Injury Prediction in Mining Industry through Applied Machine Learning Approaches

-, Akash Manjunatha (2023) Injury Prediction in Mining Industry through Applied Machine Learning Approaches. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration manual) Download (4MB) | Preview

Abstract

The mining industry is a significant contributor to the American economy, but it is also one of the most dangerous industries to work in due to the complex and risky nature of mining operations. To protect workers and reduce fatalities and accidents, the government created the Occupational Safety and Health Administration (OSHA) and the Mine Safety and Health Administration (MSHA). These organizations set safety regulations and penalties for companies that violate them. Despite the implementation of these safety measures, there are still unacceptable risks for workers in the mining industry. MSHA requires companies to record all workplace accidents and offers resources to help mine operators comply with safety regulations. Employers who violate safety regulations face steep fines from either OSHA or MSHA. Use of technology in the mining industry especially in health and safety is very minimal, given the volume of data that has become available over the years, this industry needs technology. In this research, five machine learning algorithms and one deep learning algorithm are employed to categorize the degree of injury in the mining industry. Three case studies were undertaken in order to address the research issue. Case studies 2 and 3 put the presumptions from case study 1 into practice. XGboost, decision trees, and artificial neural networks all performed admirably in case study 2’s prediction of whether or not a worker will take a day off due to injury, with an accuracy rate of about 92%. The results of case study 3 showed that multi-classification with XGboost outperformed other algorithms by accurately identifying the degree of injury with 91% in all matrices. The outcome of this research can be used in organizations to improve their health and safety practices, and more effectively forecast, and prevent injuries or accidents altogether.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Horn, Christian UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science H Social Sciences > HD Industries. Land use. Labor > Issues of Labour and Work > Health and Safety at Work. Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Tamara Malone
Date Deposited:	12 May 2023 16:43
Last Modified:	12 May 2023 16:43
URI:	https://norma.ncirl.ie/id/eprint/6559

Actions (login required)

View Item