The Lugeon Test: Methodology, Interpretation, and Applications in Geotechnical Engineering

 A digital twin is a digital replica of a physical object, person, system, or process, contextualized in a digital version of its environment. It is a virtual replica of a physical object, person, or process that can be used to simulate its behavior to better understand how it works in real life.

Digital twins are linked to real-time data sources from their physical counterparts, allowing them to update continuously and reflect the original version. They also include behavioral insights and visualizations derived from this data. When interconnected, digital twins can create an immersive digital environment that replicates and connects various aspects of an organization to optimize simulations, scenario planning, and decision-making. The combination of AI, IoT, and real-time data enables the creation of digital twins that evolve with actual conditions.

Here's a detailed overview of digital twins:

• How they work

  • Digital twins are continuously updated by AI with sensor data, which enables proactive maintenance and risk mitigation.
  • They are used to simulate real-world conditions and predict how structures will respond to stress, environmental changes, or aging.
  • AI allows for real-time analysis and simulation of different scenarios within the digital twin.
  • They take data captured from all facets of an organization’s operations and model this data to mimic physical assets, people, and processes.
  • A digital twin of a facility can visually integrate all integrity data, reports, risk matrix, and anomaly assessments into a 3D model, translating this information into heatmaps for better understanding of asset conditions.

• Types of Digital Twins

  • Product twin: A representation of a product, covering various stages of its lifecycle from initial concept design and engineering through full functionality, providing live, real-time data on the product in service.
  • Data twin: A digital twin that provides visibility into operations, profitability, and customer touchpoints, improving decision-making with real-time insights into areas like in-transit inventory, customer journeys, and staffing. An example is Google Maps, which is considered a digital twin of the Earth's surface, linking real-time traffic data to optimize commutes. Data twins can also help organizations comply with regulations requiring upstream visibility into sourcing.
  • Systems twins: These models simulate the interaction between physical and digital processes, including manufacturing, end-to-end supply chain management, store operations, and customer journeys.
  • Infrastructure twins: These represent physical infrastructure assets such as highways, buildings, or stadiums.
  • Customer digital twins: These allow customers to fully interact and immerse themselves within a company’s product, potentially leading to significant revenue increases.

• Benefits and Value

  • Enhanced Safety: By simulating outcomes and predicting behaviors, digital twins help ensure the safety and longevity of critical structures.
  • Reduced Costs: They optimize maintenance schedules and allow for predictive maintenance, significantly reducing unnecessary inspections, interventions, and costly emergency repairs. For example, the Hong Kong-Zhuhai-Macau Bridge saw maintenance costs drop by 15-20% due to its AI system, and The Shard achieved a 30% reduction in maintenance costs.
  • Extended Lifespan: Proactive maintenance based on digital twin insights helps address issues before they escalate, extending the operational life of valuable infrastructure.
  • Improved Decision-Making: Digital twins transform raw data into actionable insights, empowering engineers and policymakers to make informed decisions about infrastructure investments, upgrades, and repairs. They can increase decision-making speed by up to 90%.
  • Operational Efficiency and Resilience: They enable more agile and resilient operations, reducing transportation costs and labor by as much as 10% and increasing consumer promise by up to 20%.
  • Faster Product Development: Product digital twins can cut development times by up to 50% by allowing rapid iterations and optimizations of designs, far faster than physically testing every prototype.
  • Sustainability Improvements: They help organizations reduce material use, improve product traceability to lessen environmental waste (e.g., consumer electronics manufacturers reduced scrap waste by approximately 20%), and balance cost/speed with sustainability goals.

• Applications and Case Studies

  • In North America, digital twins are used for bridges, skyscrapers, and dams to simulate the effects of load changes, wear, or environmental factors over time, optimizing maintenance schedules.
  • Deutsche Bahn leverages smart sensors and machine learning for predictive maintenance, which contributes to an enhanced understanding of critical components and need-based maintenance, hinting at digital twin concepts.
  • Emirates Team New Zealand uses a digital twin of sailing environments, boats, and crew members to test thousands of boat designs without physically building them.
  • Anheuser-Busch InBev employs a brewing and supply chain digital twin to adjust inputs based on active conditions and automatically compensate for production bottlenecks.
  • SoFi Stadium utilizes a digital twin to aggregate multiple data sources for optimizing stadium management and operations.
  • The Space Force is creating a digital twin of space, including replicas of extraterrestrial bodies and satellites.
  • SpaceX uses a digital twin of its Dragon capsule spacecraft to monitor and adjust trajectories, loads, and propulsion systems, aiming to maximize safety and reliability during transport.
  • Mercedes-Benz (formerly Daimler) has developed customer twins that allow customers to "test drive" a vehicle virtually.

• Challenges

  • Implementing digital twins requires a high-quality data infrastructure that delivers reliable data from both testing and live environments.
  • It also necessitates new ways of working within R&D functions and beyond, making it a significant change management effort that requires senior management support and a strong project management team.

• Future Outlook

  • The future of AI-driven Structural Health Monitoring (SHM) will involve the integration of digital twins, where a virtual replica of the structure is continuously updated with real-time AI-analyzed data. This will allow for predictive simulations of damage propagation and future performance.
  • Digital twins can be used symbiotically with generative AI (gen AI), where gen AI can structure inputs and synthesize outputs for digital twins, and digital twins can provide a robust test-and-learn environment for gen AI. Large Language Models (LLMs) can create code for digital twin prototypes and process/transfer the data they need to perform.