An Appropriate Model for the Prediction of Rock Mass Deformation Modulus among Various Artificial Intelligence Models

Abstract

The rock mass deformation modulus (Em) is an essential input parameter in numerical modeling for assessing the rock mass behavior required for the sustainable design of engineering structures. The in situ methods for determining this parameter are costly and time consuming. Their results may not be reliable due to the presence of various natures of joints and following difficult field testing procedures. Therefore, it is imperative to predict the rock mass deformation modulus using alternate methods. In this research, four different predictive models were developed, i.e., one statistical model (Muti Linear Regression (MLR)) and three Artificial Intelligence models (Artificial Neural Network (ANN), Random Forest Regression (RFR), and K-Neighbor Network (KNN)) by employing Rock Mass Rating (RMR89) and Point load index (I50) as appropriate input variables selected through correlation matrix analysis among eight different variables to propose an appropriate model for the prediction of Em. The efficacy of each predictive model was evaluated by using four different performance indicators: performance coefficient R2, Mean Absolute Error (MAE), Mean Squared Error (MSE), and Median Absolute Error (MEAE). The results show that the R2, MAE, MSE, and MEAE for the ANN model are 0.999, 0.2343, 0.2873, and 0.0814, respectively, which are better than MLR, KNN, and RFR. Therefore, the ANN model is proposed as the most appropriate model for the prediction of Em. The findings of this research will provide a better understanding and foundation for the professionals working in fields during the prediction of various engineering parameters, especially Em for sustainable engineering design in the rock engineering field.

Employing Statistical Algorithms and Clustering Techniques to Assess Lithological Facies for Identifying Optimal Reservoir Rocks: A Case Study of the Mansouri Oilfields, SW Iran

“…where: E rr is the error (deviation), Y is the actual RQI of each sample, and y is the estimated RQI of each sample [21,[51][52][53]. Classification of samples for HFUs 1 to 4 are presented in Tables S1-S4.…”

Appraisal of Different Artificial Intelligence Techniques for the Prediction of Marble Strength

“…Various performance indicators such as the coefficient of determination R 2 , Root Mean Square Error (RMSE), Mean Square Error (MSE), and Mean Absolute Error (MAE) were used to evaluate the prediction performance of the Artificial Neural Network (ANN), XG Boost algorithm, Random Forest Regression (RFR), Elastic Net (EN), Lasso, Support Vector Machine (SVM), and Ridge models. The following formulas as mentioned in Equations ( 9)-( 12) were used [43,64]:…”

“…In rock mechanics, these methods are getting more attention now because they are flexible and make it easy to predict required values based on the input variables. In situations when using traditional statistical approaches for prediction is not as convenient, these methods are particularly appropriate to be applied [43][44][45][46].…”