A validation of the use of data sciences for the study of slope stability in open pit mines

TL;DR

This study explores the use of data mining techniques to predict slope stability issues in open pit mines, aiming to enhance safety and economic protection through early risk detection.

Contribution

It demonstrates the application of data mining to predict slope failures in mining, providing a validation of its effectiveness for risk assessment.

Findings

Successful prediction of a 2017 slope failure event

Identification of key risk indicators up to six months in advance

Validation of data mining as a useful tool for mine safety management

Abstract

In this work, we present an exploratory study of stability of an open pit mine in the north of Chile with the use of data mining. It is important to note that the study of slope stability is a subject of great interest to mining companies, this is due to the importance in the safety of workers and the protection of infrastructures, whether private or public, in those places susceptible to this kind of phenomena, as well as, for road slopes and close to communities or infrastructures, among others. It is also important to highlight that these phenomena can compromise important economic resources and can even cause human losses. In our case, this study seeks to increase the knowledge of these phenomena and thus, try to predict their occurrence, by means of risk indicators, potentially allowing the mining company supervision to consider predictive measures. It should be considered that there is no online test that ensures timely prediction. In previous studies conducted in other mines, it has been corroborated that the phenomena and factors associated with the movement of slopes and landslides are extremely complex and highly nonlinear, which is why the methods associated with the called "data mining", were found to be ideal for discovering new information in the data, which is recorded periodically in the continuous monitoring that mining companies have of their deposits, which allowed to find important correlations for the search of predictors of these phenomena. Some results, with data coming from different sources, are presented at the end of this work. We note that according to the information provided by the mining company, the results were favorable in the indicators of up to six months, giving as risk areas the correct sectors and predicting the April 2017 event.