Comparative performance of six supervised learning methods for the development of models of hard rock pillar stability prediction

Abstract

The prediction of pillar stability (PS) in hard rock mines is a crucial task for which many techniques and methods have been proposed in the literature including machine learning classification. In order to make the best use of the large variety of statistical and machine learning classification methods available, it is necessary to assess their performance before selecting a classifier and suggesting improvement. The objective of this paper is to compare different classification techniques for PS detection in hard rock mines. The data of this study consist of six features, namely pillar width, pillar height, the ratio of pillar width to its height, uniaxial compressive strength of the rock, pillar strength, and pillar stress. A total of 251 pillar cases between 1972 and 2011 are analyzed. Six supervised learning algorithms, including linear discriminant analysis, multinomial logistic regression, multilayer perceptron neural networks, support vector machine (SVM), random forest (RF), and gradient boosting machine, are evaluated for their ability to learn for PS based on different input parameter combinations. In this study, the available data set is randomly split into two parts: training set (70 %) and test set (30 %). A repeated tenfold cross-validation procedure (ten repeats) is applied to determine the optimal parameter values during modeling, and an external testing set is employed to validate the prediction performance of models. Two performance measures, namely classification accuracy rate and Cohen's kappa, are employed. The analysis of the accuracy together with kappa for the PS data set demonstrates that SVM and RF achieve comparable median classification accuracy rate and Cohen's kappa values. All models are fitted by ''R'' programs with the libraries and functions described in this study.

Estimation of Interfacial Tension for Geological CO2 Storage

“…For improving accuracy and execution speed of the GB with the ultimate goal of improving performance , SGB incorporates randomization in the sampling of the training data which is the idea behind Breiman's bagging technique . Literature review reveals that this new approach has been applied in different domains with considerable success .…”

Post-failure fatigue behaviour of intact sandstone from burst prone coal mine in Ostrava-Karvina Coalfield

“…Besides the numerical modelling, statistical and analytical methods, the probabilistic methods, rock mass classification approaches and artificial intelligence based or their hybrids methods have been applied successfully in designing pillars in the case of coal or hard rock mines. Also [4] recently investigated the suitability of different supervised learning (SL) algorithms for the prediction of pillar stability (PS) in underground engineering. They have considered total of 251 pillar cases between 1972 and 2011 including stable and failed one available from the reported data sources in their study.…”

Predicting underground tunnel hazards using machine learning techniques

“…Additionally, in actual practice, approaches like Scheme 3 might be also preferable if a conservative prediction is of interest. (Mirzavand et al 2015), dam behavior (Salazar et al 2015), coal pillar stability (Zhou et al 2015), slope stability (Suman et al 2016), and rock burst hazards (Zhou et al 2016a). 6.2 Selection of parameters and model construction (Goel and Singh 2011) and it is therefore often unavailable at early stages of one project; (ii) actual measurements of tunnel support pressure are unavailable until the support is constructed, and their prior estimations are often unfeasible, as suggested by the large discrepancies found between estimations and measurements in real cases (Bhasin and Grimstad 1996); and (iii) the ranges of variability of rock mass unit weights are often limited, so that they can often be assumed constant in this type of practical empirical prediction (Dwivedi et al 2013). …”

“…Previously proposed methods for long-term rock burst prediction (see Chapters 3 and 4) only considered five main parameters -H, UCS, UTS, MTS and Wet-, or their combination (Zhou et al 2015;Zhou et al 2016a). However, as those parameters are mainly selected due to their availability at early stages of a project -hence being mainly useful for long-term prediction-, they might be inadequate (or insufficient) of short-term prediction.…”

“…As the model prediction performance is not very affected by them (Genuer et al 2010;Kuhn and Johnson 2013), and because the Accuracy is only a little more sensitive to mtry than to ntree (Zhou et al 2015), we decided to fix the ntree parameter, so that only mtry needs to be tuned.…”