Abstract The mechanical behavior of rock masses in longwall mining is critically influenced by spatial correlation among material properties, yet conventional deterministic models often overlook this variability. Conventional deterministic models often overlook this spatial variability, leading to potentially misleading assessments of rock strength and stability. This research addresses the critical need for a nuanced understanding of rock mass behavior by investigating the effects of spatial correlation on stress distribution and failure mechanisms in coal seams. The primary objective is to evaluate how incorporating spatially correlated random properties can enhance the accuracy of predictions in mining operations. The study uses a three-dimensional numerical model to contrast deterministic approaches with a stochastic framework that integrates spatial correlation factors. The methodology involves generating a realistic random field database based on the Extreme Value stochastic model, which is then applied to simulate stress responses in the rock mass under various loading conditions. This 40% reduction in peak stress estimates translates to substantially different safety assessments and mining strategy recommendations compared to traditional deterministic approaches. This research underscores the necessity of adopting stochastic approaches in rock mass evaluations, as they provide a more accurate representation of real-world conditions. The insights gained from this study are essential for developing safer and more effective longwall mining strategies, highlighting the importance of considering spatial variability in rock mechanics. The findings contribute to the advancement of mining engineering by integrating advanced statistical techniques with practical applications, ultimately enhancing operational efficiency and safety in mining practices.

The objective of this study is to accurately evaluate pile bearing capacity without the use of sophisticated finite element software or wave recording by utilizing artificial neural networks and public pile driving analysis (PDA) test data. The data from 46 concrete piles in Bandar Imam and 65 steel piles in Asaluyeh will be examined using reliable tests. This article used PDA test data and a system made up of two kinds of artificial neural networks to achieve acceptable results for the pile capacity after introducing previously established relationships. In this system, the pile capacity for a chosen number of piles was finally determined by successfully combining a number of neural networks. The suggested system extends earlier neural networks for a comparable function and uses a self-organizing neural network for a set of data in training, testing, and valid datasets. Model predictability increased by 80% when parameters like hammer drop height were added. Cross-sectional area and pile length were additional improvements, and a normalized stiffness parameter provided marginal gains. However, better results were obtained when length and area were used independently. In the end, predictions were greatly improved by introducing a Bearing Capacity Index (BI) from classical equations. The suggested model is built on this index and PDA data, which provides the best prediction accuracy and efficiency for figuring out pile load-bearing capacity.

The natural silicate minerals with a ring structure were observed on a tuff breccia outcrop in Okinawa, Japan, appearing to self-repair cracks. Generally, silicate minerals are difficult to dissolve in water and their mechanical strength is relatively higher than that of other natural precipitated minerals, such as carbonate minerals. Therefore, if this natural phenomenon is induced by the microbial activities and can be artificially replicated, it could offer a self-organized geo-improvement technology characterized by insolubility and durability. In this study, the authors determined that this mineral has 26–50% rock mass quality (Q value) and 63–71% surface hardness (L value) compared to the host rock, and its growth rate is 1.21 mm/year. Therefore, this mineral is considered to have a crack repair function as it grows. According to observations by scanning electron microscope, many microorganisms were confirmed to be present in the minerals, and genera of Actinobacteria, Proteobacteria, Cyanobacteria, and others were detected by 16S rRNA analysis. Additionally, experiments using these microorganisms revealed that photoautotrophs and other microbial communities are deeply involved in the production of silicate minerals. These results significantly enhance the potential for developing a silicate-based self-healing technology for weathered and/or cracked rock in nature.

In the present investigation an attempt is made to study the axial and oblique pullout capacity of monopile and belled pile embedded in geogrid reinforced sand beds by conducting tests on model piles. A model tank of size 1.2mX0.75mX1.5 m is used to study the behaviour of an aluminium model pile under axial and oblique pull-out loads in geogrid reinforced sand bed of relative density 30%. A comparative study was made between monopile and belled piles by varying parameters such as embedment ratios (L/Ds) of 5, 7.5, and 10 respectively and diameter ratios (Db/Ds) of 1.5 & 2 with bell angles of 30°, 45°, and 60° under axial and oblique inclination angle of 30˚& 45˚.The use of the optimum size of the geogrid reinforcement placed at the bottom of the short pile and at a spacing of 4.5Ds from top in case of intermediate and long pile significantly enhanced the pullout and oblique pullout capacities of both monopile and belled piles. In the case of a monopile, the pull-out capacity increases with an increase in the load inclination angle from 0˚ to 45°, whereas in the case of the belled pile, the pullout capacity decreased with an increase in the load inclination angle. A non-dimensional parameter called the Uplift Improvement Ratio was evaluated, and the results revealed that the efficiency of the geogrid reinforcement was independent of the load inclination, bell angle, Db/Ds ratio, and L/Ds ratio when the optimum size of the geogrid was placed at optimum position of the belled pile foundation. It was also found that the obtained breakout factor increases with an increase in the L/Ds ratio and load inclination angle but the influence of the bell angle on the breakout factor was found to be marginal.The obtained values of the breakout factor were compared with the existing limit equilibrium theory, and it was found that the results of the breakout factor were in good agreement with the Meyerhof and Adam theory for both unreinforced and geogrid-reinforced sand beds.

Improving problematic soils using economical and appropriate methods is a major focus in geotechnical research. Recently, nanoparticles such as nanoclay have gained attention as a cost-effective and eco-friendly solution for enhancing soil quality. Nanoclay’s high specific surface area and ion exchange capacity make it economically viable, while its environmental friendliness comes from its derivation from bentonite minerals. This case study aims to improve a local sandy soil with low shear strength in an earthquake-prone region south of Qom City, Iran. The study uses nanoclay stabilization and compares its effectiveness on improving the local soil under both static and dynamic loading conditions, providing a novel approach. Thus, a series of direct shear and simple shear tests were performed to accomplish this specified task. After adding nanoclay to the soil, the sandy soil pores were filled with a cohesive mixture, increasing its agglomeration, which ultimately led to the following results. The results showed that incorporating nanoclay significantly increased the cohesion (c) value by an average of 12 times and raised the internal friction angle (Φ) by 15% compared to the untreated soil. Furthermore, the shear modulus (G) increased 1.4 times on average, while the damping ratio (D) decreased by 25% relative to the original soil. Overall, the use of montmorillonite nanoclay proved to be highly effective in enhancing the strength of non-cohesive sandy soils, especially under static loading. Additionally, the method of stabilizing soil with nanoclay was found to be economical for improving its strength in both static and dynamic conditions.

This study presents a comprehensive investigation of landslide susceptibility along the 83.5-km Chukyatan-Kumrat road, Upper Dir, North Pakistan. Despite its critical role in transportation and tourism, the region faces recurrent landslides due to hydrometeorological hazards, posing significant threats to stability. Employing a multidisciplinary approach, this study integrates the geological strength index (GSI) calculated from joint analysis of bedrock and landslide susceptibility index (LSI) analysis to understand the complex interactions underlying landslide occurrences. The study area contains a variety of rock formations, including metavolcanic, andesite, metarhyolite, igneous rocks, volcanic limestones, granodiorites, and spotted slates, which are overlain by remnant soils. Utilizing the landslide susceptibility index (LSI) map developed via the frequency ratio technique, regions proximal to road cuts, fault lines, and mineralogically altered and sheared lithology are identified as highly susceptible to future sliding events. GSI and rock mass rating (RMR) analyses categorized jointed bed rocks into relatively stable (zones 1 and 2; GSI 66–59, RMR classes II and III) and sheared and altered (zones 3 and 4; GSI 37–15, RMR class IV) segments, highlighting their differing susceptibilities. These zones have a moderately to highly weathered, slicken-sided jointed structure that allows rainwater and snow to infiltrate. The alteration mechanism of minerals such as chlorite, biotite, amphibole and alkali feldspar, as well as the influence of freeze–thaw cycles and precipitation on the pores and joints of bedrock, further weaken the rock, and there is a serious risk of landslide. This research contributes to the development of effective natural disaster mitigation and preparedness measures in the Chukyatan-Kumrat region. This study provides valuable insights for mapping landslide vulnerability in similar geological settings.

Under-reamed piles are deep bored cast-in-situ concrete piles with single or multiple bulbs formed by enlarging the pile shaft. Such piles are best suited to soils with significant ground movement as a result of seasonal variations, filled up ground, soft soil layers, and loose sand. These piles also improve bearing and uplift capacities, as well as anchorage at greater depths. This paper aims to investigate and analyze the behavior of under-reamed piles under inclined tension loads, as well as to compare the results with regular piles. Single and double under-reamed piles with bulb diameter ratios of 2 and 3 were used as model piles. These piles were embedded in sand with relative densities of 30, 50, and 80%. The tension load was applied at zero eccentricity above the soil surface with inclination angles of 15º, 30º, 45º, 60º, 75º, and 90º with respect to the horizontal direction. The results indicated that the ultimate inclined tension capacity of the under-reamed pile increased as sand relative density, number of bulbs, and bulb diameter ratios increased. Furthermore, when the inclination load angle decreased, the ultimate tension capacity increased due to the generated passive earth pressure in front of the pile. The horizontal load component of the inclined load has more effect on the axial capacity; and this effect decreases with the increase in sand relative density. Finally, a theoretical equation was proposed to estimate the under-reamed pile’s ultimate inclined uplift capacity. This equation correlated well with the experimental results as R2 = 0.91: 0.96.

Evaluation of embankment dam displacement (D) under earthquake loading can contribute to the safe design of the dam. Due to the complexities of modeling this problem, soft computing is an appropriate solution for predicting the embankment dam displacement under earthquake loading. In this research, Artificial Neural Networks (ANN) and Swarm Optimization Algorithm (PSOA) were integrated in an attempt to present a relationship for predicting the displacement of embankment dam (D). For this purpose, data from 102 real cases was utilized. Input parameters included the height (H) and natural period of the dam (Td), minimum required yield acceleration to slide a block of the dam body (ay), magnitude (Mw), dominant frequency (Tp), and peak acceleration (amax) of the earthquake. It was figured out that PSOA-ANN outperforms PSOA in estimating earthquake-induced dam displacement. Compared to other soft-computing methods for predicting embankment dam displacement under earthquake loading, the hybrid PSOA-ANN is more powerful and suitable.

In Japan, the disastrous effects of the 2011 Tohoku Earthquake has affected seismic design. Furthermore, there is an increasing need for dynamic nonlinear analysis that can consider large strain levels in assessing the seismic resilience of essential structures such as nuclear power plants. However, the applicability of dynamic nonlinear analysis methods to rock masses has not yet been fully evaluated. Their applicability needs to be confirmed by dynamic phenomena, such as shaking table tests. Herein, as a preliminary step, quasi-static analyses of cyclic direct shear tests were conducted to validate the constitutive model. Soft rock samples with relatively few fractures were used as the testing material. In the tests, the load was applied in multiple steps by increasing the loading amplitude at each step, with a loading frequency of 0.1 Hz. The quasi-static analyses revealed that, although an existing constitutive model generally reproduced the shear load–displacement relationship obtained in the tests from small displacement levels to peak strength, discrepancies emerged beyond peak strength. Therefore, we focused on the post-failure history damping characteristics of the soft rocks and added a damping constant after rock failure to the existing constitutive model. Consequently, the accuracy of the values of the stress history beyond peak strength was improved by more than 10%, and the reproducibility of the shear load–displacement relationship improved. Therefore, the applicability of the improved constitutive model was confirmed for conditions of quasi-static cyclic loading of soft rock with few fractures, from small displacement levels to beyond peak strength.