Rock Mass

Rock mass classification is crucial for safe and efficient underground construction, but traditional systems are limited by outdated methodologies and data scarcity. Current research focuses on leveraging high-resolution data from drilling, such as Measure While Drilling (MWD) data, combined with machine learning techniques like K-nearest neighbors, ensemble methods, and convolutional neural networks (CNNs) to automate and improve rock mass classification accuracy. This data-driven approach promises more objective and reliable assessments of rock mass quality, leading to better informed design and risk mitigation in tunneling and other underground projects. Improved accuracy in classifying rock mass properties translates to enhanced safety and cost-effectiveness in engineering projects.