Advances in Geomechanics and Geotechnical Engineering

Random forest machine learning (ML) techniques have been used within a geographic information system (GIS) to develop regional scale landslide susceptibility maps for a local government area (LGA). To do this, a landslide inventory and database of known landslides are needed to be collected and curated, the various factors contributing to landslides identified, the data combined at a given scale, the random forest trained to the database of landslides, predictions made across the LGA, post processing and validation/sanity checking performed. The ML technique predicted landslide events with an acceptably high degree of accuracy. There are some residual uncertainties in the mapping which were considered to be no greater, and probably fewer, then if the mapping was done in the traditional way using geological reasoning.

In this paper, the authors firstly discuss the issues of large settlements of foundations or artificial islands on soft soil grounds, including the cases of the Tower of Pisa, Japan Kansai Airport Reclamations, and Hong Kong reclamations. Most of such large settlements can be attributed to the viscous behaviour (creep) of clayey soil skeleton. The mechanism of creep is briefly explained. The paper then presents Hypothesis A and Hypothesis B methods for calculating consolidation settlements of clayey soils, with the history and equations of the two hypotheses, explaining the inherent logical mistakes of Hypothesis A method. Two different methods for consolidation analysis of clayey soils are described, including the fully coupled numerical method with different Elastic Visco-Plastic (EVP) model, and a simplified Hypothesis B method, namely simple method, for one-layer and multi-layers of clayey soils. Steps of how to derive this simple method are presented. Two examples of using this simple method by hand calculations or spreadsheet calculations are explained. Verifications of the simple method by comparing with lab data and values from fully coupled numerical methods are presented, including a general simple method for multiple layers and vertical drains with analytical consolidation solutions by the spectral method (Walker and Indraratna in Géotechnique 59(5):439–449, 2009 [22]). The fully coupled numerical methods and a simple method presented in this talk have been adopted in the 5th edition of “Canadian Foundation Engineering Manual” (CGS in Canadian Foundation Engineering Manual (CFEM), 5th edn., 2023 [3]).

Interferometric Synthetic Aperture Radar (InSAR) innovative satellite technology is becoming popular for the measurement of displacement and velocity of ground movement over a large area in open cut mining, tunnelling, landslide and infrastructure projects in many parts of the world. Transport for NSW (TfNSW) and Ventia are piloting a trial InSAR project for monitoring slopes and cuttings from five (5) locations across the NSW road and rail corridors. TfNSW is managing about 9000 registered slopes and cuttings across the road and rail corridors with increasing challenges, especially with a rise in slope and cutting failure incidents in recent years in extreme weather conditions (e.g. floods). Traditional investigation methods have limitations and subjectivity which sometimes lead to undetected or unaddressed slope/cutting failures. This paper presents details of InSAR monitoring results from a period of May’23 to April’24 for slopes/cuttings in areas of Waterfall Way, NSW. The results also compare with historical movements/failure of a known down slope. Following the monitoring results, it is found that InSAR is capable of measuring displacement and vector ground movement with times over a large area of slopes/cuttings. The results also suggest that InSAR technology is likely to provide significant values for slope risk management, improved safety for road/rail users, reduced maintenance cost and greater value for money in managing slopes and cuttings of TfNSW.

Early and ongoing collaboration between Sydney Metro (SM) and the property developer was key to develop a robust design to ensure that a beneficial outcome was achieved for all stakeholders involved. The project went from being a sterilised site due to complex geometry constraints and additional limitations imposed by the SM protection reserves to a successful development application following an agreed approach for concession from SM to build within the first reserve. 55 Pitt Street is a new 53-storey commercial tower located at the northern end of the Sydney CBD. The building is located immediately above one of the Sydney Metro City and Southwest (SM C&SW) running tunnel, with the basement excavation extending into the tunnel’s first reserve. Through collaborative and ongoing engagement with SM, an agreed foundation design and permanent structure located within the first reserve and a minimum 2.5 m offset from the extrados of the tunnel lining was successfully developed and constructed. The design consisted of a bridging structure—spanning the running tunnel—supported by foundation piles with diameters up to 2.4 m designed for a maximum ultimate load of 165 MN. The engagement process involved impact assessments of the excavation sequence and foundation solution as well as implementation of mitigation and monitoring plans. Numerical modelling was carried out to estimate the overall ground movements and demonstrate the overall impact to tunnel lining. The development also had direct interfaces with the Telstra Tunnel, heritage listed Tank Stream, and Ausgrid easement and substation each requiring individual assessment and engagement with their respective asset owner. The Site is mainly formed in alluvial soils overlying Hawkesbury Sandstone and adjacent to the Pittman LIV Dyke which is located along the Site’s southern boundary. The ground model including overall stratigraphy, in situ stress condition, and rock/joint parameters was developed according to site-specific site investigations, surrounding excavation mapping, and past project experience. A range of numerical packages (RS2, 3DEC) utilising continuum and discontinuum methods were utilised to develop localised and district scale models to simulate the proposed construction sequences and effects of the surrounding excavations. The numerical modelling results formed the basis of the impact assessments, monitoring strategies and a basis for a design which allowed the development to proceed and be successfully constructed.

This paper presents a case study of innovative and sustainability considerations of contaminated materials for embankment design and construction for the Coffs Harbour Bypass (CHB) project. The first consideration was development of a project specific Sustainability Management Plan (SMP) for CHB project to satisfy the Scope of Works and Technical Criteria (SWTC) requirements. The key themes outlined in the CHB SMP included: (1) climate change resilience; (2) whole of life impact; and (3) design optimisation. To efficiently integrate sustainability considerations through all stages of design, relevant themes have been selected and allocated to each individual design package. The main considerations were presented through four case studies of the project regarding the encapsulation cell design for contaminated soils and in situ uncontrolled fill within the CHB project corridor. As the encapsulation cell material includes asbestos contaminated soil and building rubble, conventional construction compaction testing by nuclear densitometer was not considered practical. For this reason, a performance-based material placement and compaction methodology was specified. It is therefore expected that the settlement profile of an embankment containing encapsulation cells will be comparable to that of an embankment constructed wholly from general earth fill. A contaminated soil mound was developed onsite due to an excessive amount of contaminated soil for encapsulation and ease of construction. During construction, the mound DCP profiling was introduced to validate the quality of compaction in lieu of plate load testing. The monitoring results of the constructed embankment are within the expectation without any surprises.

In geotechnical design, the simulation of soil behaviour is the most important facet. Often the behaviour of clay soils is modelled using the modified Cam Clay model (MCC), because it is readily available in many commercially available finite element programs used by practitioners for geotechnical design. The MCC model is formulated considering the experimentally observed remoulded clay behaviour considering isotropic consolidation conditions and developed within the critical state framework with an elegant formulation with a clear physical interpretation. However, natural clays in their in situ undisturbed state are anisotropic and have a structure, which gives an additional strength compared to the same soil in the reconstituted state. Hence, the simulation capability of the MCC model is considered inadequate for natural clays. The focus of this paper is to investigate the significance of the anisotropy and structure of natural clays using the SANICLAY family of soil models implemented in the finite element modelling program ABAQUS. The basic model behind all SANICLAY models is the MCC model. In SANICLAY models, the anisotropy of natural clays is introduced by rotational hardening where yield and plastic potential surfaces are rotated and distorted to an ellipse. Two distinct types of destructuration included in SANICLAY are isotropic and frictional. Isotropic destructuration is introduced through the softening of the yield surface. The frictional destructuration replicates the reduction in friction angle due to the breakage of soil structure. The paper presents numerical simulations for triaxial tests conducted on undisturbed Bothkennar clay samples to show the capability of the chosen soil model in simulating the experimentally observed natural clay behaviour. Then a parametric study will be presented emphasising the importance of anisotropy and structure of natural clays comparing results from the MCC and SANICLAY family of models. Results of this study show that the anisotropy and structure are important parameters to simulate the behaviour of natural clays.

In situ testing equipment and methodologies have evolved rapidly over the last five years. This paper looks at three devices that demonstrate this evolution: (1) a special purpose 3 MPa CPTu cone; (2) the automated Medusa flat plate dilatometer; and (3) the automated down-the-hole vane shear device. CPTu cones are now commercially available with special internal design and with capacities as low as 3 MPa. These “special” cones, when carefully calibrated, are capable of reliably measuring CPTu parameters in soils right to the bottom end of the very soft range. The cones come with a temperature sensor to enable the management of potential inaccuracies associated with transient temperature effects during the penetration. The Medusa DMT was developed by Marchetti to provide technicians and engineers with complete control and repeatability of the DMT diaphragm inflation and measurement process, eliminating many of the operator-dependant variables often encountered with the traditional gas-operated DMT. The vane shear test is perhaps the most relied-upon geotechnical strength test; however traditionally, it has had in-built potential errors, particularly in very soft soils, mainly due to ambiguity in friction corrections; it has been traditionally restricted to use in pure clay or clay-like soils. Equipment design has evolved (in some equipment) to eliminate the friction-correction problem. High-quality calibration is required. These advanced in situ testing tools and improved methodologies were utilised in combination to characterise a low strength soil deposit in Sydney Harbour for a major infrastructure project. The results from the use of the advanced equipment and methods are discussed and reviewed in this paper.

Concrete injection columns (CICs) are commonly used in geotechnical ground treatment to meet serviceability and stability design criteria. Together with a load transfer platform, the design objectives are achieved by bridging future fills across existing fills and soft materials to transfer loads from a higher level to a stronger underlying material layer. Efficient CIC design within brownfield sites can be challenging when there is high variability or uncertainty in the depths and extents of existing fills and soft materials. This case study presents the CIC ground treatment design development and validation required to support the future rail earthworks platform for the Sydney Metro West—Western Tunnelling Package Clyde Stabling & Maintenance Facility (SMF). The selected performance-based geotechnical ground treatment design comprised 0.45 m diameter CICs at typical 2 m centre to centre (c/c) spacings to overcome the risk of excessive site wide settlements and meet the project settlement criteria. Validation included a CIC installation trial to calibrate CIC rig torque readings intersecting soft materials and a static load test to demonstrate CIC settlement performance in line with predicted settlement behaviour, reducing uncertainty in the installation methodology and production CIC performance prior to construction.

In recent years, numerical analysis has emerged as a powerful tool for tackling a wide range of geotechnical problems, including complex ground-structure interactions. Advances in software and hardware have expanded the capabilities of numerical packages, enabling diverse design options and comprehensive output reporting. Despite these advancements, the process of developing numerical models and extracting results remains resource-intensive and time-consuming, often leading to increased costs and heightened risk of human error. Automating these procedures can accelerate the design process, reduce costs, and allow expert resources to focus on more technical tasks. Numerical packages often include built-in programming languages, such as Python, which can be used to develop automation codes. However, geotechnical practitioners frequently encounter challenges such as time constraints and a lack of programming skills. Furthermore, companies are generally hesitant to hire experienced programmers solely for developing automation codes. This raises the question of how geotechnical engineers can acquire the necessary skills to efficiently develop automation codes for their routine design tasks. The recent advancements in artificial intelligence (AI), particularly Generative AI (Gen AI), offer promising solutions for learning and coding. This paper outlines a methodology employed by geotechnical practitioners with limited coding experience to create practical automation codes. It highlights how Gen AI platforms, such as ChatGPT, act as effective virtual assistants, accelerating the coding process and enhancing productivity. The paper presents three case studies demonstrating the advantages and benefits of automating numerical procedures. Additionally, it discusses the challenges associated with using Gen AI platforms as virtual assistants for geotechnical practitioners.

Retaining walls are used to retain soil and create a separation in elevation between two areas of the road and highway network. To keep vehicular traffic on the elevated side safe from falling over, safety parapet barriers would be erected at the top of the wall. In the event a heavy vehicle collides against the parapet barrier, the impact loading may destabilize the retaining wall. The load dispersal is quite complex, and engineers often grapple with finding a practical and reasonable way of incorporating it as a pseudo-static live loading in the design of the retaining wall. Current literature on the loading distribution useful for geotechnical stability design seems to be lacking. This paper discusses an approach using finite element simulations to develop better insights of the complex impact loading issues. The simulation results are used to develop an equivalent 2D plane-strain pseudo-static loading to the impact loading by taking geotechnical stability into consideration. This approach is applied to a 1-m and a 3-m high retaining wall (both with a 1.2 m high parapet barrier integrated at the top) to serve as illustrative examples.

A valuable tool in discontinuum modelling is the concept of effective block interactions (size), where predominant structures are modelled explicitly and minor structures implicitly, i.e. as an equivalent pseudo-continuum rock block. This allows practitioners to achieve computationally effective models. However, a frequent question in this modelling approach relates to what “effective block” properties are considered adequate. As the rock mass still contains discontinuities smeared through its volumes, it cannot, in principle, be modelled as an intact block. It should not have rock mass properties estimated at the scale of the geo-structure either, i.e. using the global scale geological strength index (GSI), as this would unrealistically double up the strength and deformability downscaling (reduction). Unfortunately, no detailed guidance can be found in the international technical literature, and decisions are often based solely on the practitioner’s experience in assessing a GSI value at the effective block scale. This paper presents an example of the impact of the GSI value when using the effective block size approach, varying the scale of the block sizes in the discontinuum model from the actual rock block volumes to larger volumes that are more effective to use in practice. Although the example does not exhaustively explore all aspects associated with the problem at hand, thus not intended as a detailed how-to guide, some preliminary conclusions can be drawn on appropriate GSI values for “effective blocks”, particularly for sub-horizontally bedded rock formations with mostly sub-vertical jointing.

This paper presents the authors’ combined experience of the behaviour of Sydney Shale during the construction of surface and underground excavations in the Ashfield Shale and the Bringelly Shale in the Sydney region. It includes observations made during the excavation phase of large-scale infrastructure projects, such as WestConnex M4 East, Sydney Metro Northwest, Sydney Metro—Western Sydney Airport Station Boxes and Tunnelling and Sydney Metro West—Western Tunnelling Package. These projects involved cut and cover and mined tunnel and cavern excavations for railway stations, rail tunnels and motorways. The vertical surface excavations discussed here are up to 38 m deep and 150 m long. The largest cavern described is 26 m wide, 27 m high and has a cover of 15 m. Rock mass classification systems used in Sydney frequently classify shale and sandstone according to the same parameters: defect spacing and intact strength. This paper highlights significant differences in the engineering geological properties and rock mass behaviour of Sydney shales when compared to Sydney sandstones. Examples of observed rock mass behaviour and associated geotechnical risks relevant to Sydney shales, such as wedge failure in vertical walls, tunnel face instability and overbreak in the tunnel crown are included. Comments are made on the nature of geological defects in shale relevant to the construction of both surface and underground excavations. The authors highlight items for consideration in the design of excavations in shale, in contrast to sandstone, in terms of ground support, excavation geometry, excavation sequence, construction monitoring and field verification of expected ground conditions. Photographs documenting observed rock mass behaviour, and the character of typical geological defects are presented. Monitoring data quantifying observed movements are included.

This paper addresses key geotechnical challenges inherent to the design of deep cut and cover/retaining structures, and associated excavations, in weathered faulted Silurian age siltstone. It also presents recommendations on how to apply innovative geotechnical design approaches to deal with this unique geological setting. The paper briefly explains some the features encountered within this geological setting and the intrinsic challenges and risks they pose to design and construction. The former includes high-stress rock masses either with fractured/faulted zones or with bedding planes with an unfavourable orientation and dip angle relative to the excavation. From a design perspective, approaches on how to incorporate the impact of these geological features in a soil-structure interaction (SSI) analysis are discussed in detail, which cover not only the shortcomings of two-dimensional analytical assessments but also the need for a critical calibration between 2 and 3D analyses to derive design efficiency. The impact of bedding planes is explored as part of the design process, as well as performance during construction of a deep, multi-anchored soldier pile wall. This segment offers an insightful overview on the influence of bedding planes on the stability and serviceability performance of the wall, illustrating the intricate interplay between geological features and engineering design principles. A separate section of the paper focuses on the repercussions of highly fractured/faulted bands of rock on the design of diaphragm wall structures. The design approaches adopted to capture the effect of such features via an efficient SSI approach are discussed, along with the key findings of targeted additional investigations that may assist in de-escalating the risk profile of a project during successive stages of design.

The $5b Woolgoolga to Ballina (W2B) Pacific Highway upgrade, Australia’s largest regional infrastructure project, was successfully completed in 2020. It involved the upgrading of 155 km of the Pacific Highway. One of the biggest challenges was to construct approximately 27 km of road infrastructure on soft compressible ground in a tight construction programme. Depending on the embankment height, soft ground depth/thickness, long-term settlement criteria and construction programme, ground treatment types varying from preload only to surcharging with wick drains and rigid inclusion (CMC or CIC) were adopted at different sections. A total of approximately 3000 instruments (7 types) were installed and monitored to assess the performance of the embankments during construction, to estimate post-construction settlement in the long term, and to manage the risks of slope instability of high embankments on deep soft ground. This paper will discuss a typical procedure for instrumentation and monitoring, project specific instrumentation requirements [1], procurement process, instrument calibration and installation, monitoring and data processing, performance of various instruments, management of instrumentation contractors, decommissioning of installed instruments and handing over to the client. The lessons learned on what went well, what not and the root causes will be summarised; the best engineering practice for instrumentation will also be recommended for future projects.

The Standard Penetration Test (SPT) is arguably the most commonly used in situ test in Australia. It is quick, straightforward, and many relationships between the results (SPT N-Value) and geotechnical strength parameters have been published. However, there are a few downfalls with the SPT test; it is not repeatable in the field, drillers often use a convenient rod size even though there is a “standard size” adopted in Australia, N-Values are often used in their uncorrected form and the correlations to geotechnical parameters are not site-specific. SMEC has recently completed the geotechnical investigation for Project Energy Connect in southwestern New South Wales, which included over 1500 investigation data points, including boreholes, cone penetration tests, geophysics, and laboratory testing. These data points have been used to assess a localised, site-specific correlation between SPT N-Value and geotechnical strength parameters, notably undrained shear strength (C_u) of cohesive soils. This paper presents the correlation of SPT N-Value to C_u for the alluvial deposits of southwestern NSW and compares this correlation with those typically used within the industry.

This paper presents a summary of the time-history seismic analysis for a proposed deep foundation design consisting of a screw pile group installed into problematic deep soft ground. The ground conditions generally comprised of fill up to 4.5 m deep followed by peat swamp deposits of a very soft to soft consistency followed by very soft to soft estuarine silts and loose to medium dense sands. Alluvial and estuarine soils of a firm or better consistency were encountered at depths between 16 and 20 m. The design analysis confronted two pivotal questions: the potential seismic liquefaction of the soil mass and the dynamic seismic loading effects on the pile group. To respond to the first question, faced with the soft silts and loose sand layers on the site, analysis was conducted to identify the likelihood of the liquefaction potential. Based on the site piezocone penetrometer testing (CPTu) data obtained, liquefaction analysis was undertaken using the earthquake design method (Robertson in performance-based earthquake design using the CPT. In: International conference on performance-based design in earthquake geotechnical engineering, Tokyo, 2009 [5]). A probabilistic assessment of earthquake-induced liquefaction potential was also quantitatively conducted to demonstrate that the anticipated probability of liquefaction (P_L) at the site was less than 5% corresponding to a classification of ‘Almost certain that it will not liquefy’ based on the definition of liquefaction likelihood classification (Juang et al. in Assessing probabilistic methods for liquefaction potential evaluation. In: International conferences on recent advances in geotechnical earthquake engineering and soil dynamics, pp. 1–6, 2001 [1]). For the evaluation of dynamic actions on the pile group, a scaled time-history accelerogram of the 1989 Newcastle Earthquake, recorded at Lucus Heights was employed as the excitation source. A two-dimensional (2-D) soil-foundation-structure-interaction (SFSI) analysis was performed using finite element programme Plaxis to assess the dynamic seismic performance of the pile group. Key aspects discussed include input motions from the seismic source, consideration of tectonic and seismic settings in Newcastle, and 2-D time-history finite element analyses to capture seismically induced deformations, bending moments and shear forces along the pile group. Additionally, an independent one-dimensional (1-D) assessment including structural analysis by Spacegass, calibrated with finite element method (FEM) output, was conducted. A general agreement between the 2-D geotechnical Plaxis and 1-D structural outputs confirm the validation of the analysis. The results of the time history and SFSI analyses reveal variation ranges of bending moments, shear forces, induced settlement and horizontal movement along the pile group during seismic events. Notably, the maximum variation of bending moment and shear force remained within the proposed pile group capacity. This study contributes valuable insights and methodologies applicable to similar infrastructure designs in the Australian region, emphasising robust design and analysis practices in challenging geotechnical and seismic environments.

The problem of soft foundations on the coasts is one of the serious problems in the construction of marine structures, such as breakwaters. During construction of a breakwater, rock materials penetrate the soft clay bed, and the foundation of the breakwater is not formed on the seabed, which causes problems of excessive settlement, instability, and loss of rock materials. Estimating the depth of penetration of rock materials is necessary to evaluate the amount of settlement of the breakwater and estimate the volume of rock materials required for the implementation of the project. In this paper, an analytical solution related to the phenomenon of penetration of spherical particles into the sludge is introduced using the impact balance relationship. By considering the most important parameters effective in penetration as variable factors, a simple relationship has been developed. In addition to the rapid penetration of rock materials into the soft clay, the amount of material loss depends on other phenomena such as the consolidation of the clay layer, the flow of soft materials into the pores between the rocks, surface erosion, etc. By estimating the impact of each of the mentioned cases, the amount of loss of breakwater rock materials can be evaluated.