Is the FEM any good for geotechnics?
Triggered by a topic raised some time ago, 'which geotechnical software is good for engineering', I raise a different question. Is a 'simple' method for solving partial differential equations (ie the FEM method), after it has been enhanced by commercial companies (ie they create the graphics) any good for the geotechnical community? An example: When you buy a program people will promise that you will save a lot of time and your results will be highly accurate. No more wasted money to overconservative design from the old and useless limit equilibrium methods. So you go home and try it. The screen has lots of bright colours and the program seems user friendly. So why do you need to read the manual and spoil the fun? You can design the embankment in no time. It is really easy to do inelastic analysis with the FEM also. All the program wants is an angle of dilation. Are you going to pay for the lab test to actually find its value? Probably not, but anyway the program has a default value equal to the angle of shearing resistance. Why not... You can also include anisotropy in the analysis. That 's amazing! You can 't do that with those simple methods. But what is even more amazing is the settlement evaluation. 2-D and also 3-D. You couldn 't do that with the Terzaghi 1-D. You may not have the required triaxials but you have some UU 's. Why not estimate the stifness from these? Press some more buttons and here you are. The embankment ready in no time. Now returning to my question: how can you possibly imagine that your FEM program is any better than a limit equilibrium calculation?