Reducing Project Uncertainty: Real-Time Backanalysis with Monitoring Data

Predicting ground behavior is fundamental to geotechnical design. However, even with extensive site investigations, a high degree of uncertainty persists due to natural subsurface heterogeneity and the limitations of conventional testing methods. In this context, real-time backanalysis, based on monitoring data integrated with numerical models, has emerged as a technically advanced solution that dynamically calibrates geotechnical parameters and reduces uncertainty. This article explores this methodology from an applied engineering perspective, outlining its theoretical foundations, the limitations of traditional practices, and the added value of DAARWIN, developed by SAALG Geomechanics.

Limitations of Conventional Ground Characterization

Classical geotechnical design relies heavily on laboratory and in-situ testing. While these approaches offer essential insights, they present well-known limitations:

• Sample disturbance in triaxial or oedometer tests compromises the reliability of measured properties.

• Spatial variability is poorly captured by point-based investigations, particularly in stratified or residual soils.

• Inability to reproduce actual stress paths, especially in projects involving sequential or transient construction phases.
  

These limitations lead to simplified models that often require conservative safety factors, contributing to potential overdesign and inefficiencies.

Fundamentals of Real-Time Backanalysis

Backanalysis, or inverse analysis, involves adjusting model parameters to minimize the difference between numerical predictions and actual field measurements. When this process is implemented in an automated, iterative, real-time fashion, it becomes a powerful tool during construction.

Automated Inverse Modeling

The process is based on three main components:

• Numerical models (FEM) to simulate geotechnical and structural behavior.

• Monitoring data (I&M) from instruments such as inclinometers, piezometers, load cells, and extensometers.

• Optimization algorithms, such as genetic algorithms, capable of exploring large parameter spaces without requiring continuity or derivatives.
  

High-Performance Computational Architecture

DAARWIN executes this methodology on a parallel cloud-computing infrastructure, enabling hundreds or thousands of FEM simulations to run simultaneously. This transforms a manual, time-consuming process into an automated loop with frequent updates.

Application in Geotechnical Engineering with DAARWIN

The DAARWIN platform integrates PLAXIS models and monitoring data into a continuous analysis workflow that enables:

• Dynamic parameter calibration based on real observations.

• Comparative visualization between model predictions and field data.

• Full integration of the Observational Method, supporting model updates as construction progresses.
  

Design Validation and Control

Inverse analysis allows engineers to validate or revise initial conceptual models. In projects with unexpected stiffness changes or pore pressure responses, DAARWIN enables early detection and geotechnical reinterpretation.

Use Case: Urban Tunnels

In TBM and conventional excavations, real-time backanalysis enables:

• Estimation of local mechanical properties in transitional zones.

• Validation of structural deformation models.

• Real-time optimization of support systems.

Technical and Economic Impact

From a geotechnical risk management perspective, DAARWIN provides:

• Reduction in construction deviations (e.g., time delays, unanticipated deformations).

• Design optimization through better-informed parameter selection.

• Minimization of overdesign, resulting in material and cost savings.

• Creation of a digital knowledge base for future projects in similar conditions.
  

By centralizing historical data, test results, monitoring, and models, DAARWIN supports full data traceability and promotes evidence-based engineering.

Real-time backanalysis with monitoring data marks a paradigm shift in geotechnical practice. Beyond being a calibration tool, it becomes a critical component of the design–construction cycle, enabling uncertainty reduction, hypothesis validation, and informed decision-making in critical project stages.

DAARWIN offers a robust, scalable, and forward-looking solution aligned with the demands of modern infrastructure: efficiency, sustainability, and safety.

🔗 Learn how to implement real-time backanalysis in your project: 👉 https://www.saalg.com/real-time-backanalysis