Collapsible soils are moisture-sensitive soils that may experience sudden and significant settlements upon wetting, posing serious challenges in engineering projects. Conventional laboratory tests, such as the oedometer test, generally simulate soil wetting by full submergence and are therefore unable to represent different real-world water infiltration scenarios. In this study, a new experimental apparatus was designed and constructed to simulate four typical water infiltration patterns based on the direction of water movement (top-down and bottom-up) and the mode of water distribution (point and wide). Two laboratory-prepared collapsible soils with different collapse potentials were tested using single and double oedometer tests as well as the proposed apparatus. The results demonstrate that water infiltration patterns have a significant influence on collapse behavior. The highest collapse potential was observed under the top-point infiltration pattern, representing localized surface water sources such as irrigation, whereas the lowest collapse potential occurred under the bottom-point infiltration pattern, simulating leakage from buried pipes. For the tested soils, collapse potential induced by bottom-point infiltration was up to 35% lower than that measured by the conventional oedometer test. The findings indicate that the direction of water movement has a greater effect on collapse potential than the water distribution mode. The proposed apparatus provides a more realistic assessment of collapsible soil behavior under different infiltration conditions and highlights the limitations of conventional testing methods in predicting field settlements.

The combined piled-raft foundation (CPRF) is a cost-effective system for tall buildings, but its design remains computationally demanding. Although several closed-form approximate methods have been proposed, their reliability-especially for pre-design-has not been fully established. This study develops an analytical equation to delineate the validity ranges of common approximate CPRF methods by integrating three-dimensional finite element simulations (Abaqus) with principal component analysis (PCA) and weakly unsupervised learning to reduce the high-dimensional input space to a three-dimensional principal component domain. A total of 700 CPRF models were examined for each method-including the Poulos-David-Randolph (PDR) method, equivalent pier (EP) method, equivalent raft (ER) method, and full 3D FEM-covering 20 geometric variables such as pile length, diameter, spacing, and raft dimensions. The comparative results enable a quantitative evaluation of the accuracy of each approximate solution and support the formulation of a practical CPRF assessment tool. The primary contribution of this study is the development of a PCA-based validity-domain framework that quantitatively identifies the reliable application ranges of commonly used approximate methods for combined piled raft foundations. Rather than proposing a new predictive model, the proposed framework provides a practical pre-design assessment tool that enables engineers to evaluate whether simplified analytical or data-driven approaches can be trusted for a given foundation configuration.

We evaluate the performance of the pseudo-static method by comparing 3D nonlinear finite element (FE) analyses of the pseudo-static approach and the dynamic simulation. The pseudo-static approach accounts only for the inertial interaction, neglecting kinematic interaction as well as the coupled axial and lateral response at the soil-pile interface. The 3D dynamic FE model is validated against centrifuge measurements, demonstrating that the employed nonlinear soil constitutive model accurately captures the recorded response. The p-y curve extracted from the dynamic analysis deviates significantly from that obtained in the pseudo-static analyses, resulting in a stiffer overall response. While using the p-y curve from the 3D pseudo-static FE model provides a better fit with the dynamic analysis results compared to the API curve, a 10-30% difference in peak bending moment remains. This study reveals that applying an amplification factor to stiffen the pseudo-static p-y curve improves prediction accuracy.

The life prediction of hob cutters in tunnel boring machines (TBMs) during rock breaking has been a key issue affecting TBM boring efficiency and the timing of tool changes. By defining the service life of cutters during rock intrusion and introducing the dense core effect, an analytical method for the rock breaking force of disc cutters under the comprehensive failure mode of compression and shear was established. In addition, our work has improved Rabinowicz's abrasive wear model for predicting the wear rate and lifespan of TBM disc cutters. The engineering applicability and high precision of the prediction model were verified by comparing the predicted results with theoretical calculation results, numerical simulation results, and field monitoring data. The results indicate that the predicted values of the wear rate and lifespan of the hob cutter from the prediction models agree well with the measured values from the field. Compared with the existing models, the established model improves the prediction accuracy of the lifespan values by 7.63%, 64.15%, and 61.13%, respectively. This demonstrates the necessity of considering the dense core effect and the compression and shear failure mode in TBM rock breaking calculations. The evaluation method presented in this work can provide theoretical guidance for the design of timing and tool change plans for TBM tunnels and other similar projects.

Soil liquefaction is a destructive phenomenon resulting from earthquakes that occur in saturated loose soil. When structures are built on such soil, liquefaction can lead to significant damage, including loss of life. Therefore, understanding the behavior of structures on liquefiable soil is crucial. This study investigated the interaction of two adjacent structures on liquefiable soil using a three-dimensional finite difference method. The structures have been modeled using beam and shell elements and the soil behavior has been simulated using the Dafalias-Manzari model for stress-strain behavior. Parameters such as the distance between structures, groundwater level, number of stories, and type of seismic excitation (near-field or far-field) have been examined. The results indicate that, when the distance between the structures was equal to or exceeded their width, there was no effect on the excess pore pressure coefficient or settlement. Additionally, a groundwater level of 0.67 times the height of the structure was found to nullify the effects of settlement caused by liquefaction. An increase in the number of stories decreased the liquefaction potential but increased settlement. Far-field earthquakes were found to cause less damage than near-field ones and resulted in less settlement and liquefaction.

This research presents a new empirical rock mass classification system for sedimentary weak rocks, such as marlstones, taking into account the extent of weathering within the rock mass. For this purpose, 45 marlstone samples were systematically collected from the South Pars region in southwest Iran. The study involved measuring the geoengineering characteristics of the marls and assessing the degree of weathering they had undergone. The collected data was used to create variation weathering charts that show changes in rock mass durability and geoengineering properties with different degrees of weathering. These new charts are valuable tools for evaluating the condition of marlstone rock masses, providing a clearer understanding of how weathering affects their stability and durability. The empirical classification system introduced in this study significantly advances the assessment and management of marlstone formations in geotechnical engineering and related fields.