I think they offer more or less the same options in your case. They differ in the implementation of some feature (Ansys javascript API vs Abaqus Python API, Ansys scriptable input files vs Abaqus imperative only input file, to name a few) but overall they are both powerful FEM packages. They both offer multiphysics modules, GUIs, possibility to extend functionalities by scripting, pre-processing, post-processing...

I do think for you the choice should mainly boil down to the practicalities, as both are proprietary (and quite expensive): which one do you have available? Both offer a free student version, which is however quite constrained in the size of the model and needs proof of academic affiliation to be downloaded.

If both are available, you might discriminate based on your programming experience: if you're versed in Python, go for Abaqus; with Javascript, you might consider Ansys; with Fortran or C/C++ experience, you can go with both.

I agree with the Luca Di Stasio, they both are very strong package of FEM, but i think the use of Abaqus is the good choice to get more option in the GUI, but if u want more tutorials in the composite modelling, in that case Ansys would be useful.

It totally depends on how strong you are in scripting in ANSYS or ABAQUS. I think that usign ANSYS will be a bit more difficult than ABAQUS since finding JAVA commands is not straight forward. However, scripting in ABAQUS is a lot more convenient and all the APIs are available as well. At the end of the day it depends on you!! I will publish an article on this subject (particulate composite) soon (I'm in the process of publishing that article). You can follow me to receive a notification as soon as I published that.

It depends a bit on what you intend to do this analysis. I would be looking at Digimat as an addition to the FEA to enable multi-scale capabilities. You also are restricting yourself to Ansys and Abaqus, maybe as these are both available to you. MSC Marc has some particularly strong coupled analyses for combining structural and thermal analysis. Be aware that I work for MSC though...

Andrew Main Thank you for disclosing the conflict of interest! MSC is a good suggestion, nonetheless.

Gokul Krishna As the discussion has been extended to FEA packages other than Abaqus and Ansys, I think you should have a look at this link and save it in your bookmarks:

http://feacompare.com/

Be a little careful about this tick box approach... Because something can be done may not everything.

ANSYS is more superior in modelling composites, it has a dedicated module "AutoDyn" for composites in static and dynamic analyses with orthotropic damage models, and an-isotropic inelasticity as well and much more. I am not aware of a dedicated module for composites in abaqus.

Note that machining simulations are particularly tricky because you have to model material separation. Most people use explicit methods for that (so if I understand ANSYS correctly, you would use LS_DYNA).

Abaqus has a lot of options for composites nowadays as well.

If you have both available, I would recommend to try a simple model in both to see which one is easier for you. You might also look into the literature - composite machning surely has been done before, so check what other people use.

Dear Gokul Krishna,

Only some comments.

As Professor Martin Baeker has said, the machining simulations, the modelling of large contact surfaces, separation between materials, as well as large displacements or deformations is complex. The numerical convergence is difficult. In these cases, it is convenient to use explicit analysis.

Perhaps it might be convenient for you to use the software you know best, in which you have more experience. ABAQUS and ANSYS, both are very powerful.

Best regards,