Digital Twins’ Application for Geotechnical Engineering: A Review of Current Status and Future Directions in China

Abstract

The digital wave, represented by new technologies such as big data, IoT, and artificial intelligence, is sweeping the globe, driving all industries toward digitalization and intelligent transformation. Digital twins are becoming an indispensable opportunity for new infrastructure initiatives. As geotechnical engineering constitutes a critical component of new infrastructure, its corresponding digital transformation is essential to align with these initiatives. However, due to the difficulty of modeling, the demand for computing resources, interdisciplinary integration, and other issues, current digital twin applications in geotechnical engineering remain in their nascent stage. This paper delineates the developmental status of geotechnical digital twin technology in China, and it focuses on the advantages and disadvantages of digital twins in five application fields, identifying key challenges, including intelligent sensing and interconnectivity of multi-source heterogeneous physical entities, integrated sharing of 3D geological models and structural models, unified platforms for lifecycle information management, standardization of digital twin data protocols, and theoretical frameworks for digital twin modeling. Furthermore, this study systematically expounds future research priorities across four dimensions: intelligent sensing and interoperability technologies for geotechnical engineering; knowledge graph development and model-based systems engineering; integrated digital twin entity technologies combining 3D geological bodies with engineering structures; and precision enhancement, temporal extension, and spatial expansion of geotechnical digital twins. This paper systematically reviews the application status of digital twin technology in geotechnical engineering for the first time, reveals the common technical challenges in cross-domain implementation, and proposes a theoretical framework for digital twin accuracy improvement and spatiotemporal expansion for geotechnical engineering characteristics, which fills the knowledge gap in the adaptability of existing research in professional fields. These insights aim to provide references for advancing digitalization, intelligent transformation, and sustainable development of geotechnical engineering.

1. Introduction

With the development of new technologies such as big data, the Internet of Things and artificial intelligence, currently, the digital wave is sweeping across the globe, driving all industries toward digitalization and intelligent transformation. Digital twin (DT) technology is rapidly integrating into our lives, with extensive development and application across multiple sectors, including manufacturing [1,2], healthcare [3], transportation [4], water conservancy [5], smart cities [6], mining [7], infrastructure [8,9], and so on.

Digital twins originated from the concept of “a virtual, digital equivalent to a physical product” proposed by Professor Michael Grieves of the United States in 2002 [10,11]. In 2011, the Air Force Research Laboratory of the United States [12] first clearly defined digital twins: A digital twin is a simulation process that makes full use of data such as physical models, sensor updates, and operation history; integrates multiple disciplines, multiple physical quantities, multiple scales, and multiple probabilities; and completes mapping in the virtual space, thereby reflecting the entire lifecycle process of the corresponding physical equipment.

Generally speaking, a digital twin constructs a virtual entity that is exactly the same as the object in the physical world, through digital means, thereby achieving the understanding, analysis, and optimization of the physical entity [13].

Because digital twins are based on the interaction between the virtual and the real and the digital-model-driven approach, they can break through many physical limitations and meet application service requirements such as simulation (reflecting the real with the virtual), control (controlling the real with the virtual), prediction (predicting the real with the virtual), and optimization (optimizing the real with the virtual), achieving immediate response to demands, continuous innovation of services, and upgrading and optimization of industries. Currently, more than 50 countries around the world, over 1000 research institutions, and tens of thousands of experts and scholars are conducting related research on digital twins [14].

In recent years, China has also developed rapidly in the field of digital twins. For example, in 2019, the Institute of Industry 4.0 of China initiated the world’s first Digital Twin Consortium (DTC), while the OMG of the United States followed up and initiated a similar American Digital Twin Consortium in 2020 [15]. In March 2021, the national “14th Five-Year Plan” outline clearly proposed to “explore the construction of digital twin cities”, providing national strategic guidance for the construction of digital twin cities. Since then, the state has successively issued the “14th Five-Year Plan” in different fields, making strategic deployments on how to use digital twin technology to promote high-quality economic and social development in fields such as overall planning, information technology, industrial production, construction engineering, water conservancy and emergency response, comprehensive transportation, standard construction, energy security, urban development, and so on [16].

According to data from the China Academy of Information and Communications Technology (CAICT), the digital twin market holds significant growth potential and vast development prospects. In 2022, China’s digital twin market reached a scale of CNY 10.4 billion. As digital transformation accelerates across industries, the penetration rate of digital twin technology is expected to rise, driving domestic market growth. It is projected that the Chinese digital twin market will expand to CNY 37.5 billion by 2025 [17]. Digital twinning has become an essential driver for industrial advancement and economic development.

With the proposal of new infrastructure construction, digital twinning has become an opportunity that cannot be let go in the new infrastructure construction. Geotechnical engineering is an indispensable part of new infrastructure construction. A large number of geotechnical engineering projects, such as underground pipe galleries, rail transit, high-speed railways, and highways, have the characteristics of large investment scale, long infrastructure period, complex construction environment, and high risk. There are prominent problems such as data and information islands, multimodality, and scattered applications. Traditional engineering management models and technologies struggle to meet the needs of the development of modern geotechnical engineering, and there is an urgent need to study and utilize new information technologies to promote intelligent construction and improve the management level.

The digital twin technology is reshaping the paradigm of geotechnical engineering practice. By constructing a closed-loop system of “perception–modeling–decision-making”, three breakthroughs have been achieved: First, the problem of underground space invisibility has been overcome. With the help of three-dimensional geological modeling and real-time monitoring data fusion, the hidden engineering state has become transparent and visible. Secondly, the uncertainty risk of geological conditions is significantly reduced, and the geotechnical parameters are continuously corrected by dynamic data assimilation technology. Finally, the whole-lifecycle management and control system is established, from traditional static design to dynamic predictive maintenance. This technical system not only improves the level of engineering safety but also promotes geotechnical engineering to enter a new stage of digital intelligence.

Introducing digital twin technology into the field of geotechnical engineering is a countermeasure to solve the complex geological conditions and construction environment of geotechnical engineering, and it is also the best way to achieve the digitization of geotechnical engineering investigation, the interactivity of design, the virtualization of construction, the intelligence of decision-making, the networking of monitoring, and the superiority of performance [18]. Digital twin technology builds a virtual model based on the physical integration of various kinds of information and the dual drive of data and models to present the entire lifecycle state of the entity in the real physical world, thereby achieving the simulation, prediction, monitoring, and optimization of the entire process. Only through such digitization can geotechnical engineering match the infrastructure construction and promote the sustainable development and improvement of the new infrastructure construction.

The purpose of this study is to review the application status of digital twin technology in geotechnical engineering, revealing the common technical challenges in cross-domain implementation, and proposing a theoretical framework for digital twin accuracy improvement and spatiotemporal expansion for geotechnical engineering characteristics, to provide references for advancing the digitalization, intelligent transformation, and sustainable development of geotechnical engineering.

2. The Current Situation of Digital Twins in Geotechnical Engineering

Geotechnical engineering is composed of soil mechanics, rock mechanics, engineering geology, and corresponding engineering and environmental disciplines. It serves different engineering categories, such as construction, water conservancy, hydropower, transportation, railway, aviation airports, water transport, oceans, petroleum, mining, the environment, and the military, and even aerospace and other engineering fields all cannot do without geotechnical engineering [19]. With the rapid advancement of projects such as urban rail transit, underground pipe galleries, high-speed railways, and highways, traditional engineering technologies and management models struggle to meet the needs of modern geotechnical engineering. The refined management and informatization construction of geotechnical engineering have become inevitable, and digital twins have begun to enter the field of geotechnical engineering. The current situation of digital twins in geotechnical engineering in various fields is as follows [20,21]:

(1)

In the field of urban construction:

After entering the 21st century, with the improvement of computing power and the development of big data technology, digital twins gradually expanded to the field of urban management [22]. In the early 21st century, the concept of smart cities gradually emerged, and various places began to explore how to use information technology to improve urban management efficiency and quality of life. Digital twin technology became an important tool in this trend. By the 2010s, the application of digital twins in cities began to receive wider attention. The virtual model of the city was no longer a static three-dimensional model but a dynamic system integrating multiple technologies, such as sensor data, real-time monitoring, and simulation prediction. At present, the application scenarios of digital twins have relatively mature practical cases in the fields of transportation, parks, urban emergencies, and so on [23]. The basic technology platform architecture of digital twin cities is shown in Figure 1. Cities such as Singapore and Amsterdam took the lead in realizing the application of digital twins, helping urban managers make better decisions.

Exploration of digital twin cities has also been carried out successively in China. For example, Hao Shang et al. [6] took the Spring Area in Jinan as an example and established a four-dimensional geological environment database and digital twin model based on the theory of digital twins, which can achieve the visualization of three-dimensional geographic and geomorphic models, water level models, karst models, borehole models, fracture models, geological models, and hydrological models, as well as above-ground buildings, underground structures, municipal pipelines, and subway models. This model can also change over time.

Qiang Li [24] proposed an urban digital twin model based on digital twin technology, which includes five modules—physical city, virtual city, intelligent service, twin data, and virtual–real interaction—and established an urban flood disaster assessment and early warning digital twin system (Figure 2), which helps to realize the functions of automatic monitoring and early warning of urban flood disasters, real-time intelligent command and dispatch, and improving the capabilities of existing cities in flood disaster assessment, early warning system perception, dynamic response, and real-time decision-making.

Danuta Szpilko et al. [25] proposed a digital twin-based framework for smart city energy management, which significantly improves renewable energy integration efficiency and grid stability through real-time data integration and AI-driven simulation optimization. By developing dynamic digital twin models, the system achieves predictive energy demand fluctuation analysis and optimized distributed energy allocation, providing decision-support tools for smart city infrastructure planning.

Feng Shao et al. [26] applied digital twin technology to underground space visualization, underground construction, and underground disaster prevention monitoring and analysis, and they established an intelligent overall planning model for underground space based on a digital twin model.

The Xiongan New Area, which is currently under construction, uses digital twin technology to simulate urban construction and traffic flows, helping planners design more efficient and sustainable transportation systems to enhance citizens’ travel experience.

The application of digital twin technology in urban construction not only helps to improve efficiency and optimize resources but also enhances the resilience and sustainability of cities. By creating a virtual city model, it can monitor, predict, and respond to various challenges in urban development in real time, and can promote the development of cities in the direction of intelligence, greenness, and sustainability.

(2)

In the field of transportation:

In the field of traffic and tunnel engineering, digital twin systems can achieve intelligent and optimized traffic management, from traffic flow monitoring and intelligent signal control to accident prevention and infrastructure management. Through real-time data collection and analysis, digital twins provide decision-makers with more accurate decision support, greatly improving the safety, efficiency, and sustainability of road traffic. This is conducive to the overall goal of “measurable, knowable, controllable and serviceable” in transportation and tunnel engineering. For example, the Helsinki Metro Line Extension Project in Finland has improved the efficiency of all stages of project planning, design, construction, and operation through the use of digital twin technology. The Zayed Port Infrastructure Project in Abu Dhabi has improved the design accuracy, construction efficiency, and operation management level of the project through digital twin technology. Daxing International Airport in Beijing has also widely adopted digital twin technology in all stages, from design and construction to operation, which not only improves the construction efficiency and quality of the project but also provides intelligent solutions for the long-term operation and maintenance of the airport.

Guowen Liu [27] used the AI digital twin monitoring linkage system to achieve all-weather and all-elements monitoring of expressway tunnels, completing real-time monitoring of traffic parameter data and traffic events in the tunnel, and synchronizing warnings with LED screens, smart cloud broadcasting, induction devices, lane control lights, etc., to achieve the safe operation supervision level and efficiency of vehicles in expressway tunnel sections.

Qingrong Liu et al. [28] established a set of expressway tunnel early warning platforms based on digital twin technology, AI, AR, VR, MR, and other technologies, which can be applied to different scenarios of expressway tunnels to achieve all-weather monitoring of expressway tunnels and solve the problem of safety hazards in expressway tunnels.

Zhaohui Wu et al. [29] proposed a digital twin architecture of highway traffic. Through this architecture, a digital twin platform for highway traffic supporting “state perception–digital experience-aided decision-optimization control” was established and applied to road engineering projects in Nanjing, Wuxi, Dalian, and other cities. Figure 3 shows the Digital Twin System Architecture of Highway Traffic.

Hehua Zhu’s team [30,31] proposed the concept of an intelligent tunnel construction system characterized by holographic perception, fusion modeling, intelligent analysis, closed-loop control, and continuous optimization learning based on digital twin technology, and they constructed the intelligent tunnel construction system (iS3), which can be used for integrated intelligent decision-making of data collection, transmission, processing, expression, analysis, and service throughout the lifecycle of the infrastructure. The intelligent construction system (iS3) was used for the support design of the Grand Canyon Tunnel of the Ehan Expressway. It only takes 10 min from the extraction of site information to the automatic feedback of the design plan to the on-site construction personnel, verifying the good applicability of the iS3 system. Figure 4 shows the intelligent tunnel construction system based on information flow.

Lei Shi et al. [32] carried out the application of “full-process and full-professional” BIM technology in the major overhaul project of the western runway at Beijing Capital Airport, and they explored the construction of the “digital twin” west runway, which provided a model and data foundation for the future operation of a “smart airport”.

Urumqi Airport has built a digital twin service platform based on GIS + BIM and other technologies. In the whole-lifecycle construction process of airport planning, design, construction, and operation, it meets the needs of multiple entities and units, such as builders and operators, and realizes the concept of “integration of smart airport construction and operation” [33].

In the field of transportation and tunnel engineering, the application of digital twin technology covers the whole process, from design and construction to operation and maintenance. Through real-time monitoring, data analytics, and virtual simulation, digital twins can provide intelligent support for transportation systems, reduce costs and risks, improve operational efficiency, and ensure the long-term safety and efficient operation of transportation facilities.

(3)

In the field of hydraulic engineering:

Digital twin technology in the field of hydraulic engineering is developing in multiple fields and dimensions, especially in water resources management, flood prediction, irrigation optimization, dam monitoring, etc. In order to achieve high-quality development of water conservancy and meet the requirements of intelligent management, it is necessary to establish a digital hydraulic model based on digital twin technology and construct a smart water resources system with capabilities for real-time monitoring, diagnosis, analysis, decision-making, forecasting, early warning, simulation, and contingency planning.

Hadzalic E et al. [34] innovatively applied digital twin technology to establish a multi-physical field-coupling monitoring system for the Salakovac concrete gravity dam. By integrating advanced numerical modeling and real-time monitoring data, the accurate evaluation of the structural performance of aging dams was achieved, including the dynamic prediction of key parameters such as temperature field distribution and displacement response, which provides reliable technical support for dam safety monitoring.

Baodong Lou et al. [5] constructed a digital smart water conservancy application solution for the whole process from model establishment, data transmission, and data analytics to end-point display based on digital twin technology, such as the digital twin of the Gezhouba Hydropower Station.

Tan Yaosheng et al. [35] developed a BIM-based full-process information model for hydropower slope construction and implemented it at the Baihetan Hydropower Station.

Yuanlin Deng et al. [36] proposed a digital twin-based holistic framework and data integration model for intelligent dam construction management through GIS-BIM data fusion, microservices architecture design, and big data analytics, with practical implementation cases.

Rui Xu et al. [37] established a 3D GIS and spatiotemporal data-driven visual safety monitoring system for hydraulic projects based on digital twin technology. This system enables intuitive visualization of terrain, geomorphology, safety monitoring models, and analytical results, while achieving 3D data simulation and visualization.

Ting Zhang et al. [38] developed an intelligent integrated simulation cloud platform for hydraulic engineering through customized development for industry-specific scenarios. Building upon the Tianhe Engineering Simulation Cloud Platform, they integrated cloud computing, big data, and AI technologies with field-measured and experimental data, addressing the growing cloud-based trends in engineering simulations and digital twin platforms.

Ding S.L. et al. [14] proposed a digital twin technology framework for safety monitoring and management of earth-rock dams. This study significantly improved the accuracy of dam deformation prediction by integrating Finite Element Modeling (FEM) with real-time monitoring data and a Bayesian updating method. The research team verified the effectiveness of the technology in the right-bank core-wall rockfill dam project in Danjiangkou, and they achieved three functions of real-time simulation, future prediction, and extreme scene deduction, which provided a dynamic and intelligent solution for the safety management of aging dams. This achievement not only promotes the transformation of dam monitoring from static–passive to dynamic–active but also provides an important technical reference for 94 digital twin dam projects in China.

Yuntao Ye et al. [39] defined the concept of “digital twin watershed” and established its fundamental model, comprising physical watershed, virtual watershed, real-time interaction interfaces, digital-enabled services, and twin watershed data/knowledge. Their work identified critical scientific challenges and technical frameworks for digital twin watershed implementation, providing valuable insights for smart watershed research and digital technology applications in basin governance. Figure 5 shows the basic framework of the digital twin watershed.

Digital twin technology plays a pivotal role across the design, construction, and operational phases of hydraulic engineering. Through real-time data integration and virtual model establishment, DT technology not only enhances management efficacy but also empowers hydraulic systems to better mitigate natural disasters, optimize resource allocation, improve ecological conservation, and advance the sector toward intelligent, refined, and sustainable development.

(4)

In the field of the mining industry:

Digital twin technology is progressively demonstrating transformative potential in mining operations, particularly in mineral extraction, resource management, equipment maintenance, safety monitoring, and environmental protection. With the deepening integration of intelligent mining and virtual reality technologies, DT-enabled unmanned precision mining and transparent extraction systems are emerging as frontier solutions.

Long Chen et al. [40] pioneered China’s first integrated, unmanned, and intelligent solution for open-pit mines—the Intelligent Mining Operating System (IMOS)—by synergizing smart mining philosophies, ACP parallel intelligence theory, and next-generation AI technologies, thereby proposing a unified framework for parallel mine management.

Dan Meng et al. [41] elaborated on DT’s applications in the digital transformation of mining, showcasing innovative implementations through case studies. These included 3D mine visualization, digital mining workflows, cloud-based mine surveying and mapping services, vehicle positioning and scheduling systems, production ore blending optimization, and slope stability monitoring

Enjie Ding et al. [7] established a smart mining service framework encompassing four critical dimensions: intelligent sensing and smart equipment, edge computing and network services, digital twin knowledge modeling, and platform and application systems. This architecture demonstrates the potential to achieve real-time measurability, precise control, and data-driven decision-making for physical mining environments, thereby laying a technological foundation for advancing intelligent mining systems. Figure 6 shows the service framework of an intelligent mine based on a digital twin.

Fan Zhang et al. [42] and Shirong Ge et al. [43] proposed a comprehensive framework for mining digital twins based on digital twin technology and parallel intelligence theory, including a conceptual framework, system architecture, core technologies, fundamental theories, and methodological systems. They further developed an evolutionary theoretical model of mining digital twins through the interactive coupling of physical models, simulation models, mechanistic models, and data models. This innovation achieves digital mirroring and model optimization of physical entities in intelligent, fully mechanized mining faces, enabling precise cyber–physical synchronization. Figure 7 shows an evolutionary example of a digital twin at the face of intelligent mining.

Unlike intelligent mining systems or 5G-enabled mines, digital twin-enabled mining operations exhibit three defining characteristics: bidirectional mapping, real-time interaction, and data-driven optimization. These features enable superior technological inclusivity, effectively overcoming limitations inherent to single-technology solutions. The emergence of digital twin mines has been catalyzed by breakthroughs in core technologies such as dynamic digital twinning, low-cost wireless communication systems, geological digital twin entities, etc. [44].

The application of digital twins in the mining sector is becoming a critical technology for enhancing mine safety, production efficiency, environmental protection, and resource management. Through deep integration of the virtual and physical domains, digital twins enable real-time monitoring, optimized decision-making, and predictive maintenance, thereby accelerating the industry’s transition toward intelligent, eco-friendly, and sustainable development.

(5)

In the field of petroleum engineering:

The application of digital twin technology in petroleum engineering is gaining increasing traction, demonstrating robust potential, particularly in oil exploration, extraction, production management, equipment maintenance, safety monitoring, and energy optimization. For instance, digital twins serve as the core technology for digital transformation and intelligent upgrades of offshore drilling platforms. By establishing digital twin entities of drilling platforms with sensing, analytical, and executive capabilities, this technology enables intelligent monitoring, diagnostics, and early warning safety systems for drilling platform equipment installations.

China National Offshore Oil Corporation (CNOOC) has developed an intelligent system for semi-submersible drilling platforms based on DT technology. This system achieves intelligent safety diagnostics, condition-based early warning for critical equipment, visualized safety control, etc. [45,46]. As illustrated in Figure 8, the drilling platform digital twin architecture comprises the physical drilling platform, virtual counterpart system, twin data repository, offshore platform O&M services, and interconnected subsystem networks.

Yuancheng Lin et al. [47] proposed a hybrid modeling approach combining data-driven and mechanism-driven methods, utilizing digital twin technology to achieve real-time optimization and predictive maintenance in energy systems.

The application of digital twin technology in petroleum engineering has enabled enterprises to enhance production efficiency, reduce operational costs, elevate safety standards, and achieve resource optimization with environmental protection across multiple operational phases. Looking ahead, the deep integration of digital twins with cutting-edge technologies such as artificial intelligence (AI), big data, and the Internet of Things (IoT) is poised to propel the transformation of petroleum engineering toward greater intelligence, greener practices, and sustainable development trajectories.

The advantages and disadvantages of digital twins in applications across various fields are shown in

Table 1. The characteristics of the applications of digital twins in geotechnical engineering across various fields.

Domain	Key Example	DT Benefits	Remaining Gap
Urban Construction	The four-dimensional geological model of Jinan’s Spring Area. The construction of a digital twin model in Xiong’an New Area.	Optimize planning and design. Visualization of construction process. Real-time monitoring of operation and maintenance stage.	Difficult to obtain urban geological data. Multi-source data fusion is hard. Model update is not timely.
Transportation	Digital twin system architecture of Ehan Expressway, Urumqi Airport	Geological risk pre-control. Accurate decision-making of operations and maintenance. Construction dynamic optimization.	High cost of data acquisition. Dynamic load simulation limitations.
Hydraulic Engineering	Digital twin of Gezhouba Hydropower Station and Baihetan Hydropower Station	Seepage–stress coupling analysis. Predictive maintenance decisions.	The computational complexity of hydraulic–rock coupling is high.
Mining Industry	Intelligent mine operating system of Wangjialing coal mine and Sanshandao gold mine.	Slope slip warning (multi-parameter fusion). Real-time optimization of mining scheme.	Difficult to realize three-dimensional geological–tectonic integration and real-time data assimilation.
Petroleum Engineering	Digital twin system of China petroleum oil and gas exploration platform.	Real-time correction of drilling trajectory. Wellbore stability prediction.	The acquisition cost of deep-sea geological data is high, and the update is delayed.

.

In summary, digital twin technology, through real-time data interaction between virtual models and the physical world, helps resolve persistent challenges in geotechnical engineering while enhancing engineering efficiency and safety. Consequently, geotechnical digital twin platforms are being actively developed across various domains. This trend stems from the technology’s demonstrated capacity to address the following critical issues in geotechnical engineering:

(1)

Real-Time Monitoring and Data Analytics

By integrating sensor networks (including seismic, pressure, and displacement sensors) with digital twin platforms, this approach enables continuous tracking of critical geotechnical parameters such as geological shifts, structural deformations, and subsurface water flow dynamics. The system facilitates rapid identification of geohazards (e.g., landslides, subsidence, surrounding rock instability) through predictive analytics-driven risk assessment, while implementing pre-failure alert mechanisms to substantially mitigate catastrophic engineering failures and associated economic losses.

(2)

Precision Modeling and Predictive Analytics

During the investigation phase, digital twin technology can integrate heterogeneous information such as geotechnical exploration data and seismic data from the field, optimize work deployment, and establish precise virtual models by combining technologies like Building Information Modeling (BIM), Geographic Information Systems (GISs), and Finite Element Modeling (FEM). These computational models facilitate comprehensive mechanical simulations and stability assessments under diverse operational scenarios, empowering engineers to conduct predictive modeling of potential geotechnical complications during the preliminary design phases. This capability drives preemptive design optimization, effectively eliminating latent safety risks that might otherwise manifest during subsequent construction or operational stages.

(3)

Design Optimization and Decision Support

During the design phase of geotechnical engineering, digital twin technology integrates field investigation data and laboratory test results with machine learning algorithms to conduct comparative multi-scenario analyses. This approach optimizes critical design parameters (e.g., support treatment schemes, material selection) through data-driven decision-making, enabling scientifically rigorous and precision-engineered solutions. Additionally, it leverages virtual reality environments to enable safe and efficient structural exploration, rapid design iteration, and construction process simulation. The technology significantly reduces design errors while concurrently enhancing construction feasibility and operational safety—particularly in high-risk geological conditions, where traditional design methods show limitations.

(4)

Real-Time Feedback and Process Adjustment

Through real-time synchronization of digital twin models with on-site construction data streams during geotechnical operations, this system implements continuous tracking of critical construction metrics, including progress rates, quality compliance indices, and deformation thresholds. The cyber–physical integration enables instantaneous anomaly detection through predictive analytics, allowing for the dynamic adjustment of construction methodologies to mitigate risks associated with non-conforming quality benchmarks or schedule deviations. This precise process control framework reduces unplanned operational interruptions while ensuring project continuity through proactive risk mitigation protocols.

(5)

Lifecycle Management and Maintenance

During the operations and maintenance (O&M) phase of geotechnical projects, digital twin technology enables comprehensive lifecycle management through continuous data assimilation and model recalibration. This framework facilitates predictive maintenance scheduling, structural integrity diagnostics, and emergency response protocols. By implementing condition-based maintenance strategies, the system can reduce maintenance expenditure, enhance operational efficiency, and extend infrastructure service life through proactive degradation mitigation. Crucially, it ensures the long-term stability and safety of geotechnical projects.

(6)

Post-Disaster Recovery and Safety Assessment

When geotechnical engineering projects confront disaster scenarios, digital twin technology executes high-resolution simulations of post-event geomechanical alterations and structural integrity degradation. This capability provides quantitative safety evaluations and real-time damage quantification, guides rapid damage assessment and rehabilitation planning, identifies latent hazards through scenario-based failure simulations, and develops high-efficiency emergency response and rehabilitation plans.

(7)

Multi-Stakeholder Collaboration

Digital twin technology enables unified integration of multi-stakeholder data (including designers, construction teams, and operators) to facilitate collaborative work. This approach significantly reduces communication overheads and minimizes information transfer errors through real-time data synchronization. By implementing a centralized digital thread architecture, the technology enhances cross-disciplinary coordination efficiency, effectively eliminating information silos and communication bottlenecks. The resulting collaborative environment ensures on-schedule, high-quality project delivery throughout the project lifecycle.

The key examples, DT benefits, and remaining gaps in each field of digital twin application in geotechnical engineering are summarized in Table 1. In conclusion, digital twin technology can provide more precise, real-time, and predictable tools for geotechnical engineering, thereby effectively helping to solve problems such as inaccurate design, construction risks, difficult operation and maintenance, and delayed disaster response in geotechnical engineering. The safety, economy, and sustainability of geotechnical engineering have been significantly improved by digital twin technology, making it an important development direction in the field of geotechnical engineering in the future.

3. Challenges in Geotechnical Digital Twin Development

Digital twin technology, as an inevitable outcome of advanced informatization, is emerging as a revolutionary tool for deconstructing, characterizing, and understanding the physical world.

Scholars worldwide have conducted extensive research on geotechnical DT systems, including DT frameworks [18,48,49,50], 3D geological modeling [51,52,53], BIM [54,55,56], simulation and geotechnical numerical analysis [57,58], geotechnical monitoring and sensing [59,60,61], etc. For instance, Jiaming Wu et al. [48] developed a BIM-based data integration and sharing mechanism for geotechnical DT systems. Their work established an integrated geological–structural model and created a simulation analysis module, forming a preliminary geotechnical DT framework.

The implementation of geotechnical digital twins relies on both commercial and open-source platforms. Industry solutions like ANSYS Twin Builder (for system simulation), Dassault 3DEXPERIENCE (BIM integration), and Siemens Simcenter (IoT data fusion) provide robust toolchains, while emerging open frameworks like OpenTWIN offer modular architectures for customized deployments. Notably, IFC-Geo standard extensions (e.g., IFC-Soil and IFC-Tunnel) are bridging BIM gaps in geotechnical data interoperability.

However, there remain significant challenges to be addressed in the practical implementation of digital twins in geotechnical engineering. Current research exhibits notable deficiencies in several critical areas, including intelligent perception of physical entities, integration of multi-source heterogeneous data, construction of multidimensional information models, integrated application of geological body models and structural body models, standardization of geotechnical digital twins, and whole-lifecycle management. These limitations are primarily manifested in the following aspects:

(1)

Intelligent perception and interconnectivity of multi-source heterogeneous physical entities in geotechnical engineering remain weak:

Currently, the sensing of various geotechnical parameters operates independently and in isolation, making comprehensive processing highly challenging due to the absence of integrated intelligent sensing devices. Geotechnical engineering involves massive multidimensional heterogeneous data that vary in format across multi-scale, multi-temporal, and multi-scenario dimensions, posing significant challenges for data fusion. There exists a critical lack of unified interface support and highly standardized, modularized theories/methods for comprehensive data–model integration.

(2)

Integrated sharing of 3D geological and structural models faces substantial obstacles and reliability problems:

Digital twins in geotechnical engineering comprise geological and structural models, which exhibit fundamental disparities in organizational frameworks. The absence of a unified spatial data model and data integration standards results in poor compatibility and consistency. Difficulties in data integration and information sharing between 3D geological models and structural models directly compromise geotechnical engineering analysis and decision-making. There are also scalability issues when combining high-fidelity geological models with structural simulations (e.g., soil–structure interface grid complexity). Secondly, the existing constitutive models (such as the Mohr–Coulomb model and Hoek–Brown model) cannot capture the time-dependent behavior of soft soil creep. Finally, the digital twin component based on AI/ML has the characteristics of a “black box” and lacks interpretability in key scenarios such as landslide triggering. There is also an unresolved propagation problem of model uncertainty in DT-based predictions, such as the effect of soil heterogeneity on slope stability simulations. These will seriously affect the reliability of digital twins.

(3)

There remains a conspicuous absence of a unified information management platform comprehensively covering the geotechnical engineering lifecycle:

Currently, numerous information platforms have emerged in geotechnical engineering, primarily serving critical functions including visualization, construction safety assurance and quality control, resource allocation and cost management, monitoring data integration, and construction progress tracking. These platforms aim to fulfill diverse requirements such as design communication, project demonstration, and construction simulation. However, most existing platforms are confined to specific project phases, failing to achieve comprehensive lifecycle management. Their capabilities in design optimization refinement and on-site construction guidance require further enhancement and development.

(4)

Standardization frameworks for geotechnical digital twin data remain critically underdeveloped:

As geotechnical digital twins are still in their nascent stage, there exists an urgent need to establish a comprehensive standard system encompassing six key dimensions: fundamental common standards, core technical specifications, tool/platform requirements, evaluation metrics, security protocols, and application guidelines. Each category demands intensified research efforts to ensure systematic development.

(5)

A universally recognized theoretical framework for geotechnical digital twin modeling is notably lacking:

Given the cross-disciplinary nature of digital twin technology and its inherent integration complexity, current implementations suffer from insufficient interdisciplinary collaboration and the absence of a generalized technical infrastructure. For example, there are some problems in geotechnical engineering: incompatibility between BIM (Building Information Modeling) and geotechnical data formats, e.g., IFC vs. GeoSciML; lack of protocols for reconciling multi-scale data, e.g., integration of surface models derived from LiDAR with borehole logs; and ethical and legal implications of data ownership in multilateral DT projects. The exploration of geotechnical digital twins remains in its preliminary stages, necessitating immediate establishment of consensus-driven standards and theoretical systems to provide coherent technical direction and implementation guidance.

4. Future Research Perspectives for Geotechnical Digital Twins

(1)

Advancing Intelligent Perception and Interconnectivity Technologies in Geotechnical Engineering

Continuous advancement in sensing and data acquisition technologies remains the cornerstone for driving the evolution of digital twins. Future development trajectories should prioritize sensor miniaturization to enable embedded deployment in intelligent systems, thereby achieving deeper data acquisition capabilities. Concurrently, advancing multi-sensor fusion technologies to integrate diverse sensing functionalities into unified modules will enhance the timeliness and accuracy of geotechnical decision-making.

Adopting reverse digital thread technologies, such as Administration Shell (AAS) frameworks, can facilitate standardized semantic recognition, definition, and validation of existing data/models by leveraging multi-domain engineering integration standards. The development of unified interfaces for data–information interoperability will promote cross-domain compatibility and seamless interaction among multi-source heterogeneous models, enabling holistic information exchange across systems.

(2)

Developing Integrated Geotechnical Digital Twin Technologies for 3D Geological–Structural Systems

Given the inherent disparities in modeling methodologies between 3D geological bodies and engineered structural systems, the precise construction of geotechnical digital twins necessitates advanced “modular assembly” techniques. The application of cross-disciplinary, cross-domain, and cross-scale model convergence technologies is imperative.

While significant progress has been made in meso-scale and macro-mechanistic modeling for geotechnical systems, effective integration of multi-scale models remains elusive. Breakthroughs in macro/meso cross-scale model fusion would substantially elevate digital twins’ capabilities, advancing fundamental geotechnical mechanism studies. Furthermore, such integration enables collaborative simulation frameworks for design–construction synergy, supporting dynamic model reconfiguration, process emulation, and predictive evaluation. These capabilities allow for the anticipatory analysis of feedback mechanisms and evolutionary trends between geological and structural systems, thereby guiding construction practices and achieving integrated lifecycle management spanning the planning, design, construction, and operation phases.

(3)

Advancing Geotechnical Knowledge Graphs and Model-Based Systems Engineering (MBSE) Technologies

Leveraging the distinctive characteristics of geotechnical engineering, Model-Driven Systems Engineering (MDSE) should be adopted as a paradigm. At the initial stages of user requirement analysis, the Unified Modeling Language (UML) should be systematically employed to precisely define data specifications, model standards, and topological relationships among model objects. This approach facilitates semantic data representation and architectural framework development for geotechnical digital twins, laying a robust foundation for comprehensive data–model integration throughout the lifecycle.

Accelerating integration with industrial Internet platforms is crucial to establish a “Geotechnical Internet Platform + MBSE” technical ecosystem. Migrating MBSE tools to unified platforms serves dual purposes: (1) standardizing syntax/semantics across heterogeneous models through MBSE tools, and (2) enabling seamless integration with IoT data collected by platforms. This synergy maximizes the application value of data–model integration, empowering geotechnical digital twins with advanced capabilities, including data management, model representation, simulation, scenario projection, intelligent prediction, and autonomous decision-making.

(4)

Enhancing the Precision, Temporal Application, and Spatial Expansion of Geotechnical Digital Twins

Digital twins’ precision can be categorized into four hierarchical levels: basic descriptive, general diagnostic, intelligent decision-making, and autonomous control. Current applications predominantly reside at the basic descriptive and general diagnostic levels, collectively accounting for 71% of implementations [62]. Intelligent decision-making applications remain relatively scarce, while autonomous control applications are minimal. Geotechnical digital twins currently concentrate on the first two levels, with significant progress required to achieve advanced capabilities.

Furthermore, lifecycle optimization of geotechnical digital twins warrants extensive exploration to enable comprehensive lifecycle management and extend their temporal application spans. Regarding spatial expansion, applications primarily encompass two categories: same-scale twin object coordination, and cross-scale twin object coordination. In geotechnical contexts, research on group tunnel excavation effects exemplifies same-scale twin challenges, while optimization of excavation–support–monitoring systems represents cross-scale coordination problems, both requiring in-depth investigation.

5. Conclusions

In this era of universal datafication, digital twin technology has unveiled a transformative paradigm, progressively blurring the boundaries between the physical world and its data-driven virtual counterpart. As technological advancements continue to mature digital twins’ capabilities, the realization of digitalization and intelligentization in geotechnical engineering appears inevitable.

Historically, the genesis of digital twins can be traced to innovations in modeling and simulation technologies, while their exponential growth has been propelled by breakthroughs in sensing technologies. With the collective advancement and deep integration of next-generation information technologies, digital twins are poised to enter an expansive phase of development. The transition towards digitalization and informatization has become an inescapable imperative across industries.

As an integral component of new infrastructure development, geotechnical engineering has found that digital twin technology is to be an indispensable pathway towards intelligentization and digital transformation. However, the core challenges currently faced by geotechnical engineering digital twins include the following: the strong heterogeneity and invisibility of underground environments make high-quality data acquisition, installation, and validation difficult; the highly complex, multi-physics, and nonlinear behavior of geotechnical materials poses challenges in constructing, computing, and validating high-precision physical models; there is a contradiction between the information silos of multi-source heterogeneous systems and the seamless integration required by real-world demands; methods for data–model fusion and uncertainty quantification/management remain immature; the inherent conflict between real-time requirements and the massive computational power needed for high-fidelity models cannot be ignored; and cost–benefit models and clear definitions of engineering application scenarios are still under exploration and development. Despite these significant challenges, addressing them is essential for advancing geotechnical engineering toward intelligence and precision. With progress in sensor technology, edge computing, AI, cloud computing, and industry standards, these bottlenecks are expected to be gradually overcome.

This paper has systematically reviewed the current state of digital twin applications in geotechnical engineering in China, elucidated existing challenges, and outlined future research directions. It is hoped that this work will serve as a catalyst for further advancements in geotechnical digital twin technologies, stimulating more in-depth research and innovation in this field.