I am not familiar with foundations but have seen where thin shells are considered anything with a 10:1 ratio (span versus thickness). Some have considered 3:1 as well. This comes into play with Plane Stress versus Plane Strain when considering the material properties such as fracture mechanics.

TRY TO DESIGN BOTH CASES AND U GONNA KNOW THE DIFFERENCE
U'LL FIND SOME OF THE INTERNAL FORCES NEGLECTED WITH COMPARE TO THE OTHERS...

some recent elements developed, to my knowledge, in 1990 by Zienciewich and others (journal of computer and structures)namely the element by the name 'DRM' will work fine for both situations and you don't have to diferentiate. please also see the book by zienciewicz and taylor 'finite element method - vol II'.

For the case of thick plate assumption, the stresses in the thickness direction is considerably high that one can not ignor that of stresses whereas the stresses in the direction of the thickness is neglegted for the shell assumption.

If you are familar with ANSYS just go through the element notes of SHELL43 and SHELL63 which will give idea regarding thick and thin concept used in ANSYS.

The difference between shell and thick plate refers mainly to the mathematical solving theory used to find strain in the plates. The border between the two types is allocated on the total amount of plate deflection in service.
A rule of thumb says to solve the plate as "thick plate" when deflection perpendicular to surface is less than half the thickness and use results of "thin plate" theory when deflection is greater.

U. Fassio Mechanical Engineer

I used to think that thick shell approach is about the same idea like deep beam approach where shears start to mean something and deflections also would depend on shear rigidity whereas thin shells simply mean flexure dominant stiffness.

Am I wrong?

Hey Consulting, I now seem to share your understanding. I know that there are two kinds of plate theories, one for small deflection and one for large deflection. Many textbooks do categorize as such,including the reference that Zhol quoted, but it never occured to me that thick and thin had the same meaning as small and large. Strange huh!

I usually model matt foundation using thick plate sitting on springs representing piles.  In usual situation, there is no inplane forces induced in such foundation. thick shell is therefore unnecessary.  But if you want to know when to use thick/thin plates to analyse a matt, try both to see the difference.

Here is some explanation for the terms "plate, general plate, curved shell".

1) plate (bending) has a deflection, two rotations
2) general plate is consist of plane stress (two membrane displacements) + plate
3) shell has three displacements and two(or three) rotations

And the term "thick" means that the aspect ratio of the structure is less than 100, where there is no numerical instability using thick standard finite element.

In foundation engineering thick plate is a common occurence as a mat foundation. Here "Shear deformation is more prominent than bending deformation ". Hence I would select thick plate element for mat foundation and thin plate element for overhead slab. SAP 2000and SAFE has thick plate element with shear deformation considered.

If U r going to model raft foundation ofcourse it will be thin shell..due the aspect ratio between the thick. and its dimension...and if u r going to model isolated footings and u want an accurate model for research purpose It is recommended to use soild element (3d) ..tis will give u a good results.
Thick shells are used to model what named a hole cover at the roads...

In the analysis views, the difference between thin and thick shell is to consider transverse shear deformation or not to.