This paper presents an attempt to conceptually develop unorthodox Markov chains to model the probability of failure evolution of tailings storage facilities (TSFs), focusing to linking TSF Quantitative Risk Assessments (QRA) to data observed after its compilation. The driving idea is to offer users a way to optimize costly QRA updates frequency and to guide the selection of adequate TARP types along the life of a considered dam. A companion paper (Markov Processes for Tailings Dams. Part 2: Applications, Oboni, 2026 in this Journal) delivers the results of various analyses on real dams, including the simulation of aging under decaying care, which show the model capabilities and offer insight on possible practical uses, for instance to guide the time to QRA updating. Markov chains, including various degrees of unorthodox assumptions, are used in numerous engineering and business applications. For instance, bridges, pavements and buildings degradation under service or seismic loadings have been modelled using various unorthodox extensions of Markov chains but applications to TSF QRAs have not been found to date. QRAs are necessarily based on carefully selected parameters and should be complemented by many observations made on an ongoing basis, for instance during inspections. However, observations made after the first QRA are difficult to integrate in a swift manner, especially if new monitoring techniques are implemented delivering data which may not be compatible with the original engineering analyses feeding the prior QRA, for example deformation velocity by satellite observation vs. limit equilibrium stability analyses. The result is a blatant gap between the prior QRA results and day-to-day observations, which may prove difficult to fill due to time constraints and costs. Recent papers (Oboni, 2026a, b) discuss these difficulties with reference to Trigger Action Response Plans (TARPs) and in this paper it is postulated that Markovian processes can help fill the gap provided some formulation extensions are performed. The unorthodox Markovian model proposed in this paper is agnostic to the QRA procedure, provided it can deliver the required data. A list of twelve possible observable physical conditions pertinent to TSF dams is selected based on experience, ranging from deformations, water management, accidental overloads, leaks and erosion, and spillway issues. This list constitutes an unorthodox construct with respect to strict Markovian definition, but successful precedents exist under the form of Hidden Markovian Models (HMM). The list allows to build a Markov transition matrix, which is considered to be generally applicable to TSFs of various construction types. The paper shows how to build the transition matrix and confirms its behaviour in terms of convergence and steady state evaluation. The paper closes with a section covering possible future development and with a list of takeaways from the present stage leaving the application to its companion cited above (Part 2).
The automation of numerical analysis in geotechnical engineering provides a practical pathway for improving design efficiency and supporting more robust decision-making. This study presents an AI-assisted workflow integrating Python scripting and finite element modelling to automate model generation, parametric analysis, and results extraction. Two case studies are used to demonstrate the application of the framework: a representative parametric assessment of pipeline response adjacent to an excavation and a real project involving a piled raft foundation system. The automated approach enables rapid evaluation of a large number of scenarios, reduces manual effort, and supports systematic exploration of design alternatives. The results show that automation significantly reduces analysis and post-processing time while enabling evaluation of up to 240 scenarios within a single workflow. This capability supports more informed and potentially more resource-efficient design decisions, while facilitating improved multidisciplinary coordination. The framework also provides the capability to incorporate additional scenarios, such as variations in groundwater conditions and loading scenarios, where required. The study highlights the role of automation in enhancing the efficiency, scalability, and consistency of geotechnical design workflows, offering a practical tool for modern engineering practice.
Whenever I find myself in a new experience, doing something that once felt beyond me, I pause and think, I never imagined I would be here. That is how I measure how far I have grown. Over the past seven years, I have been learning to find my voice. As someone who is naturally introverted, I often wished I showed up with more confidence and presence. That wish sent me on a journey filled with meaningful lessons and many aha moments. This is my attempt to capture some of those lessons and reflections, in the hope that another young engineer who is still finding their own voice might find something here that resonates.
This article presents a review of the book “Dispersive Soils - Processes, Impact and Management”, Edited by Pichu Rengasamy and Ehsan Tavakkoli. Publisher: CSIRO Publishing (AU & NZ), Melbourne, Australia.
A 59 storey tower underlain by a 30m deep basement is currently under construction within the expanding central business district of Parramatta, NSW. The tower is surrounded by and abuts two multi-storey buildings with basement levels, a heritage listed church, a 100 year old water main, and is set back approximately 50m from the Parramatta River. A comprehensive site investigation program was carried out to characterise the complex geotechnical conditions of the site, including deep cored boreholes, insitu permeability testing and borehole imaging. Due to the challenging ground conditions encountered, proximity of the neighbouring structures and shallow groundwater, the shoring system consists of secant pile walls terminated above bulk excavation level in competent bedrock, below which a grout curtain extends approximately 50m below surface level to limit seepage inflows. An Instrumentation and Monitoring (I&M) program was implemented to assess actual basement wall movements against those predicted from the numerical analyses.
Shotcrete linings usually achieve good bond to the underlying rock substrate when first applied during construction. This bond usually remains sound through many years of service. However, in some cases the bond may be inadequate or diminish with the passage of time causing the lining to detach from the substrate. To prevent detachment, and possible collapse in these circumstances, a shotcrete lining needs to be secured in place using a secondary means of attachment which, in the case of linings on hard rock, usually comprises rock bolts with external cover plates or spider plates on the ends of bolts. The potential causes of bond loss are many both prior to spraying and in service. Testing of bond strength frequently leads to apparent ‘zero bond’ results that may reveal an actual absence of bond or may be an outcome of the coring and testing process. If bond is relied upon in design it is important to consider the significance of zero bond results rather than simply dismiss them as artefacts of the testing process. This paper considers how partial or total bond loss can be considered in shotcrete lining design.
Excavation of a deep underground station required support of high and narrow sand backfill that was placed by others during earlier construction of an adjacent 26m deep basement. The sand backfill was supported by a complex retaining system including a rock pillar and two levels of props. This paper describes design of the support system, monitoring results, retaining system operation and dismantling of the props. The design employed a number of design methodologies developed from first principles including application of three methods for calculation of earth pressure from narrow sand backfill, using beam analogy for preliminary calculations of stresses in the rock pillar and of prop force, design criteria for assessment of stresses in the rock pillar and analyses of potential influence of temperature changes on prop forces and the rock pillar stresses. Construction and operation of the retaining system included systems for movement monitoring and prop force monitoring. The paper describes how monitoring results confirmed adequacy of design methodologies and enabled flexible operation of the retaining system when needed, such as more frequent adjustments of prop forces which was required due to higher than assumed degree of prop restraint. Dismantling of the propping system included its modification to suit construction methodology of the Station while monitoring results during that stage further indicated agreement with design assumptions.
Geotechnical engineers have traditionally relied on engineering charts for the analysis and design of specific geotechnical problems. However, interpolating target design parameters, particularly on logarithmic scale charts, can be time consuming and susceptible to human error. Recent advancements in machine learning enable engineers to efficiently approximate design parameters by training models on extensive datasets, thereby minimizing both time and manual intervention. Furthermore, coefficients for closed-form equations can be derived from these models in some cases, streamlining computational analysis and enhancing design workflows. This paper presents two case studies: one focused on shallow footing settlement assessment and the other on single pile settlement assessment. It illustrates the application of non-linear regression, high-degree polynomial regression, Gaussian process regression and fully connected neural networks in developing effective machine learning models for graphical approximation.
Progressive brittle fracture plays an important role in rockfall initiation. Although the magnitude of in situ stress near the surface of a slope may be low relative to the strength of intact rock, gravitational stresses become concentrated near the tips of pre-existing discontinuities, promoting the slow process of subcritical crack growth. Over time, subcritical crack propagation reduces the size of intact rock bridges that interrupt an incipient failure surface, increasing the stress intensity at the advancing crack front. Eventually the stress intensity reaches a critical threshold where the fracture toughness of intact rock is exceeded, causing a rapid acceleration of crack growth and sudden failure. This paper demonstrates a modified fracture mechanics model for time-dependent subcritical crack growth, applied using the bonded block discrete element method. A series of conceptual numerical models are developed based on the geological environment of Blue Mountains National Park in New South Wales, where the Triassic sandstone formations of the Narrabeen Group outcrop in cliffs up to 200 m high. The landscape of the Blue Mountains has been shaped by stream incision and escarpment retreat processes that now are dominated by gravity-driven slope failures, including rockfalls that vary in magnitude from discrete minor block falls up to rock mass scale cliff collapse events. This investigation first explores a simplified model for progressive failure of an overhanging sandstone slab, driven by time-dependent propagation of a vertical rear release joint. The methodology is then extended to consider collapse of a sandstone cliff by undermining failure of a weaker underlying shale layer. The models demonstrate how the time required for rockfall initiation varies depending on the geometry and persistence of pre-existing discontinuities and the intact rock bridges that must fail for an incipient rockfall block to detach. When combined with empirical methods for estimating rockfall magnitude-frequency relationships, the proposed methodology can help to improve the temporal estimates of rockfall probability that form a critical input to rockfall risk assessment.
Tunnelling-induced ground surface settlement (GSS) poses potential risks to buildings and underground utilities in urban areas. Commonly used approaches for GSS assessment, including three-dimensional (3D) numerical simulations and artificial intelligence techniques, are often limited by high computational costs or poor generalizability. To overcome these limitations, this study proposes a simplified and efficient two-dimensional (2D) numerical method for rapid estimation of GSS at different tunnel cross-sections. The approach replaces the detailed 3D step-by-step excavation processes by employing 2D plane strain conditions, while incorporating convergence patterns characterized by the gap parameter (g) and volume loss (V-L). Several typical tunnel deformation models are numerically analysed to evaluate their influence on the resulting settlement troughs. The results demonstrate that the proposed method can effectively capture key GSS characteristics while improving computational efficiency. In addition, this method enables probabilistic analysis in the absence of comprehensive field data, supporting early design decisions. Future research can further refine this approach by incorporating more advanced soil constitutive models and considering tunnel-soil interaction effects to improve the accuracy of GSS evaluation.
This article presents a review of the book “Guidelines for Open Pit and Waste Dump Closure”, Edited by Phil de Graaf, Geoff Beale and Dr Trevor Carter. Publisher: CSIRO Publishing, Melbourne, Australia.
This article presents a review of the book “Exploration Magnetics – Theory and Practice”, Edited by Phil Schmidt, James Austin, David Clark, Keith Leslie, Mark Lackie and Clive Foss. Publisher: CSIRO Publishing, Melbourne, Australia.
Investigative drilling (ID) is a modern measurement while drilling (MWD) technique that has been effectively used in site investigations for several long corridor projects in Australia. The drilling data collected through the ID method has provided clients with valuable in situ strata verification for earthworks and footings. However, the use of drilling data from ID has been primarily qualitative. This paper presents a preliminary assessment of the ID method's potential for quantitative site characterisation through case studies involving soils and rocks. Comparative analysis with conventional boreholes demonstrates that ID data provides significant insights into ground conditions and can be used to characterise soil and rock properties. Specifically, the penetration rate, as a single drilling parameter, proves effective in identifying rock fractures, while compound indices such as the soil-rock resistance, Somerton index, and drilling energy correlate with the standard penetration test (SPT) values and rock strength. Despite the promising results, further rigorous testing is necessary to validate these findings, given the inherent uncertainties in subsurface conditions and conventional testing results.
The METRONET Yanchep Rail Extension (YRE) is one section of Perth's most ambitious public transport program of works. The passenger rail infrastructure runs 15 km north from the existing Butler Station to the Yanchep stowage yard and includes three new station precincts and over a dozen bridge structures. The rail alignment passes through regions known for karstic limestone conditions and is primarily formed within a large cut setting requiring bulk excavation in the order of up to 10 m to 15 m below the existing ground level. Due to project and site constraints, contiguous piled walls without propping, anchoring or tiebacks have been constructed to support the ground within these large cuts. Fully cantilevering piled retaining walls of this height and within these geological settings are unconventional. Importantly, they require consideration of post-peak strength loss associated with the predicted strains and lateral wall movements, coupled with effective management of karstic risk. This paper discusses the development, application, advantages and limitations of an innovative "strain-softening" numerical analysis approach developed for the wall design. An efficient and pragmatic approach used to mitigate karstic risks during design and construction is presented, along with the instrumentation and monitoring that was adopted to monitor the wall performance.
Cognitive dissonance is the incompatibility between any two elements of knowledge or belief. We often filter information that conflicts with what we already believe, to avoid contradictory statements occurring. The term can also be loosely used when contradictory statements occur. There are many contradictory statements in Geotechnical Engineering-a few of these will be discussed. There are many common geotechnical practices that have historically served the engineering community well, but continue to be used in common practice, even when later knowledge has shown inconsistencies. Can risk management substitute for data? But even with data, why would a group of Engineers presented with the same data so often have different conclusions? When did data become a point of view? A sampling of such dissonance in common geotechnical practice is presented through case studies.
This paper presents a probabilistic vulnerability analysis of two pit profiles within an active mine in Austral Africa. The profiles exhibit differing geological conditions, necessitating a nuanced assessment of their stability under varying operational and care scenarios. The study employs a prior published quantitative semi-empirical methodology to evaluate the evolution of annualized probability of failure (PoF) under different factors of safety (FoS) and levels of maintenance. The analysis encompasses four hypothetical scenarios representing varying degrees of care, maintenance, and dewatering activities for each profile. As expected, results indicate that maintenance practices, particularly dewatering, significantly influence slope stability, with neglect leading to elevated PoF levels. The study extends across multiple time horizons, from yearly assessments to long-term projections spanning 5, 10, 20, and 50 years. Findings reveal that while increasing FoS generally reduces PoF, scenarios with reduced maintenance and FoS exhibit persistent vulnerabilities, especially over longer time frames. Profiles in more difficult geological conditions, demonstrate heightened vulnerability, underscoring the importance of tailored maintenance strategies. Comparisons with benchmark values derived from industry standards provide further insights into the effectiveness of current maintenance practices. This analysis reaffirms the significance of the current standard of care within the examined pit. Moreover, the proposed approach, complemented by Bayesian updates, holds promise for informing long-term and closure designs, facilitating risk-informed decision-making, and supporting efforts to achieve ALARP (As Low As Reasonably Practicable) conditions.
This study presents the results of static load tests on expanded base piles, installed using the novel PAILE method. This method facilitates the installation of a range of foundation and ground improvement piles with enlarged bases and, therefore, at a significantly reduced material consumption compared with current methods. The results of the static load tests in loose to medium-dense sands demonstrate that the performance of piles installed in loose to medium dense sands using the PAILE method can be predicted using direct CPT methods. This allows the determination of pile capacity and load/settlement factors corresponding to drilled displacement and precast driven piles, respectively. Additionally, the PAILE method allows the resistance of each expanded base pile to be measured during installation. As a result, the risk related to insufficient geotechnical capacity is minimised and the resistance is recorded for every single substructure. Advanced analytics can be performed on the multi-sensor installation data using Artificial Intelligence, hence the "AI" in the PAILE name. By utilising proven, existing technology in a novel way, expanded base piles can be installed using smaller machinery, with higher productivity and less material than with traditional piling equipment.
Ison Road Overpass bridge with high approach embankment at each abutment has been proposed to carry the existing Ison Road over Geelong to Melbourne rail corridor, extending the road south towards Browns Road in Werribee, Victoria. The high approach embankment behind each abutment requires a retaining structure to retain its earth fill. At the same time, collision protection shall be considered in accordance with AS 5100.1 for all structures within 20 m adjacent to existing or proposed rail track centrelines. An innovative hybrid retaining system consisting of L-shape wall, reinforced soil structure (RSS) block and required ground improvement works has been progressively developed through value engineering process as part of design development which was then built successfully. This paper commences with a brief description of geological conditions along the proposed project alignment and then discusses the optioneering process to develop the innovative and cost-effective retaining system to support the high abutment approach. The paper moves further to discuss key geotechnical issues and challenges encountered in design, basis of design, design approaches and assumptions, ground improvement works needed, subsequently presents key observational views from construction of the retaining system and finally presents some technical insights for future design and construction of similar retaining system as conclusions.
This study investigates the evolution of the porosity of crushed sedimentary rock masses in one-dimensional compression for dry and inundated conditions with wetting-induced collapse at different stress levels in the context of underground pumped hydro energy storage (UPHES) facilities, where the collapsed underground void of a disused mine serves as the lower reservoir. Three fine-grained rock lithologies typically present in underground mine roofs, indurated mudstone, interbedded sandstone-mudstone, and massive siltstone with unconfined compressive strength (UCS) values in the range of 25 to 157 MPa, were subjected to constant strain rate compression tests up to a stress of 12 MPa, with different specimens of each lithology inundated at stresses of 50 kPa, 2 MPa, 5 MPa or 12 MPa. The results show that the weaker the rock, the greater the reduction in porosity under stress. The porosity reduction upon collapse is independent of the magnitude of the applied stress for the stress range considered in the tests, but its magnitude depends on the material. It is also observed that the porosity of the crushed rock samples in dry conditions under the maximum stress level (i.e., 12 MPa) increases linearly with the UCS on a semi-logarithmic scale. The stress-strain curves were scaled to account for the size of the rock boulders forming a real goaf, which is usually found in disused underground mines. It was found that the change in porosity upon collapse is not scale-dependent and, hence, can be satisfactorily characterised at the laboratory scale, provided that the particle size distribution is adequate.
The typical strength descriptor and parameter used for rock is Uniaxial Compressive Strength (UCS). In many small to medium size projects laboratory tests to determine UCS are only conducted on a few rock core samples, if at all. Point Load Strength Index (PLI) testing is commonly conducted at regular intervals on rock core. Correlations between PLI and UCS are used to assess rock strength and as input to rock mass classification systems. PLI tests are usually made parallel (Axial) and perpendicular (Diametral) to the core axis. The Axial results are commonly used to estimate rock strength. The ratio of Axial to Diametral PLI results can be used to assess the degree of anisotropy. It is common to compare UCS with PLI test results that are close together to establish site specific correlations but limitations in datasets often do not result in clear cut correlations. This paper presents a methodology for assessing UCS and PLI datasets and comparing these to the relationship between UCS and PLI assumed in AS1726 Geotechnical Site Investigation. There are three main bedrock units encountered in the Sydney Region: Hawkesbury Sandstone, Ashfield Shale, and Bringelly Shale. This paper presents assessments of the relationship between UCS and PLI for these bedrock units. The degree of anisotropy of each of the main Sydney bedrock units is also assessed and compared with published values.
