Predictive Models for Rock Strength using Microstructural Properties: A Review

Abstract

Mining and geotechnical engineering depend on the understanding of rock strength since it influences the excavation design, stability evaluation, and reserve extraction productivity. This review summarizes advances in predictive modeling of rock strength using microstructural properties, with emphasis on machine learning approaches such as  Artificial Neural Networks (ANN),  Support Vector Machines (SVM), and hybrid optimization algorithms. It highlights key influencing factors (grain size, porosity) and discusses modeling challenges and future directions. The latest developments have improved rock strength prediction models in machine learning and microstructural research. The combination of microstructural features with data-driven approaches is the main emphasis of this paper's overview of current techniques and technology utilized in the predictive modelling of rock strength. Microstructural characteristics of rock have a great and direct influence on predicting the mechanical properties and failure mechanism under loading. This paper reviewed the current model techniques that have been adopted, future directions, and the major problems. However, the authors strongly advised that the statistical methods of regression analysis should be encouraged to save time and cost.