NGI /
Artificial intelligence at NGI

Artificial intelligence at NGI

At NGI, we see great potential in harnessing the power of machine learning (ML) to tackle complex geotechnical challenges.

Our expertise spans a broad spectrum of projects, including reinforcement learning for rock tunnelling, predicting rock mass parameters from drilling data, ML-driven simulations of earthquake-induced landslides, image-based rock mass classification, and particle reconstruction from rock samples. By integrating advanced ML techniques, we strive to enhance efficiency, gain deeper insights, improve predictive accuracy, and optimise decision-making in geotechnical engineering. Our approach ensures AI is applied where it delivers tangible value, complementing traditional engineering methods with data-driven intelligence.

Figure 1. Predicting landslide locations in Kvam (Liu et al., 2021)

In NGI, we work with models in all branches of machine learning, from supervised to unsupervised and reinforcement learning.

Figure 2. Branches of machine learning.

Machine learning courses in NGI Code Academy

The NGI Code Academy is NGI’s educational program that offers a range of courses to help geotechnicians, geologists, and other geoscience professionals in the ongoing digitalisation. The NGI Code Academy’s motto is “Geoscience experts empowering geoscience experts to achieve more with coding and modern technologies”. Two of the courses in NGI Code Academy deal with machine learning:

Contact information

At NGI, we continue to explore and refine ML applications to enhance geotechnical engineering, ensuring that our research and industrial projects benefit from the latest advancements in artificial intelligence. If you want to learn more about our work, attend our ML courses, write a master thesis with us, or explore potential collaborations, please get in touch with us. See contact info at the bottom of this page.