Search

Internet of Things-based real-time slope stability forecast using a digital twin and conversational AI

This project develops an operational, real-time framework for slope stability forecasting by combining digital twins, IoT-based monitoring, numerical modelling, and data-driven methods. By integrating local and regional scales within a cloud-based system, the work supports more reliable landslide early warning under changing climatic conditions.

Submenu

About the project

The research activities have been carried out within the EU Project The HuT - The Human-Tech Nexus - Building a Safe Haven to cope with Climate Extremes (ID101073957, https://thehut-nexus.eu/), which played a crucial role in facilitating and advancing our research.

Landslides affecting natural and engineered slopes pose a growing challenge for disaster risk reduction, particularly under the increasing frequency and intensity of rainfall and snowmelt events driven by climate change. Accurate and operational slope stability forecasting requires the integration of meteo-hydro-geological data sources, physical understanding of failure mechanisms, and scalable computational frameworks capable of delivering timely predictions. Our research activities aim at creating an integrated real-time cloud-based operational framework that combines slope-and regional-scales digital twins for landslide forecasting, leveraging real-time monitoring, numerical modelling, data-driven methods and conversational AI.

By integrating detailed slope-scale digital twins with scalable regional-scale forecasting, our work illustrates a coherent multi-scale approach to landslide prediction. The proposed framework is readily transferable to other sites and regions, offering a practical pathway for enhancing real-time landslide early warning and risk management.

 

Local scale

Slope-scale forecasting is achieved through the development of a digital twin of an instrumented slope in Norway (Piciullo et al., 2022; Piciullo et al., 2025). The digital twin integrates real-time monitoring of hydrological variables, such as volumetric water content and pore water pressure, with numerical slope stability modelling and machine learning. The numerical model is continuously validated against monitored data and used to calculate the FoS. To enable efficient operational forecasting, data-driven models, including Polynomial Regression and Random Forest, are trained on simulated FoS values, monitored hydrological conditions, and meteorological inputs to forecast the rolling three days FoS. These data-driven models replace the computationally intensive numerical model within the cloud service, enabling rapid and reliable FoS forecasts. A performance evaluation demonstrates that the data-driven surrogates provide accurate and robust FoS predictions comparable to the numerical model, highlighting their suitability for operational early warning applications.

Regional scale

At the regional scale, slope stability forecasting is addressed through a hybrid methodology that merges physically-based infinite slope models with data-driven landslide susceptibility and probability models (Abraham et al., 2025). The regional framework operates across multiple first-order catchments within a selected study area in Norway. A physically-based model computes pixel-wise Factor of Safety (FoS) values using precipitation, topography, and subsurface parameters, calibrated through back-analysis and applied in forward forecasting model. In parallel, a machine learning data-driven model estimates the probability of landslide occurrence at the catchment scale. Both model types are deployed as automated cloud services that generate daily forecasts, overcoming key operational challenges related to model integration, parameter updating, and large-scale data handling. Forecast outputs are disseminated through NGI Live, the Norwegian Geotechnical Institute’s data platform, supporting Landslide Early Warning Systems (LEWS).

Additionally, the developed Three-Dimensional Translational Slide (3DTS), a physically-based model that combines rainfall infiltration and 3D slope stability analysis, can simulate regional-scale rainfall-induced shallow landslides while accounting for the following: terrain complexity (curving or turning surfaces), vegetation root reinforcement, soil side resistance, non-uniform rainfall history, and spatially varying hydro-geotechnical properties. Given the computational efficiency of the 3DTS model relative to traditional 3D slope stability models, 3DTS is suitable for reliability-based probabilistic analysis for regional scale. The RALP software provides a graphical user interface (GUI), modelling tools, and interactive animation plots; hence, enabling engineers to use the 3DTS model with ease.  
 

Conversational AI

The growing availability of IoT-enabled sensor networks has transformed how geotechnical systems are monitored, producing vast datasets on pore pressures, groundwater levels, displacements, and external drivers such as rainfall. In the literature there is an increased trend to use advanced data analysis approaches and surrogate models for geotechnical engineering tasks. However, the sheer volume and complexity of these data often limit users’ ability to interact with them in a flexible and intuitive way.

Emerging advances in generative AI models and agentic frameworks suggest a new paradigm: chat-with-your-data. In this approach, users interact directly with monitoring data through natural language, requesting tailored visualizations, summaries, analysis or predictions without needing coding or rigid workflows. The potential benefits include a more direct access to data-centric analyses and increased use of monitoring data for tailored decision-making. Within this project we aim to prototype agentic frameworks built on generative AI models that provide both intuitive and reliable insight into monitoring data. While slope monitoring is selected as one illustrative use case, the broader framework promises impact across many geotechnical areas. 

Publications

2022 - Piciullo, L., Capobianco, V., and Heyerdahl, H. (2022). A first step towards a IoT-based local early warning system for an unsaturated slope in Norway. Nat. Hazards 114. https://doi.org/10.1007/s11069-022-05524-3. 

2025- Piciullo, L., Abraham, M. T., Drøsdal, I. N., and Paulsen, E. S. (2025). An operational IoT-based slope stability forecast using a digital twin. Environ. Model. Softw. 183, 106228. https://doi.org/10.1016/j.envsoft.2024.106228.

2025 - Abraham, M. T., Piciullo, L., Liu, Z., Drøsdal, I. N., Robinson, H., Rudjord, Z. C., Albright Blomberg, A. E., Maio, E. C., Ribeiro, W. N., and de Mendonça, M. B. (2025). Operational regional scale landslide forecasts: Physics-based and data-driven models. Proceedings of the 9th International Symposium on Geotechnical Safety and Risk (ISGSR 2025). Research Publishing, Singapore. https://doi.org/10.3850/981-973-0000-00-0-isgsr2025-paper.

2025 - Cheon, E., Ahmet Oguz, E., DiBiagio, A., Piciullo, L., Kwon, T. H., and Lee, S. R. (2025). A Physically-based 3D Landslide Susceptibility Model for Shallow Translational Landslides using DEM, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17059, https://doi.org/10.5194/egusphere-egu25-17059  

2025 – Cheon, E., Oguz, E. A., DiBiagio, A., and Piciullo, L. (2025). A 3D Shallow Translational Landslide Susceptibility Model with DEM Cells Accounting for Side Resistance and Vegetation Effects. The 9th International Symposium for Geotechnical Safety and Risk (ISGSR) at Oslo, Norway. https://doi.org/10.3850/GRF-25280825_isgsr-007-P354-cd 

 

The team

Luca PICIULLO – NGI PM
luca.piciullo@ngi.no

Minu Treesa ABRAHAM – Slope stability model, ML
minu.treesa.abraham@ngi.no

Vittoria CAPOBIANCO – Slope stability model
vittoria.capobianco@ngi.no

Andreas-Nizar GRANITZER – AI conversational model
andreas.nizar.granitzer@ngi.no

Andreas MATHISEN – AI conversational model
Andreas.mathisen@ngi.no 

Enok CHEON – Physically-based model
enok.cheon@ngi.no

Ida Norderhaug DRØSDAL – NGI live implementation
ida.norderhaug.drosdal@ngi.no

  • About the project

    The research activities have been carried out within the EU Project The HuT - The Human-Tech Nexus - Building a Safe Haven to cope with Climate Extremes (ID101073957, https://thehut-nexus.eu/), which played a crucial role in facilitating and advancing our research.

    Landslides affecting natural and engineered slopes pose a growing challenge for disaster risk reduction, particularly under the increasing frequency and intensity of rainfall and snowmelt events driven by climate change. Accurate and operational slope stability forecasting requires the integration of meteo-hydro-geological data sources, physical understanding of failure mechanisms, and scalable computational frameworks capable of delivering timely predictions. Our research activities aim at creating an integrated real-time cloud-based operational framework that combines slope-and regional-scales digital twins for landslide forecasting, leveraging real-time monitoring, numerical modelling, data-driven methods and conversational AI.

    By integrating detailed slope-scale digital twins with scalable regional-scale forecasting, our work illustrates a coherent multi-scale approach to landslide prediction. The proposed framework is readily transferable to other sites and regions, offering a practical pathway for enhancing real-time landslide early warning and risk management.

     

  • Local scale

    Slope-scale forecasting is achieved through the development of a digital twin of an instrumented slope in Norway (Piciullo et al., 2022; Piciullo et al., 2025). The digital twin integrates real-time monitoring of hydrological variables, such as volumetric water content and pore water pressure, with numerical slope stability modelling and machine learning. The numerical model is continuously validated against monitored data and used to calculate the FoS. To enable efficient operational forecasting, data-driven models, including Polynomial Regression and Random Forest, are trained on simulated FoS values, monitored hydrological conditions, and meteorological inputs to forecast the rolling three days FoS. These data-driven models replace the computationally intensive numerical model within the cloud service, enabling rapid and reliable FoS forecasts. A performance evaluation demonstrates that the data-driven surrogates provide accurate and robust FoS predictions comparable to the numerical model, highlighting their suitability for operational early warning applications.

    Regional scale

    At the regional scale, slope stability forecasting is addressed through a hybrid methodology that merges physically-based infinite slope models with data-driven landslide susceptibility and probability models (Abraham et al., 2025). The regional framework operates across multiple first-order catchments within a selected study area in Norway. A physically-based model computes pixel-wise Factor of Safety (FoS) values using precipitation, topography, and subsurface parameters, calibrated through back-analysis and applied in forward forecasting model. In parallel, a machine learning data-driven model estimates the probability of landslide occurrence at the catchment scale. Both model types are deployed as automated cloud services that generate daily forecasts, overcoming key operational challenges related to model integration, parameter updating, and large-scale data handling. Forecast outputs are disseminated through NGI Live, the Norwegian Geotechnical Institute’s data platform, supporting Landslide Early Warning Systems (LEWS).

    Additionally, the developed Three-Dimensional Translational Slide (3DTS), a physically-based model that combines rainfall infiltration and 3D slope stability analysis, can simulate regional-scale rainfall-induced shallow landslides while accounting for the following: terrain complexity (curving or turning surfaces), vegetation root reinforcement, soil side resistance, non-uniform rainfall history, and spatially varying hydro-geotechnical properties. Given the computational efficiency of the 3DTS model relative to traditional 3D slope stability models, 3DTS is suitable for reliability-based probabilistic analysis for regional scale. The RALP software provides a graphical user interface (GUI), modelling tools, and interactive animation plots; hence, enabling engineers to use the 3DTS model with ease.  
     

  • Conversational AI

    The growing availability of IoT-enabled sensor networks has transformed how geotechnical systems are monitored, producing vast datasets on pore pressures, groundwater levels, displacements, and external drivers such as rainfall. In the literature there is an increased trend to use advanced data analysis approaches and surrogate models for geotechnical engineering tasks. However, the sheer volume and complexity of these data often limit users’ ability to interact with them in a flexible and intuitive way.

    Emerging advances in generative AI models and agentic frameworks suggest a new paradigm: chat-with-your-data. In this approach, users interact directly with monitoring data through natural language, requesting tailored visualizations, summaries, analysis or predictions without needing coding or rigid workflows. The potential benefits include a more direct access to data-centric analyses and increased use of monitoring data for tailored decision-making. Within this project we aim to prototype agentic frameworks built on generative AI models that provide both intuitive and reliable insight into monitoring data. While slope monitoring is selected as one illustrative use case, the broader framework promises impact across many geotechnical areas. 

  • Publications

    2022 - Piciullo, L., Capobianco, V., and Heyerdahl, H. (2022). A first step towards a IoT-based local early warning system for an unsaturated slope in Norway. Nat. Hazards 114. https://doi.org/10.1007/s11069-022-05524-3. 

    2025- Piciullo, L., Abraham, M. T., Drøsdal, I. N., and Paulsen, E. S. (2025). An operational IoT-based slope stability forecast using a digital twin. Environ. Model. Softw. 183, 106228. https://doi.org/10.1016/j.envsoft.2024.106228.

    2025 - Abraham, M. T., Piciullo, L., Liu, Z., Drøsdal, I. N., Robinson, H., Rudjord, Z. C., Albright Blomberg, A. E., Maio, E. C., Ribeiro, W. N., and de Mendonça, M. B. (2025). Operational regional scale landslide forecasts: Physics-based and data-driven models. Proceedings of the 9th International Symposium on Geotechnical Safety and Risk (ISGSR 2025). Research Publishing, Singapore. https://doi.org/10.3850/981-973-0000-00-0-isgsr2025-paper.

    2025 - Cheon, E., Ahmet Oguz, E., DiBiagio, A., Piciullo, L., Kwon, T. H., and Lee, S. R. (2025). A Physically-based 3D Landslide Susceptibility Model for Shallow Translational Landslides using DEM, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-17059, https://doi.org/10.5194/egusphere-egu25-17059  

    2025 – Cheon, E., Oguz, E. A., DiBiagio, A., and Piciullo, L. (2025). A 3D Shallow Translational Landslide Susceptibility Model with DEM Cells Accounting for Side Resistance and Vegetation Effects. The 9th International Symposium for Geotechnical Safety and Risk (ISGSR) at Oslo, Norway. https://doi.org/10.3850/GRF-25280825_isgsr-007-P354-cd 

     

  • The team

    Luca PICIULLO – NGI PM
    luca.piciullo@ngi.no

    Minu Treesa ABRAHAM – Slope stability model, ML
    minu.treesa.abraham@ngi.no

    Vittoria CAPOBIANCO – Slope stability model
    vittoria.capobianco@ngi.no

    Andreas-Nizar GRANITZER – AI conversational model
    andreas.nizar.granitzer@ngi.no

    Andreas MATHISEN – AI conversational model
    Andreas.mathisen@ngi.no 

    Enok CHEON – Physically-based model
    enok.cheon@ngi.no

    Ida Norderhaug DRØSDAL – NGI live implementation
    ida.norderhaug.drosdal@ngi.no

Portrait of Luca Piciullo

Luca Piciullo

Lead Engineer / Lead Researcher Dam Safety luca.piciullo@ngi.no
 +47 467 42 144