couple of comments like to raised:
1. during cutting, the soil pretty much is undrained (using Cu instead of c') cause the construction would not be long for accounting drained(c' and phi). try to use Cu.
2. for deformation, the stiffness of soil play a very important role. and refer to Vermeer column, HS model would be better to modelling the Unloading case which split the loading and unloading modulus hense improve the analysis results. but you may happen to have no enough data for predicting the soil paramers required for HS model
3. you have c'=19.3~28.7, do you have more description of those material, but it just seems not right to me. the c' and Phi are very senstive in slope kind of analysis, and make sure you got it right in the first place.
4. the poisson ratio you using is basically like the odometer loading poission ratio(0.3), for unloading 0.2 would give better results
5. for structure member, the nails have been modelled using plate element, and the spacing should be also taken into account, pressumely you did not. and posson's ration for nail is too high. the input modulus should be the equivelant modulus.
Try to adopt the model based the comments above to see if there is any improvement
Cheers, yk
