Limited monitoring data of newly-established monitoring slopes in universal landslide monitoring projects and their unavoidable poor representation ability of the deformation patterns have made the traditional single-slope modelling impossible. This paper proposes to classify the multi-slope integrated monitoring dataset based on the lithology of slopes and thus construct pre-trained models to apply to the newly-established monitoring slopes to improve the prediction performance. By integrating the monitoring data, the pre-train models can learn more deformation characteristics from the dataset than from only single-slope data. Moreover, by further classifying the integrated dataset based on the lithology of slopes, constructing different pre-training models, and applying them to newly-established slopes with corresponding lithology, it is feasible to enhance the classified dataset’s ability to represent the deformation patterns of corresponding kind of slopes while still ensuring the volume of dataset of each class is reasonable and, ultimately, to improve the pre-trained models by enhancing the consistency of the pre-training data and target domain data. A case study based on actual monitoring data shows that the pre-training models based on lithology-classified dataset perform overall significantly better on newly-established monitoring slopes with corresponding lithology than single-slope models or pre-training models based on other integrated dataset and may provide effective support for displacement prediction of newly-established monitoring slopes.

landslide, short-time displacement prediction, lithology-classified integrated dataset, pre-training model, newly-established monitoring slope, universal landslide monitoring