Mousa Attom, Rami Haweeleh, Mohammad Yamin, Philip Virgil Astillo
Internal erosion of soil is the main factor behind embankment dam failure. Such failure has a devastating impact on properties and human lives. The susceptibility of internal erosion is represented by an erosion rate index (IHET) value. Evaluation of IHET experimentally is expensive, time-consuming, and affected by many physical and mechanical soil properties. In this research, Artificial Neural Network (ANN) model based on experimentally obtained values of liquid limit, plasticity index, initial water content, initial dry density, internal friction, and cohesion for sixteen types of clay soils have been developed to predict IHET. The IHET corresponding to these different parameters was obtained using the Hole Erosion test and compared with those obtained from the ANN model. The erosion rate index of these soils obtained from ANN models correlated with that obtained from experimental results. The performance indices such as the coefficient of determination (R2) and mean square error (MSE) were used to evaluate the performance of the prediction capacity of the models developed in this study. The developed models achieved R2 and MSE values of 0.9508 and 0.040216, respectively. Thus, it can be concluded that the ANN is an efficient tool to predict the erosion rate index of clay soil. © 2025 The Authors.
Department of Civil Engineering, American University of Sharjah, Sharjah, United Arab Emirates; Department of Civil Engineering, Minnesota State University, Mankato, United States; Department of Computer Engineering, University of San Carlos, Cebu, Philippines
