Proceedings of the Indian Geotechnical Conference 2022 Volume 9

The present study considers the two-dimensional seismic analysis of a concrete gravity dam along with massed foundation and the reservoir. The coupled system is subjected to spectrally matched mainshock-aftershock sequences, and the response thus obtained has been compared with the case when only the mainshock earthquake ground motion is considered. Response parameters considered are the crest displacement, stress at heel, crack length, dissipated energy and contact displacements. The results suggest that aftershocks lead to an enhancement of these response parameters in a dam that has been damaged due to mainshock but has not yet been repaired. An attempt has also been made to discuss two types of damage indices that are suitable for consideration of the cumulative effect of damage due to aftershocks.

Failure of earthen embankments during earthquake has been reported often in most of the major events (Loma Prieta Earthquake (1989), Hyogo Ken Nambu Earthquake (1995), Bhuj Earthquake (2001), Sumatra Earthquake (2004), Great Eastern Japan Earthquake (2011), etc.). Most of these failures are reported to be due to the vulnerability of foundation soil upon which the earthen embankment is founded. Liquefaction of subsoil has been attributed to be one of the major causes for such vulnerability. These embankments play a major role in a country’s growth and any damage to these facilities will hamper the economic development of infrastructure, especially with respect to rural development. Considering this fact, there is a requirement to strengthen the subsoil by densification or use of an alternate solution to prevent structural failures by making subsoil strong. It becomes even more necessary to plan the strengthening of existing structures so that earthen embankments with weak subsoil founded in seismically active zones can be retrofitted considering future seismic event. Different mitigation measures can be proposed such as use of geotextiles, use of sheet pile walls or providing dense wall at appropriate locations in the subsoil to strengthen it. In this paper, an attempt is made to use GEOSTUDIO, a finite element software with two dimensional plane strain idealization to model the earthen embankment with and without dense wall and study the liquefaction vulnerability of subsoil and hence the overall performance of earthen embankment. The parameters such as location, stiffness, geometry of dense wall, initial condition of subsoil and intensity of earthquake shaking will be assessed from the overall seismic performance consideration of earthen embankment. The results suggest that the provision of dense wall can be one convenient measure to strengthen the subsoil from liquefaction vulnerability.

India has been prone to several major to minor earthquakes from ancient times. Although the prediction of earthquake time is unreliable, prior seismic hazard studies may help against significant earthquake loss. Seismic hazard studies effectively exhibit solutions for future events at a particular location by combining different mathematical models and addressing uncertainties and challenges. This paper has carried out the seismic hazard analysis of the Tindharia landslide, Darjeeling Sikkim Himalayas and its surroundings based on the deterministic approach considering an updated historic earthquake database and models within a 300 km radius. The method determines the worst-case ground motion for the best maximum magnitude (M_max) at the shortest source-to-site distance (R_min). The hazard analysis was carried out using linear seismo-tectonic data considering faults, lineaments and shears zones collected from different sources. The M_max is calculated using the five most widely used empirical relations involving magnitude and fault rupture length. The maximum among them is considered the maximum magnitude of the fault. The peak ground acceleration for each source has been calculated using the ANBU-13 attenuation equation at 30, 50 and 75 km focal depth. The controlling earthquake of the site is carried out by comparing the levels of PGA produced by various earthquake sources.

Occurrence of repeated liquefaction caused severe damage to the foundations and structures (e.g. 2012 Emilia-Romagna seismic sequence). Limited research works have been reported in understanding the mechanism behind the occurrence of repeated liquefaction. This paper presents the mechanism of liquefaction subjected to sequence of seismic events using 1-g uniaxial shaking table apparatus. A rigid watertight container of dimension 1050 mm × 600 mm in plan and depth of 600 mm was used for testing the sand specimen. The specimen was prepared at 25 and 50% relative density and tested under repeated uniform sinusoidal loading conditions. Loading was applied in the sequence of 0.3 g (1), 0.3 g (2), 0.3 g (3) and 0.3 g (4) with constant dynamic frequency 5 Hz and 1-min shaking duration. Excess pore water pressure was measured using the pore pressure transducers at different depths to observe the occurrence of repeated liquefaction. In addition, induced sand densification after each shaking event was estimated. Reliquefaction mechanism was complex and different from that of the liquefaction subjected to independent shaking events. It is found that the effect of preshaking and shaking induced soil densification are critical in influencing the reliquefaction resistance. Despite the obtained beneficial effects of repeated shaking events, pore pressure ratio increases with the successive application of uniform sinusoidal loading events.

Site response analysis heavily weighs dynamic soil characteristics including liquefaction potential, shear modulus, and soil damping ratio. The current work studies the response of clayey sand owing to dynamic loads. Saturated undrained cyclic triaxial tests have been conducted using strain-controlled loading at different relative densities (50–80%) representing the field conditions and by varying shear strain amplitudes and confining pressure under 1 Hz loading frequency. Test results have shown that soil's ability to liquefy is reduced when confining pressure and relative density increased. For higher shear strains liquefaction susceptibility, damping ratio increased significantly and shear modulus reduced.

The extent of an earthquake destruction varies from location to location because it is influenced by the local site impacts and the nature of the bedrock motion. The behavior of the soil beneath a structure during the propagation of earthquake waves determines the safety of the building at any location. When it comes to the ground surface, the amplitude, frequency, and duration of seismic motion at the bedrock changes owing to the topography of the site and the soil geomorphology. To determine the impact of local site conditions, borehole data from Sri Damodaram Sanjeevaiah Thermal Power Plant at Nellore is collected and one-dimensional ground response analysis is performed using both equivalent-linear and non-linear analysis using DEEPSOIL. SPT “N” value is used to correlate shear wave velocity. In this study, the amplitude content and frequency content of the Chi-Chi motion, Loma-Gilroy motion, and Kobe motion are taken into consideration. Peak surface acceleration and design response spectra are used to compare and illustrate the results of the two analyses.

Heritage sites in India are well known for brilliant architecture, massive structures and stone carvings which symbolizes its cultural and civilizational aspects. Unfortunately, the construction of the glorious past has suffered degradation due to events such as earthquake, weathering, vandalism and human negligence, etc. Conservation of these structures are of utmost importance with respect to our civilizational and economic value. In the present study, the risk level for 207 heritage sites in Gujarat falling under UNESCO and ASI monuments are evaluated based on the seismic zonation map of India, the seismic hazard and liquefaction hazard study for Gujarat region from literature. The study aims to identify the vulnerable sites by assigning a risk factor to each site with respect to the PGA values of the area based on the three aspects: earthquake zones, probabilistic seismic hazard and liquefaction hazard. The different weightages of these three aspects are accounted and the mean risk for each site is evaluated to prepare the hazard risk map of heritage sites in Gujarat. The future direction on how to account the structural condition assessment of the heritage structure to the evaluated mean risk is also provided by performing condition assessment of UNESCO world heritage site Rani ki Vav.

An attempt is made in this study to understand and quantify the liquefaction-induced lateral spreading potential of Guwahati city soil in the Indian state of Assam. Past studies have shown that Guwahati city soil is prone to liquefaction. Liquefaction-induced lateral spreading in Guwahati city was reported during the 1897 Assam EQ (MW-8.1). However, no studies have been conducted to identify the probable locations or to estimate the displacement lengths of lateral spreading during similar future earthquakes. In this study contour maps developed by previous researchers for scenario earthquakes of 1869 Cachar EQ (MW-7.5) and 1897 Assam EQ are used to identify sites with high liquefaction potential. The identified sites are further classified as prone to lateral spreading based on sub-soil properties, layer thickness, fines content, depth of ground water table, ground slope, earthquake magnitude, and source-to-site distance. The aforementioned parameters are then employed in a Multiple Linear Regression (MLR) model to estimate the amounts of displacement at the classified locations. The displacement lengths are found to range between 0.02 and 0.11 m. Based on case studies the flow direction of lateral spreading at some of the locations is also identified.

Quay wall failures are mainly attributed to the properties of backfill soil and foundation soil. From the research available in the field of earthquake geotechnical engineering, it is evident that the failure of quay walls during Niigata earthquake of 1964, Bhuj Earthquake of 2001, Sumatra earthquake of 2004 and The Great Tohoku earthquake of 2011 are caused due to the failure of foundation soil and/or backfill soil. It has been observed that the improvement in the ground conditions leads to lesser deformation of seismic structures. It therefore becomes essential to strengthen the backfill soil and the foundation soil by suitable improvement methods and mitigate such catastrophic failures. The problem of liquefaction and associated deformation of backfill and foundation soil can be controlled by various ground improvement techniques such as densification of soil, providing drainage at suitable locations, dewatering, and introducing reinforcements (Kramer and Holtz, National Science Foundation, 1991). As an attempt to achieve improved performance, a detailed study is carried out on the effect of ground improvement adopting reinforced earth technique on backfill soil and foundation soil of quay wall systems. For this purpose, GEOSTUDIO, a finite element software has been used. An analytical model is prepared to represent a quay wall system comprising of the wall, the backfill soil and foundation soil and analyzed in 2D plane strain idealization. The influence of ground improvement on the permanent horizontal displacement, vertical settlement and amount of tilting of quay wall subjected to strong ground motion is studied. A comparative study is also made on the influence of strength and stiffness characteristics of foundation soil and backfill soil on the seismic displacement of the quay wall which will contribute towards performance-based design of seismic structures such as quay walls.

The silty sand obtained from Waynad, Kerala was found to be liquefaction susceptible under saturated condition when subjected to cyclic loading. The stabilization of the soil using agar biopolymer was found to impart significant resistance to liquefaction since a great reduction in excess pore pressure build-up was observed. In this study, the effect of biopolymer treatment on the secant shear modulus and damping behavior of the soil under varying biopolymer content, curing period and over-consolidation ratio (OCR) have been studied. A set of cyclic triaxial tests were performed on the untreated and biopolymer-treated silty sand. It was observed that the shear modulus increased with increase in the biopolymer content, curing period, and OCR. Furthermore, the biopolymer-treated soil was found to exhibit a higher damping when compared to the untreated silty sand.

Ground vibration is generally caused by construction activities, e.g., pile driving, sheet pile driving, dynamic compaction of soil, blasting; natural activities, e.g., earthquake, landslide, etc. In urban region, construction-induced vibration is mainly caused by pile driving, sheet pile driving, etc. Pile driving induces vibration in the soil which is transmitted to neighbouring structures, thereby causing potential damage, such as unequal settlement of foundation and distortions or cracks within the building. In the present study, field observation has been conducted at two sites during pile casing driving by (i) Impact Hammer upto a depth of 28 m beneath the ground surface, (ii) Hydraulic Vibratory Hammer upto a depth of 10 m beneath the ground surface. A comparative study has been made between two sites based on the following observations. Peak particle velocity (ppv, mm/s) has been measured in the sites with varying distances from the pile driving location. Ground motion amplitude has also been acquired at different depths while driving pile casings. Based on the site observation, regression model has been created to predict the peak particle velocity. The model is very useful for predicting PPV for site-specific parameters. Thereafter, it has been compared with the existing ground vibration model obtained from field study.

Rock socketed piles efficiently transfer the heavy loads from high-rise buildings, long-span bridges, and offshore structures to bearing strata. In the present study, A 3D finite element analysis of a rock socketed pile subjected to earthquake loading is carried out by PLAXIS 3D, and the rock socketed pile is embedded in a jointed rockmass and weathered rockmass. The pile material is modeled as linear elastic, whereas Mohr’s Coulomb and Hoek- Brown material models are used for the simulation of soil and rock. The acceleration-time history of the 1991 Uttarkashi earthquake is considered for the seismic analysis. The observations are recorded to understand the effect of dynamic loading on the displacement of socketed piles for different socket lengths. The effect of soil cover depth and socket length are also studied in the present work. The socket length considered is 3, 5, and 10 times the diameter of the pile. It is observed that the settlement of the rock socketed pile increases with an increase in weathering grade from W0 to W4. The settlement of the rock socketed pile is also dependent upon the GSI parameter of rockmass.

This study is devoted to the risk associated to earthquake of an urban area which is great deal for the local authorities. As India has a relatively high frequency of having great earthquake with low frequency of moderate earthquake so there is need for efficient tools to assess associated with it. Patna sahib which is a Lok Sabha constituency located at Patna District in Bihar which falls under seismically active zone IV, classified as a severe zone by National Center for seismology. Patna Sahib is well known as the birthplace of Guru Gobind Singh, the tenth guru of Sikhs. The assessment of seismic activity in Patna Sahib is done by RADIUS software, which includes making of Grid mess with the use of software Global Mapper, distribution of region. Assessment of loss of lives, building damage is done by using various datasets such as soil types, number of buildings, and earthquake scenario. The deterministic analysis is done considering an earthquake of moment magnitude 6.5 using RADIUS software. The result is shown in the form of tables and various thematic maps in the GIS environment.

The soil desaturation method for mitigation of sand liquefaction has gained interest in recent years. The application of microbes for various ground improvement studies has proven to be cost-effective and non-disruptive compared to other conventional methods. The potential of soil desaturation by introduction of non-pathogenic denitrifying bacteria is studied in this paper. This bio-chemical process used to induce partial saturation leads to the nucleation of gas bubbles within the soil pores which generate gas as final product. This evolved gas occupying the pore spaces shifts the soil matrix to quasi-saturated state affecting the rate of pore-pressure. Prior to the evaluation of soil strength properties, batch tests were conducted to understand the effect of initial nitrate concentration on generated gas volume. To understand the effect of microbial de-saturation on static behavior of the soil, undrained strain-controlled static triaxial tests were conducted for loose sand condition. Results showed increase in peak deviatoric stress after treatment, change being prominent in loose condition indicating enhanced resistance to static liquefaction. B-value and water content measurements were taken to monitor the change in degree of saturation. The rate of excess pore pressure development in undrained compression loading also showed a decreasing trend.

This paper presents the results of numerical analysis by using 10-noded PLAXIS 3D software. A drained analysis was carried out using Mohr–Coulomb’s model for soft soil and granular piles. 120 × 4 × 40 m soft soil deposit is used to place a raft of size 16 × 4 m at 2 m below the ground level. The soil was reinforced with a group of six granular piles arranged in square pattern, each having diameter 1 m, at 2.5 m c/c spacing and of length 10 m. The ground response (in terms of amplification) is studied for the imposed 1990 Upland, USA earthquake of magnitude 5.4 data (acceleration versus time) under two situations namely the one without the granular piles below the raft and the other with the granular piles. Amplification is the ratio of value at surface to ground input motion in the form acceleration for this study. The peak amplification value corresponding to unimproved and improved soil is obtained as 41.12 and 4.36 respectively. Thus, significant reduction in peak amplification is exhibited by inclusion of granular piles in soft soils.

In the recent past, more focus has been given to the practical utilization of Artificial Neural Networks (ANN) in solving diverse geotechnical engineering prob-lems. The present study mainly aims to evaluate liquefaction potential for Visakhapatnam city based on IS 1893 Part-1 2016 method using an artificial neural network. Earlier researchers have developed back propagation artificial neural networks to predict the liquefaction potential of subsoil and concluded that modeling of any com-plex relationship between seismic, soil parameters, and liquefaction potential is possi-ble with neural networks. These models are reported to be simpler and more reliable than conventional methods of evaluating liquefaction potential. In the above context, an attempt has been made on a total of 10 boreholes data in the city premises of Visakhapatnam at different locations, which spreads over the coast-line. The most critical input parameter identified in the modelling of the network is the Standard Penetration N-Value. The data set in the model was trained, validated, and tested in the ratio of 60:20:20. The final results showed that neural networks are a powerful tool in predicting the occurrence of liquefaction potential. These predictions are almost 90% similar with an acceptable confidence level to the IS 1893 Part-1 2016 method.

The proper disposal of used tyres is a serious and genuine ecological problem. The level of reuse is significantly less when compared with the accumulating consistently. It is well known that rubber serves as an excellent damping material. Therefore, there is a huge possibility for the utilisation of these rubber pieces for earthquake-resistant design purposes. This study suggests developing an economic seismic isolation system utilising soil and waste tyre scrap. The inclusion of tyre chips modifies the damping properties of the soil by expanding energy dispersion. A FEM investigation of the isolated framework is analyzed by PLAXIS 2D. The time history of acceleration of some well-known seismic tremors is utilised as an input motion in the study. Due to the damping effect produced by soil-rubber tyre isolators, peak ground acceleration response is observed to decrease considerably. The de-amplification values found for the acceleration at the model’s peak of the isolated building are promising.

Tunnels are constructed in mountainous region to provide straight and comfortable routes. Often, mountains consist of rock mass, which are of highly vulnerable in nature because of the presence of random discontinuities. In Himalayan region of India, many tunnels are proposed to be constructed due to their high strategic importance because these tunnels help to provide the troops quick access to the border areas in all weather conditions. Himalayan region is seismically very active and falls in seismic zones IV and V. Seismic analysis of tunnels situated in Himalayan region considering jointed rock mass is hardly reported in the literature and the same has been studied in the present research work. In this research work, the dynamic response in terms of acceleration and vertical deformations in static as well as in dynamic conditions are evaluated at the crown and invert of the tunnel. The discontinuity in the rock mass is considered by jointed rock model with different orientation of joints. In the present study, FE analysis using 15-noded triangular elements has been carried out using PLAXIS 2D software to examine the harmonic response of tunnel in jointed rock mass and then effect of frequency of excitation is also examined.

This study introduces a novel micromechanical-based contact modeling approach to investigate granular material behavior and fabric formation. By deriving particle contact properties from micromechanical tests, the study aims to enhance the accuracy of discrete element method (DEM) simulations and gain insights into the multiscale behavior of granular materials. The evaluation of small-strain shear modulus (G_max) behavior in quartz sand, along with the assessment of particle contact parameters, offers new perspectives on the roughness characteristics of granular materials. This innovative approach has the potential to advance our understanding of granular material behavior and inform the development of more accurate models with broad applications in science and engineering.

Stone columns are fundamental geotechnical units for improving soft soil deposits. It is a common application for supporting an isolated footing over stone columns when subjected to static loading. The same can also be used for earthquake loading. However, when the same element is subjected to an earthquake loading, the behaviour would not be the same. In such cases, the geometry requirement needs to be studied. The study can be done numerically using finite element analysis software PLAXIS 3D. In this study, the earthquake of magnitude 6.9 on the behaviour of the stone column was analysed. Parametric analyses were carried out to determine the adequacy of the stone column geom-etry for both OSC and ESC. The parameters studied are the area replacement ratio and the encasement stiffness. The results proposed the optimum values for the area replacement ratio as 40% and 35% for earthquake loading for OSC and ESC, respectively. The optimum encasement stiffness is proposed as 1000 kN/m for static loading and stiffness greater than 2000 kN/m for earthquake loading. In this paper, the procedure for the earthquake analysis using PLAXIS 3D is described to provide better clarity to the readers.

This study focuses on controlled implosion, a novel demolition technique to raze large structures at a reasonable cost and short time duration. Implosion is the controlled placement of explosives and timing of their detonation to cause a structure to quickly collapse in on itself with the least amount of physical harm to its immediate surroundings. This demolition technique was used for the deconstruction of the unlawful Supertech Twin Towers in Noida, India as per the Supreme Court verdict. The UK company Vibrock estimated the intensity of vibrations likely to be created when the building hit the ground based on the blast implosion plan details and the type of soil. Aster-2, a B + G + 12 storey, neighbouring structure just 9 m away from the tower to be demolished was analysed to ensure its safety due to touchdown vibrations resulting from demolition. Aster-2 was modelled in ETABS software and subjected to different loading conditions including blast-induced vibration and earth quake loadings. The comparison between behaviour of structure under blast loading and seismic loadings is presented, and the safety of the structure is evaluated as per codal provisions. The findings indicate that the structure is safe under the blast loading induced. Crack meters were installed to monitor the variation if any after the blast and it was found that structure didn’t had any additional damages due to the implosion. The vibration was monitored using geophones at the basement level of Aster 2 and the peak particle velocity noted was well within the permissible limits.

Failures in rock masses are one of the leading causes of recurrent disasters like landslides, rock falls, instability during tunnelling, dam foundation failures, etc. As the frequency of such catastrophic events increases, thorough forensic investigations of such failure events are essential to obtain representative information of the failure regions and improve the understanding of the factors controlling the stability of rock masses. This paper reviews reported case studies on forensic investigations of rock failures to assess the various tools and techniques employed to inspect the different types of failures. The importance of detailed forensic investigations to developing a better understanding of the factors responsible for the instability of rock masses is also highlighted in this paper. The review of these failure observations can also aid as a guide for selection of appropriate mitigation measures to avoid such disasters that cause significant socio-economic losses.

The Central University of Tamil Nadu (CUTN), Thiruvarur constructed 30 numbers of Staff Quarters with load bearing wall in 2013. During the end of 2016, considerable vertical, horizontal and diagonal cracks of 5 to 10 mm wide were noticed, in one building, on the exterior walls of living room and also caving-in of floor tiles and plinth protection slabs to a depth of 50 to 100 mm. The type of foundation was strip footing of 2 m width placed at 2 m depth (Footing pressure is 68.98 kN/m²). Bore log data revealed that the top 4 m soil below the footing is dominantly expansive clay of N-value 4 to 7 underlain by sandy layer of N-value 4 to 7. The consolidation settlement of 142 mm and swell pressure of 410 kN/m² were computed for the soil below the footing. Hence, it was evident that excess settlement and swell pressure due to untreated expansive soil are the primary causes of distress in building. To remediate distress, it was recommended to stabilize the foundation soil by lime injection in the entire footing area of building. In the severely affected area, under reamed piles with single bulb was recommended up to a depth of 3.6 m.

In recent years Bengaluru city has witnessed a series of building collapses, unsurprisingly in monsoon. The majority of these collapses appear to be due to the failure of the foundation. The high cost of land in urban areas has motivated constructors to build in the vertical direction of the site rather than in the horizontal direction, even on the smallest piece of land to maximize the floor area. In an effort to save some money, land developers have been ignoring proper geotechnical investigation and structural planning before starting construction. This lack of awareness of soil investigation’s importance has led to many building failures in various modes. After studying the forensic research done on case histories of building failures in Delhi, the various modes of building failure have been identified as (i) Lateral soil movement, (ii) Unequal settlement, (iii) Draw down and heave, (iv) Change in water level, etc. This study aims to identify the various reasons for building collapses in Bengaluru city through a forensic geotechnical approach and their remedial measures through suitable ground improvement techniques.

Foundation failure of structures is rare until the structure comes under the influence of groundwater. Lack of geotechnical field knowledge may lead the structures to collapse in a short duration just after construction. It is unfortunate that many failed structures in India are not reported because they are buried under administrative procedures. This practice has influenced the constructors to neglect the soil investigation process before planning and also ignore the national building bylaws so as to obtain a maximum built area to save some money. The current study discusses the forensic analysis of a building collapse that occurred in Bangalore and a review of foundation failure that occurred in a 13-storey building in Shanghai, China. The study consists of the location of a site, failure date, details of the structure, the casualties that occurred and general causes, geotechnical aspects and remedial measures to the failures, which would provide answers to several confusions like what went wrong, when, why, where, how and by whom the structure faced the issue. This study visualize a strong input to accelerate future designs and the importance of different civil engineering fields and expert op (Subramanian (2009) Rare Foundation Failure of a Building in Shanghai, China, NBM and CW Journal.) inions in construction and create awareness about soil investigation.

Foundation is a vital element of a structural system, which helps to transfer the structural load to soil uniformly. Therefore, the strength and stability of structure depend on the behaviour and properties of soil. The collapse of any structure mainly due to the differential foundation settlement and also due to the punching shear failure is in the case of loose soil deposits. Geotechnical investigation is important to control the failure of any structures. It includes surface and sub-surface exploration of the site. The cost of site investigation depends on the size of the structure, built-up area and other factors. Even though approximately 30% of India consists of problematic soil, the scope of soil investigation is underestimated. This paper aims to focus on the impact of the inadequate soil investigation report on the structural stability and financial risk caused due to over-designed foundations. The forensic analysis of a recent building collapse in India and review of the foundation failure of a case study clearly show the result of limited site investigation. The study consists of the location of the site, details of failure, common reasons and the general remedial measures to be taken to control these failures.

A reliability-based design and analysis of reinforced anchors for transmission tower foundations  is presented using the Kriging-based response surface method. The improvement of the uplift capacity of anchors in reinforced soil is demonstrated using a three-dimensional numerical model. The initial design of the anchor plate is based on the uplift forces exerted at the foundation of a typical tower subjected to lateral wind forces. Next, a Kriging metamodel is constructed to predict the uplift capacity of anchors in unreinforced and reinforced soils to conduct reliability-based analysis considering the random nature of the uplift load, soil properties, and reinforcement stiffness. The mean uplift capacity of the anchor at 15% uplift displacement is increased by 1.35 times in the presence of reinforcement on the top of the foundation. The probabilistic analysis results indicate that the failure probability of the foundation in reinforced soil reduces significantly for the uplift forces obtained at the transmission tower foundation level. The influence of the variability of different load and resistance parameters on the stability of the foundation is examined. The variation of the reliability index with the factor of safety is presented, and the limitations of the conventional safety factor approach are discussed.

Soil is a naturally occurring material, so considering spatial variability residing in soil properties seems logical. Assessment of Probability of Failure (PF) adds more aspect to the determination of the safety of a system. In this study, probabilistic analysis along with deterministic analysis is carried out for different slope angles and variation levels of shear strength properties. The deterministic factor of safety, probabilistic factor of safety and probability of failure are calculated for different slope geometries with and without consideration of cross-correlation between cohesion and friction angle under static conditions. The effect of different variation levels on slope under dynamic conditions is also studied. Three different LEMs have been utilized for analysis, and random variables are considered to follow normal distribution. The results prove that FOS and PF have an inverse relationship which is non-linear, but their non-linearity is greatly affected by COV levels. Results prove that the critical slip surface with minimum FOS obtained through deterministic analysis may not be the critical slip surface after all. Probability of failure (PF) more than 10% is not acceptable in any case. Probabilistic analysis along with deterministic analysis should be carried out to fully consider the geotechnical risk and hazard.

Response surfaces are commonly adopted as a surrogate for complex performance functions due to their ability to solve rock slope reliability problems with low computational costs. Most widely used methods employ a static scheme that uses input (rock properties) and output (factor of safety) samples generated using some experimental design to obtain the best fit parameters of the response surface. However, since the selection of input samples is not optimized, the increase in accuracy of the response surface often comes at the cost of an increased number of performance function evaluations, particularly, for slopes having a low probability of failure (\({P}_{f}\)). This is addressed by an active learning scheme that iteratively selects input samples that improve the prediction of the response surface around the failure region. In this paper, an active learning scheme with support vector machine (SVM) is adopted for estimating the \({P}_{f}\) of a rock slope along the Rishikesh–Badrinath highway against planar failure. The analytical expression for the factor of safety is utilized for conducting Monte Carlo simulation to estimate \({P}_{f}\), which is treated as a benchmark for determining the accuracy of the proposed method. Comparison with static scheme SVM illustrates the advantages of active learning scheme in increasing the accuracy in estimating the \({P}_{f}\) for a similar number of performance function evaluations.