Novel intelligence technique analyzing slope stability based on Harris Hawks' optimization and artificial neural network

The evaluation and precise prediction of safety factors can be used to design and assess slope structures'safety. However, due to the complexity of slope structure and the uncertainty of related aspects, the simple prediction model dissatisfies the accuracy requirements. Hence, the main objective of this study is to introduce a new heuristic algorithm, i.e., Harris Hawks' Optimization (HHO), to advance the accuracy of traditional neural networks (ANN) in predicting slope safety factors. HHO was used to adjust the weights and biases of the ANN model. To clarify the accuracy of the HHO-ANN algorithm in the prediction of slope safety factors, the GA-ANN model was established and assessed as the control group. Three performance indexes of mean absolute error (MAE), root mean square error (RMSE), and determination coefficient were calculated to evaluate the accuracy of the prediction model. The results clarified that the HHO-ANN model had the most dominant accuracy than other prediction models. This consequence indicates that the HHO algorithm can improve the performance of ANN and raise the reliability of the model in predicting slope SF values.