Breakdown and Productivity Prediction of Dragline using Machine Learning Algorithms

Abstract

Dragline operations play a major role in the overall production of coal in open cast mining. Hence, it becomes necessary to maximize the working hours and minimize the idle and breakdown hours as it affects the overall production of a mine. There is also a shortage of skilled labour for dragline operations and combined with the time-to-time breakdown of dragline, it results in a production deficit. In this study, extensive research is carried out using machine learning algorithms on data obtained from one of the largest opencast mines in Singrauli. The data consists of the parameters that were maintained by the staff on a regular basis, and the algorithm tried to learn the underlying patterns between the independent and dependent variables and find the correlation between the parameters that have a significant impact on productivity and breakdown, which were the dependent variables. The results obtained from the algorithms are encouraging and, with certain improvements in data collection procedures, can improve the prediction accuracy to an effective level. An increase in the frequency of data collection and expanding the data recording using sensors to the electrical and mechanical parameters along with the specific type of failure in the dragline machine will further improve the accuracy of the model and can provide beforehand information so that the machine could be handed over to maintenance department for the change of faulty parts and necessary precautions that can be taken to prevent the breakdown which will result in an overall reduction of idle and breakdown hours and increase in overall production.