Abstract: The horizontal slice method is commonly used for examining the trench stability in layered soils. Unsafe results may be given by the 3D horizontal slice method when the ratio of trench length to depth is relatively small. In addition, the effective soil strength parameters are recommended for trench stability analysis. However, the soil strength parameters in engineering reports are often obtained from the direct shear tests. Further investigation is needed to incorporate the trench stability analysis with the current practice. In this study, the theoretical framework of the horizontal slice method is introduced, and its connection with the sliding wedge method is discussed. The deficiency of the 3D sliding wedge method in undrained clay is shown by a field test, and the improvement by an equivalent 3D wedge method is proposed. A discussion on influences of the differences between 2D and 3D methods, selection of soil strength parameters and water pressure calculation on the calculated results is carried out through 10 collected case studies.