Comments on ‘Two-dimensional slope stability analysis by limit equilibrium and strength reduction methods’ by Y.M. Cheng, T. Lansivaara and W.B. Wei [Computers and Geotechnics 34 (2007) 137–150] J. Bojorque * , G. De Roeck, J. Maertens Department of Civil Engineering, Katholieke Universiteit Leuven, Kasteelpark Arenberg 40, B-3001 Leuven, Belgium Received 18 April 2007 Available online 11 June 2007 In the first part of the paper the authors present a com- parative analysis between the location of the critical slid- ing surface computed by limit equilibrium method (LEM) using an optimization procedure (annealing technique) and the location of the maximum shear strains (total/ incremental) computed by finite element analysis (FE) determined from the strength reduction method (SRM). For the FE-SRM both non-associated flow rule (SRM1) and associated (SRM2) are used. Furthermore, in this sec- tion the authors present a comparison between the factor of safety values (FOS) computed by Spencer’s method and the SRM1 and SRM2. Next, the authors have con- structed three interesting cases where the applicability of SRM might be limited. The first example presents the influence of a soft band in SRM. In the two other exam- ples the presence of different sliding surfaces is high- lighted. In the last section a small parametric study with respect to the influence of the elastic modulus for SRM is presented. 1. Comparative analysis between LEM and FE-SRM The authors conclude from the first part that for most of the cases the FOS obtained by SRM are slightly larger than those obtained by LEM with only few exceptions. How- ever, by performing a FE-SRM with PLAXIS (commercial finite element package) [1], the discussers found that for all of the cases the FOS values are lower than those obtained by the authors’ method, and for most of the cases slightly lower than LEM. It is worth mentioning that the FOS computed from Spencer’s method without optimization procedure (anneal- ing technique) are higher than SRM. However, for some cases (especially case 16 and other cases with low friction angle) using the optimization technique, the FOS value might be 10% lower than SRM, and for those cases the location of the sliding surface differs. These cases should be considered for further analysis. It is assumed that the authors performed the FE-SRM calculations for a maximum tensile strength equal to the cohesion divided by the tangent of the friction angle, related to Mohr–Coulomb failure criteria. Most of the soils do not support or support small tensile strength and a tension-cut off equal to zero, i.e. tensile strength equal to zero, is more realistic and conservative. To extend the comparative analysis, a tension-cut off for both non- associated flow rule (SRM1T) and associated (SRM2T) has been introduced. The presence or absence of a ten- sion-cut off influences the computation of the factor of safety in SRM and at the same time affects the location of the sliding surface especially at the crest of the slope. Moreover, the tension-cut off might cause concentration of strains at the toe of the slope for a associated flow rule. For this case, the determination of the sliding surface derived from the total/incremental shear strain contours needs extra care. Table 1 present the comparative analysis using PLAXIS with and without the inclusion of the tension-cut off. Based on Table 1 some summary conclusions can be made as follows: 0266-352X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.compgeo.2007.04.005 * Corresponding author. E-mail address: jaime.bojorque@bwk.kuleuven.be (J. Bojorque). www.elsevier.com/locate/compgeo Available online at www.sciencedirect.com Computers and Geotechnics 35 (2008) 305–308