Application and Challenges of Machine Learning Techniques in Mining Engineering and Material Science

Abstract

The ultimate objective of modern engineering applications in mining and material science is to develop good quality novel materials with desirable qualities. Machine Learning (ML) is used in the mining industry to provide solutions to complex problems of the mining industry and improve the efficiency of the overall system. ML methods are increasingly being used by materials scientists to uncover hidden trends in data and generate predictions. Furthermore, data centric techniques can provide useful insights into the basic processes that influence material behaviour while simultaneously reducing human labour in large data processing. The ability of persons to find new materials and infer complex relationships is important for the development of new materials. Large amounts of machine-readable data must be available to use statistical methodologies to speed materials research. In mining engineering, ML can be used for analyzing geographical data, assessing the risk of rock fall, predicting equipment failures and impact of mining activities on the environment etc. Material science data may be used in a variety of ways, including property prediction, the search for new materials and discovering synthesis methods. Selecting proper machine learning techniques to provide solutions is very important and that is discussed here. The purposes of this paper are to provide a comprehensive list of different ML techniques which are applied for the mining and material science domain.