Search

In their highly influential paper, Yarkoni, Tal, and Jacob Westfall. 2017. "Choosing Prediction over Explanation in Psychology: Lessons from Machine Learning." Perspectives on Psychological Science 12 (6):1100–1122. the authors highlight difficulties in traditional explanatory research in the field of psychology and argue in favour of novel data-driven science. By applying machine-learning methods to large data sets, predictive power has been shown to increase significantly. Geological engineers are responsible for a wide range of applications, including the design of tunnels, dams, foundations, and mines. While the field of geological engineering stands on solid mechanistic grounds, we argue that its predictive aspect aligns more closely with psychology than other mechanistic sciences. We therefore propose a paradigm shift in geological engineering research towards a prediction-centric approach. Potentially, this could enhance cost-effectiveness in structural design and lead to substantial societal savings. [ABSTRACT FROM AUTHOR]

Numerical modeling is increasingly used to analyze practical rock engineering problems. The geological strength index (GSI) is a critical input for many rock engineering problems. However, no available method allows the quantification of GSI input parameters, and engineers must consider a range of values. As projects progress, these ranges can be narrowed down. Machine learning (ML) algorithms have been coupled with numerical modeling to create surrogate models. The concept of surrogate models aligns well with the deductive nature of data availability in rock engineering projects. In this paper, we demonstrated the use of surrogate models to analyze two common rock slope stability problems: (1) determining the maximum stable depth of a vertical excavation and (2) determining the allowable angle of a slope with a fixed height. Compared with support vector machines and K-nearest algorithms, the random forest model performs best on a data set of 800 numerical models for the problems discussed in the paper. For all these models, regression-type models outperform classification models. Once the surrogate model is confirmed to preform accurately, instantaneous predictions of maximum excavation depth and slope angle can be achieved according to any range of input parameters. This capability is used to investigate the impact of narrowing GSI range estimation. [ABSTRACT FROM AUTHOR]

Smuggling and warfare tunnels are unique structures that have rarely been studied from an engineering perspective. A notable example is the vast networks of tunnels that were secretly constructed underneath the Gaza Strip. Particularly because these tunnels were not designed and constructed via traditional engineering practice, they constitute an interesting case study. The tunnels are supported by thin precast concrete elements, with the wall elements being the critical structural element. While some instances of structural failure and collapse have been reported in the media, a great number of the tunnels have remained stable. In this paper, we attempt to conduct a forward analysis to estimate the load and response of the wall elements. We estimate the range of problem input parameters based on multiple sources, including media accounts, geological research papers, and geotechnical reports obtained from the vicinity of the Gaza tunnels. The problem is then analyzed using two approaches: (1) a simplified structural analysis based on lateral earth-pressure theory and (2) numerical modeling. Both analysis methods show that the wall elements should fail due to compression even under the most favorable estimates of input parameters, in contrast to actual reality. We discuss possible explanations for this disparity. While it is not possible to pinpoint the exact explanation, we argue that current geotechnical practice is generally biased toward conservatism, even prior to the application of safety factors. [ABSTRACT FROM AUTHOR]