• Journals

Predicting Collapse Potential in Gypseous Soils: A Hybrid Approach Combining Random Forest Modeling and Uncertainty Quantification

The present research deals with the important problem of landslide forecast for collapsible gypseous soils. The current traditional methods of a deterministic nature often lack an account of the uncertainties intrinsic in the properties of gypseous soils and, hence, are unreliable. In this work, a new hybrid approach that incorporates random forest modeling and Monte Carlo simulations was introduced. To take into account the uncertainties in those parameters, it is quite an effective method for catching the nonlinear relationship between key soil parameters such as compression index, initial void ratio, dry unit weight, gypsum content, and initial soil saturation with collapse potential. A comprehensive experimental framework involving single, double, and modified collapse tests is used to study the collapse behavior and to validate model predictions. These results have shown that the proposed approach improves the prediction accuracy and provides a more probabilistic assessment of the risk of collapse in gypseous soils. This study has placed a strong emphasis on how uncertainty quantification is a must to develop better foundation design that addresses the safety and durability of structures within these hostile gypseous soil environments.

• Journals

Experimental Investigation of the Effects of Cement Treatments on the Swelling Potential of Clayey Soils

Expansive soils challenge various design aspects of many projects around the world, including slabs, hardscapes, road construction, etc. One of the common mitigation techniques regarding the expansiveness of soil is mixing cement into the soil with varying quantities depending on the expansion potential of the soil. This project focused on systematic laboratory testing of cement treated clays in southern California. Three samples were taken from different locations and tested in the lab with different cement mixtures. Each of the samples was tested for expansion index (EI) with varying amounts of cement, and the expansion mitigation was quantified. The purpose of the project was to better understand the optimum and most economical amount of cement, where the expansion potential decreases significantly. The laboratory testing demonstrated that each material has a specific amount that yields a significant decrease in the EI. The results are presented in charts, and comparisons were conducted for various mixtures. A practical, two-variable equation was also developed as a guide for the design engineers to predict the EI improvement as a function of cement content and the plasticity index of the clay soil.

• Journals

Estimating Preconsolidation Pressure: Comparative Analysis of Multiple Linear Regression and Neural Network Models

The pre-consolidation pressure (P_c) is one of the most important geotechnical properties, determining the stress history of fine-grained sediments and governing their future behavior. This paper presents a comparative analytical study that develops predictive models using artificial neural network (ANN) and multiple linear regression (MLR) techniques for the determination of the P_c of fine-grained sediments based on soil index variables. Soil data were gathered from the eastern part of the Basra region, southern Iraq, as a case study. A dataset consisting of 110 samples includes P_c as the dependent variable and independent variables consisting of liquid limit, water content, overburden pressure, and void ratio. Both models were analyzed using Python in this work. The best correlations in the study of MLR and ANNs were identified by using three independent variables for each model. The ANN models have been trained and compared. An accuracy test revealed that the ANN model performed more effectively. Both models are valuable tools for determining P_c, while ANN performs well in predicting complex soil behavior.

• Journals

Thermal Performance Analysis of Energy Piles With W-Shaped Heat Exchange Pipe

To enhance the efficiency of geothermal energy utilization and optimize building usage, this study proposes a heat-storage energy pile system that integrates a ground-source heat pump with a building pile foundation. A simulation model of a W-type heat exchange pipe energy pile is developed, and the impacts of inlet water temperature, circulating fluid flow rate, heat exchange pipe diameter, shape of pile group, and spacing of piles on the heat exchange performance of the energy pile are investigated. The results indicate that the inlet water temperature, fluid flow rate, and pile body temperature exhibit a positive correlation; however, the growth rate of the pile body temperature gradually decelerates as the flow rate increases. Meanwhile, an increase in the heat-exchange pipe diameter reduces the growth rate of the heat-exchange capacity. Regarding pile group arrangement, the triangular arrangement results in less heat loss compared to the square arrangement. During the heat-storage stage, heat continuously diffuses into the pile bodies and the surrounding soil, and the thermal storage capacity of the system increases gradually. During the heat-extraction stage, the temperature of the pile bodies initially decreases, and there is a lag in the thermal response of the surrounding soil. Eventually, the system enters a stable heat-exchange state. In the heat extraction stage, the temperature of the pile body decreases first, and the response of the peripheral soil lags. Moreover, the system finally enters the steady state heat exchange state. The findings of this study can offer a theoretical foundation for the optimized design of an energy pile heat-storage system.

• Journals

Impact of Fine Soil on Soybean Calcite Precipitation: A Multifaceted Analysis of Mechanical and Chemical Interaction in Sandy Soil Stabilization

Enzyme-induced calcite precipitation (EICP) using urease extracted from soybeans has been proven to enhance the strength of sandy soil. However, most studies on the application of this method have used pure sandy soil without fine soil. Previous evaluations have indicated that fine soil particles can influence the rate of calcium carbonate production. This study evaluated the use of 20% fine soil particles in sandy soil on the effectiveness of the soybean calcite precipitation method, using parameters such as unconfined compressive strength (UCS), calcium carbonate content, and microstructural analysis based on the mechanical and chemical aspects of the soil. The results show that the addition of 20% fine soil to sandy soil reduced the UCS by 70%, from 122 kPa to 37 kPa, and decreased the calcium carbonate content by 24%, from approximately 4.1% to 3.1%, compared to pure sandy soil at the same density. Furthermore, microstructural analysis revealed that calcium carbonate formed interparticle contacts in sandy soil, but these were absent in sandy soil containing fine soil. Soil density affects the increase in soil strength, but reducing the fine soil content and using additive materials does not have a significant impact. Fine soil at a concentration of 10 g/L also reduced the calcium ion concentration by 57 mg/L. For improvement, a reagent concentration of 2.5 mol/L yielded the best results, increasing the UCS from 37 kPa to 52 kPa and the calcium carbonate content from 3% to 4%, corresponding to improvements of approximately 41% and 29%, respectively. However, the use of excessively high reagent concentrations can disrupt the precipitation reaction equilibrium and lower the solution pH.

• Journals

Experimental Investigation of the Impact of Tire-Derived Aggregate Sizes on Tunnel Behavior Under Seismic Loadings

The present study aims to examine the impact of tire-derived aggregate (TDA) size on the response of tunnels subjected to seismic loads. In this work, a modified tunnel-sand-pile interaction model is used to assess the bending moments at the tunnel crown and invert under seismic excitations resulting from the Kobe and Loma Prieta earthquakes. Two different TDA sizes, 25 × 25 mm and 50 × 50 mm, were analyzed. The results show a substantial influence of the size of the TDA on the observed bending moments. The 50 × 50 mm TDA consistently produced a greater bending moment than did the 25 × 25 mm TDA. This may happen due to the larger voids of 50 × 50 mm size with less compact arrangement to cause a reduction in stiffness. The 25 × 25 mm TDA has smaller voids and dense packing due to its smaller size. This resulted in higher stiffness and lower bending moments. These findings indicate that smaller sizes of TDA may offer better seismic force resistance by the tunnel to enhance overall structural stability. This research suggests the need for future study by using different TDA sizes with additional factors, such as overburden pressure and liquefaction, to optimize the use of TDA in seismic design.

• Journals

Relationship Between Strength and Durability of Quartzite Ballasts from Nepal Himalaya

This study evaluates the petrographic, mechanical, and durability properties of quartzite ballasts from the Fagfog, Dunga, Pandrang, and Chisapani Quartzite units, focusing on their suitability for use in construction and railway applications. These quartzite units, predominantly monomictic and composed mainly of quartz, exhibit high hardness, low porosity, and excellent resistance to weathering, key attributes for load-bearing and stability applications. The petrographic analysis indicates that the Fagfog and Pandrang Quartzites are high-grade metamorphic rocks characterized by coarse, recrystallized quartz with granoblastic textures, whereas the Dunga and the Chisapani Quartzites display lower metamorphic grades with finer grains. These textural and mineralogical differences directly influence their strength and durability characteristics. Mechanical testing revealed significant variability in the Point Load Strength Index (PLSI) across the samples, ranging from medium to extremely strong, with PLSI values showing strong correlations with other durability indicators, such as the Los Angeles Abrasion Value (LAAV) and Aggregate Impact Value (AIV). Higher-strength quartzites, characterized by tightly interlocked grains and minimal secondary minerals, exhibited superior abrasion resistance (a negative correlation between PLSI and LAAV, r = -0.70). The study also identified a significant positive correlation between AIV and LAAV (r = 0.71), suggesting that samples more susceptible to impact are also prone to abrasion. The Slake Durability Index results highlight the high resilience of these quartzites across multiple cycles, with the Fagfog and the Dunga Quartzite units showing minimal reduction in durability through cyclic wetting-drying tests. The comprehensive analysis underlines the importance of site-specific evaluations in selecting quartzites for engineering applications, given the substantial variability within each formation. Understanding the intricate relationships between petrographic features, mechanical properties, and durability is critical for optimizing the use of quartzite as ballast in demanding environmental conditions.

• Journals

Strength Estimation of Cement and Bentonite Stabilized Soft Soil

The paper is going to develop a proposed prediction model for the unconfined compression strength (UCS) of cement and bentonite stabilized soft soil based on gene-expression programming (GEP). The model evaluates the effect of three independent variables, including cement content, bentonite content, and curing time, on the UCS of the stabilized clay. The selected optimal GEP model shows its accurate prediction ability with a high correlation coefficient (R = 0.992), and low errors (Root Mean Squared Error and Mean Squared Error). The proposed prediction equation formulated from the selected GEP model is ready to be applied in practice in predicting the UCS value of cement and bentonite-stabilized soil material. The sensitivity analysis shows that the strength of stabilized clay could be significantly enhanced by increasing the cement content. In addition, the parametric analyses indicate that the bentonite content has a negative effect on the UCS of the stabilized soil. With the high bentonite/cement ratio, the strength of stabilized clay decreases significantly. The research findings could help engineers choose suitable cement and bentonite content to optimize the UCS of stabilized clay.

• Journals

Establishing a Regional Landslide Early Warning System Based on Rainfall and Soil Moisture Measurement

Landslide-prone areas are often located in mountainous rural regions where residents come from various backgrounds but maintain strong social networks. For these communities, a simple, community-based Landslide Early Warning System (LEWS) that is easy for everyone to understand is ideal. This study introduces a novel approach to regional-LEWS by using rainfall and soil moisture as the triggering factors. Statistical analysis based on historical landslide events was used to evaluate the performance of this threshold. The results show that the area under the curve (AUC) for rainfall-only thresholds was 0.862, while the combined rainfall and soil moisture thresholds achieved a higher AUC of 0.90, demonstrating improved predictive performance. Two threshold lines were established in this study: a lower threshold to distinguish landslide from non-landslide events and an upper threshold using five exceedance probabilities (10%, 20%, 30%, 40%, and 50%) to refine the prediction. The lower threshold achieved a hit rate (HR) of 1, an accuracy of 0.6, and an Euclidean distance (Ed) of 0.4. The best results for the upper threshold were observed at a 40% exceedance probability, which achieved an accuracy of 93%, a HR of 0.591, and Ed of 0.415. These thresholds classified the area into three levels, which were then integrated with a landslide susceptibility map using logical operators to develop five warning levels: null, attention, watch, alert, and emergency. This combined system was tested using landslide inventory data, yielding promising results. The system successfully triggered warning levels of "Watch," "Alert," or "Emergency" for approximately 80% of the landslide inventories. This indicates that the model proposed in this study has strong performance and could be applied in other mountainous regions.

• Journals

Assessment of Geotechnical and Geoenvironmental Properties of an Expansive Soil Treated with Red Mud

To identify the potential reuse of industrial byproducts in order to achieve an environmentally sustainable society, this study deals with the utilization of red mud (RM) for stabilization of expansive soil. This research explores the practicality of treating expansive soil by considering different percentages of RM addition to the expansive soil. The characteristic studies on RM revealed that it possesses no harmful elements, establishing the way for utilization in stabilization applications. The experimental studies conducted on stabilized soils of varying percentages of RM showed improvement in the index and engineering properties of expansive soil. A detailed microstructural analysis using scanning electron microscopy and toxicity studies using leaching tests established that the expansive soil incorporating RM falls well within the standards, assuring less chances of problems associated with environmental concerns. The engineering properties showed a noticeable improvement of 138% in compressive strength, a 225% increase in the soaked California Bearing Ratio values and a reduction in the permeability values after 28 days of curing. The research findings establish the viability of using RM up to 12% for the stabilization of expansive soil. The use of RM in soil stabilization applications provides a sustainable method to reduce the deposition of RM in enormous quantities, thereby fostering eco-friendly and economically viable solutions in reducing waste disposal.