Abstract: Abstract: With rapid industrialization and urbanization in China, water and soil losses caused by engineering constructions become very serious. Engineering excavated slope is one of the main sources of water and soil loss in construction sites. Generally, the underlying surface of excavated slope has high compactness and bulk density, the slope gradient varies in the range from 30° to 70°, the length of slope is approximately 10 meters, and the time of excavating is generally within two years. Because of underlying surfaces changing greatly after exposure, excavated slope not only has general features of water and soil losses of traditional soil slope, but also has its own unique features. Little knowledge on quantitative soil erosion of engineering excavated slope has certainly impacted our ability to accurately assess soil erosion and harmless of excavated slope. Therefore, it is urgent to carry out research to develop soil erosion model of engineering excavated slope. The objective of this study was to develop a model to predict soil erosion in the engineering excavated slope. The model was established based on soil erosion theory in combination with field experiments. It incorporated factors that could affect soil erosion such as characteristics of excavated slope, engineering operation of simple and practical principle, and rainfall event. RUSLE was used as a reference for our model development. Fifteen different plots of excavated slope in Zhanjiang of Guangdong province and ten different plots of excavated slope in Fangcheng of Henan province were constructed and used for model fit. Key factors of soil erosion for excavated slope were discussed based on the field site observations. The model and its parameters were tested by measured data. The results showed that erosive forces and underlying slope had important impact on soil erosion of excavated slope. There was significant (p<0.05) partial correlation or correlation between soil loss per unit area and factors such as steepness of slope, length of slope, soil texture, and rainfall intensity and duration. The correlation analysis results were used to validate the applicability of the model and parameters analyzed. The predicted values from soil erosion model of engineering excavated slope under one rainfall event agreed well with the measured data. The model was simple and its parameters were measurable and easy to be obtained, and had physical meaning. The model was also easy to use and very good in practicability. For the application across the whole country, the model need further test and improvement in different environmental conditions. Nevertheless, this model provided the valuable information for effective controlling water and soil loss in the excavated slopes, implementing related policy, and supervising and executing law on soil and water conservation.