In geotechnics, CPTs are used at discrete locations to gain insight into the soil composition. When visualizing a cross-section between these CPTs, the challenge arises to create a realistic interpolation of the stratigraphy. Classical methods, such as linear interpolation, take limited account of geological patterns or discontinuities.
By using machine learning on site investigation data, we can significantly reduce the time required for geotechnical assessments, improve accuracy, and make better use of existing data. This project will contribute to Witteveen+Bos’ ongoing innovation efforts in the field of geotechnical engineering.
During your assignment, you will have the opportunity to:
- Be part of a team of geotechnical engineers and data scientists.
- Learn about a wide range of projects at Witteveen+Bos.
- Participate in social and professional development events.
- Receive an internship allowance.
- Gain firsthand experience in machine learning applications within geotechnical engineering.
This project will involve working closely with geotechnical engineers to understand the necessary soil parameters and data requirements. You will also collaborate with data scientists to develop machine learning models and validate them with field data.
Objective
In this thesis assignment, you will investigate whether deep learning methods, such as Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs), can be employed to realistically generate intermediate stratigraphy.
Supervision
Are you interested in this assignment? Contact the Master thesis coordinator:
