It depends on the condition of your rock mass. If there are too make joint sets in rock to be modeled, UDEC/3DEC can be used in which discontinuous medium is represented as an assemblage of discrete blocks while the discontinuities are treated as boundary conditions between blocks. If there are not too many joint sets to be modelled, then FLAC, which is based on Lagrangian analysis can be used effectively. The solution scheme in both these codes is based on finite difference method. Many people have used these codes successfully to simulate excavations in rock mass. One of my papers on this topic:

Sitharam,T.G. and G. Madhavi Latha, G. (2002).Simulation of excavations in jointed rock masses using a practical equivalent continuum approach, International Journal of Rock Mechanics and Mining Sciences, Elsevier, Vol. 39(4), pp. 517-525. DOI: 10.1016/S1365-1609(02)00024-2.

I agree with Madhavi's answer, and add that there are many more possibilities.  However, be careful if your model  goes beyond the elastic limit to not predetermine plastic localization by how you constructed your mesh or grid.

Le deseo buena suerte en su decision. He tenido más experiencia con los códigos de modelado de Itasca como Madhavi. Pero es muy necesario para mantener tamaño de la zona constante en el área de interés.

Flac 3D is commonly used and Finite Elemental Analysis Software is ultimate.

• In order to evaluate underground structures both FEM or DEM analyses are adequate. There very good software from ITASCA, Z-Soil and PLAXIs, as an example.

I felt very confortable when I used Phase (Rocscience) (recently they´ve just released a new version called RS2, equivalent to Phase2 9.0). But first of all, take into account if you want to perform a 2-D or a 3-D analysis. Phase is suitable only for 2-D analysis.

I agree with the answers already provided. The three main questions that need to be asked are: 1. do you want to do the analysis in 2D or 3D? 2. are there more than a few discontinuities that need to be modeled (joint sets, faults, etc.)? 3. how much time do you have? If there is enough time, you can use FLAC3D (1-2 faults) or 3DEC (multiple joint sets, faults, etc.), bearing in mind that the learning curve is somehow steep and model construction will take the most of your time. If there's only a few months to do this in and 2D is sufficient, I'd recommend going with the latest version of Phase2.

Try with RS2 and RS3 of rockscience package, which is a userfrendly and also have chance to carry out sensitivity analysis. However, results from numerical modeling are largely dependent on the way you select the input variables.   

Is Phase2 (Rocscience) is suitable for dynamic stability assessment of jointed rock slope?

In this website you can find side by side comparison chart of 78 Finite Element Analysis (FEA) programs and their sorting by features:

http://feacompare.com/sortbyfeatures.html

https://www.researchgate.net/project/ADONIS-A-Free-Nonlinear-Finite-Element-Program-for-Geo-Engineers