Rockburst prediction is of crucial importance to the design and construction of many underground projects. Insufficient knowledge, lack of characterizing information, and noisy data restrain rock mechanics engineers from achieving optimal prediction results. In this paper, a data set of 246 rockburst events was examined for rockburst classification using supervised learning (SL) methods. The data set was analyzed with 8 potentially relevant indicators. Eleven algorithms from 10 categories of SL algorithms were evaluated for their ability to learn rockburst, including linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), partial least-squares discriminant analysis (PLSDA), naïve Bayes (NB), k-nearest neighbor (KNN), multilayer perceptron neural network (MLPNN), classification tree (CT), support vector machine (SVM), random forest (RF), and gradient-boosting machine (GBM). The data set was randomly split into two parts: training (70%) and test (30%). A 10-fold cross-validation (...

Ground settlements above a tunnel as a result of tunnel construction can be predicted with the help of input variables that have direct physical significance. Several empirical and artificial intelligence methods for estimating ground settlements have been established by researchers. However, these methods have some limitations because the large number of influential factors involved makes tunnel–ground interaction complicated. In this work, a random forest (RF) was developed and employed to predict ground settlements above tunnels. To achieve this goal, tunnel geometry, geological properties, and construction parameters were investigated as input variables to utilize in the RF modeling, resulting in the maximum surface settlement value (Smax) and trough width (i) as the ground surface settlement index. To demonstrate the applicability of the RF model, two data sets associated with different features, which were obtained from a detailed investigation of different tunnel projects published in litera...

The paper examines the dynamic response of flexible conventional pavement systems to single wheel traffic loads in terms of the pavement design criteria, namely the fatigue strain at the bottom of the asphalt concrete layer and rutting strain at the top of the subgrade material. Model setup including geometry, boundary conditions, and load wave characterization are presented. The effect of elastoplasticity of the base material and elastoplasticity with strain hardening of the subgrade material on the dynamic response of the pavement system are first investigated. A detailed model parametric study then follows to show the effect of the base strength and thickness and the subgrade quality on the fatigue and rutting strains and the vertical surface deflection. The study, conducted with program ADINA, employs a three-dimensional, implicit dynamic, finite element method.

A series of finite element (FE) simulations are carried out to evaluate the benefits of integrating a high modulus geosynthetic into the pavement foundation. The simulations are conducted under a parametric study to investigate the beneficial effects of geosynthetic reinforcement to the fatigue and rutting strain criteria, and to determine how such effects are influenced by the base quality and thickness as well as the subgrade quality. Three locations of the geosynthetic reinforcement are studied, namely the base–asphalt concrete interface, the base–subgrade interface, and inside the base layer at a height of 1/3 of its thickness from the bottom. It is found that placing the geosynthetic reinforcement at the base–asphalt concrete interface leads to the highest reduction of the fatigue strain (46–48%). The placement of geosynthetic reinforcement in thin bases is particularly effective; the highest decrease of rutting strain (16–34%) occurs when the reinforcement is placed at a height of 1/3 of the base th...