It depends mainly on how much the company can spend (usually in the form of an annual subscription) and what kind of problems will be solved - each software has different capabilities in terms of advanced analyses and in some cases programs are sold in the form of separate modules. Personally, I would choose Abaqus for highly nonlinear solid mechanics problems involving contact and complex material models (including fracture mechanics). But for CFD and some other types of analyses, I would vote for ANSYS. Sometimes even CAD-embedded FEA modules are enough and the company can work with SolidWorks Simulation, for example (to reduce cost and avoid buying unnecessarily advanced software). Well, in some (rather rare) cases open-source software may suffice - OpenFOAM solver's popularity is a good example.
There's an article titled "5 Tips for buying FEA Software" on Simuleon's blog - you may find something interesting there.
to follow up on greg's post ... it depends on what you want to do, and how much effort you're going to put in the learn it.
civil eng or other ? civil seem to have their own preferred codes (RISA, no recommendation), other will use lots of codes (ANSYS, SolidWorks, FeMap, PATRAN/NASTRAN).
simple FEA or something more colourful (dynamics, hyper-elastic, ...)
There is, IMHO, no "best" code. All should give the same answers (with the same mesh), some are easier to learn than others. The biggest difference is the pre- and post- processing.
another day in paradise, or is paradise one day closer ?
Most comprehensive and popular packages- Abaqus and Ansys for most mechanical products right from Pre-processing, solving and post-processing.
Based on your requirement (Your product, what you want to do CFD/FEA or combined together, just design based introductory analysis, full fledged complex analysis like crash of car, fracture mechanics, non-linear material simulation like hyper-elastic/plastic, plastics, visco-elastic/plastic etc.), there are different packages which can be used.
This link may help in shortlisting and decision making.
What software I would choose is easy. I would choose the software I have used for many years. But I think you need to ask yourself a few questions to start with.
- What is your budget?
- What are your analysis requirements? It ranges from linear static and up to non-linear dynamics including multi-physics. And there are also "side-tracks".
- How will you interact with the software, or requirements for pre/post software? Manually build a model or import a solid and auto-mesh?
My impression is that you think there is a single software that is the best for all types of applications. My experience is that there is no such software.
