A new logistic model tree (LMT) model is developed to predict slope stability status based on an updated database including 627 slope stability cases with input parameters of unit weight, cohesion, angle of internal friction, slope angle, slope height and pore pressure ratio. The performance of the LMT model was assessed using statistical metrics, including accuracy (Acc), Matthews correlation coefficient (Mcc), area under the receiver operating characteristic curve (AUC) and F-score. The analysis of the Acc together with Mcc, AUC and F-score values for the slope stability suggests that the proposed LMT achieved better prediction results (Acc = 85.6%, Mcc = 0.713, AUC = 0.907, F-score for stable state = 0.967 and F-score for failed state = 0.923) as compared to other methods previously employed in the literature. Two case studies with ten slope stability events were used to verify the proposed LMT. It was found that the prediction results are completely consistent with the actual situation at the site. Finally, risk analysis was carried out, and the result also agrees with the actual conditions. Such probability results can be incorporated into risk analysis with the corresponding failure cost assessment later.
The California bearing ratio (CBR) is one of the basic subgrade strength characterization properties in road pavement design for evaluating the bearing capacity of pavement subgrade materials. In this research, a new model based on the Gaussian process regression (GPR) computing technique was trained and developed to predict CBR value of hydrated lime-activated rice husk ash (HARHA) treated soil. An experimental database containing 121 data points have been used. The dataset contains input parameters namely HARHA-a hybrid geometrical binder, liquid limit, plastic limit, plastic index, optimum moisture content, activity and maximum dry density while the output parameter for the model is CBR. The performance of the GPR model is assessed using statistical parameters, including the coefficient of determination (R2), mean absolute error (MAE), root mean square error (RMSE), Relative Root Mean Square Error (RRMSE), and performance indicator (ρ). The obtained results through GPR model yield higher accuracy as compare to recently establish artificial neural network (ANN) and gene expression programming (GEP) models in the literature. The analysis of the R2 together with MAE, RMSE, RRMSE, and ρ values for the CBR demonstrates that the GPR achieved a better prediction performance in training phase with (R2 = 0.9999, MAE = 0.0920, RMSE = 0.13907, RRMSE = 0.0078 and ρ = 0.00391) succeeded by the ANN model with (R2 = 0.9998, MAE = 0.0962, RMSE = 4.98, RRMSE = 0.20, and ρ = 0.100) and GEP model with (R2 = 0.9972, MAE = 0.5, RMSE = 4.94, RRMSE = 0.202, and ρ = 0.101). Furthermore, the sensitivity analysis result shows that HARHA was the key parameter affecting the CBR.