Abstract: To prevent the loss of properties and human lives, it is of great significance to establish an accurate and reliable model for predicting the landslide time of high fill slope. On the basis of the summary of the displacement and velocity characteristics during the creep failure process of high fill slope, this paper proposes a practical model for predicting the medium and short-term landslide. The proposed model is based on the improved artificial bee colony algorithm by modifying the strain rate formula of Saito model. The time series of landslide displacement after entering the accelerated deformation stage are taken as the input, and then the predicted landslide time is obtained by inverting the practical model parameters through the artificial bee colony algorithm. Three cases of high fill slope landslides are taken as examples, and the accuracy and reliability of the practical model are verified by the displacement data of the monitoring points. Meanwhile, the results of landslide time predicted by this model are compared with those predicted by the traditional Saito series model. The results show that the practical model is more accurate and reliable than the two general Saito models in predicting the landslide time through the time series of landslide displacement.

Key words: high fill slope, landslide, failure prediction, artificial bee colony algorithm