward the canal side. This phenomenon produces active forces on the slope of the road embankment along the canal. As a result, the road embankment along the canal side can collapse due to a disturbance in the sensitive clay layer with rapid drawdown. The result was agreed with the study findings obtained by the finite element model.

Machine learning (ML) algorithms are increasingly applied to structure health monitoring (SHM) problems. However, their application to pile damage detection (PDD) is hindered by the complexity of the problem. A novel multi-sensor pile damage detection (MSPDD) method is proposed in this paper to extend the application of ML algorithms in the automatic identification of PDD. The time-series signals collected by multiple sensors during the pile integrity test are first processed by the traveling wave decomposition (TWD) theory and are then input into a hybrid one-dimensional (1D) convolutional and recurrent neural network. The hybrid neural network can achieve the automatic multi-task identification of pile damage detection based on the time series of MSPDD results. Finally, the analytical solution-based sample set is utilized to evaluate the performance of the proposed hybrid model. The outputs of the multi-task learning framework can provide a detailed description of the actual pile quality and provide strong support for the classification of pile quality as well.