Search for: All records

Soil mixing is a ground improvement method that consists of mixing cementitious binders with soil in-situ to create soilcrete. A key parameter in the design and construction of this method is the Unconfined Compressive Strength (UCS) of the soilcrete after a given curing time. This paper explores the intersection of Machine Learning (ML) with geotechnical engineering and soilcrete applications. A database of soilcrete UCS and site/soil/means/methods metadata is compiled from recent projects in the western United States and leveraged to explore UCS prediction with the eXtreme Gradient Boosting (XGBoost) ML algorithm which resulted in a ML model with a R2 value of 88%. To achieve insights from the ML model, the Explainable ML model SHapley Additive exPlanations (SHAP) was then applied to the XGBoost model to explain variable importances and influences for the final UCS prediction value. From this ML application, a blueprint of how to scaffold, feature engineer, and prepare soilcrete data for ML is showcased. Furthermore, the insights obtained from the SHAP model can be further pursued in traditional geotechnical research approaches to expand soil mixing knowledge. 

The field of biogeotechnics has emerged from the realization that processes intrinsic to natural systems can provide new approaches and inspiration through which the efficiency, sustainability, and functionality of geotechnical systems can be improved. Of these processes, microbially induced calcite precipitation (MICP) has advanced the most rapidly with the use of ureolytic microbial activity providing an opportunity to control the precipitation of calcium carbonate minerals throughout a soil matrix, thereby significantly improving soil engineering behaviors. The process affords increases in soil stiffness, strength, and dilatancy, with utility across a breadth of geotechnical and geoenvironmental applications, including mitigation of earthquake- induced soil liquefaction. This state of the art paper first covers: (1) enabling scientific processes, (2) treatment methods, and (3) monitoring techniques, which are broadly useful for different engineering applications. The second part focuses on how MICP can: (1) improve engineering behaviors at the element scale, (2) be modeled at the particle- and continuum-scales, (3) be applied at the field-scale, and (4) improve the resistance to liquefaction triggering and reduce the consequences when it does occur. 

Calibrations of the PM4Silt constitutive model are presented for two low-plasticity fine-grained soils that exhibit significantly different cyclic loading be-haviors. The PM4Silt model is a stress-ratio controlled, critical state compatible, bounding surface plasticity model that was recently developed for representing low-plasticity silts and clays in geotechnical earthquake engineering applications. The low-plasticity clayey silt and silty clay examined herein were reconstituted mixtures of silica silt and kaolin with plasticity indices (PIs) of 6 and 20. Un-drained monotonic and undrained cyclic direct simple shear (DSS) tests were per-formed on normally consolidated, slurry deposited specimens. Calibration of the PM4Silt model was based on the monotonic and cyclic DSS test data, plus em-pirical relationships for strain-dependent secant shear moduli and equivalent damping ratios. The calibration process and performance of the PM4Silt constitu-tive model are described for each soil. The results illustrate that PM4Silt is capa-ble of reasonably approximating a range of monotonic and cyclic loading behav-iors important to many earthquake engineering applications and is relatively easy to calibrate. 

Engineering centre research faculty and staff value the importance of performing educational outreach and mentoring graduate students. However, these activities are often less structured than research projects, which leads to variable and less effective results. The geotechnical group at the University of California, Davis (UC Davis), which includes research faculty and staff at the Center for Geotechnical Modeling and the Center for Bio-mediated and Bio-inspired Geotechnics, developed a Ladder Mentoring Model (LMM) for mentoring graduate students in academic environments to enrich graduate student development while minimizing additional demands on centre personnel. The LMM is a combination of several existing mentoring models and relies on six core principles where the outcome is students receiving guidance from a variety of mentors with different areas and levels of expertise or experience. This paper provides a brief overview of the UC Davis LMM and describes how it is integrated into three critical areas of graduate student development: technical training, professional skills, and educational outreach.