If it was this easy, you wouldn't have to study it, would you?
In a biatomic molecule it is omega=sqrt(k/mu) so a redshift (=omega decreases) either means the force constant k decreased or the reduced mass increased; vice versa fo a blueshift.
If a molecule has more than two atoms, all vibrational modes are collective modes. If it has symmetry you can use group theory (=not simple) in order to manually calculate the influences of specific atoms on the vibrations. If it is of low symmetry you will need explicit ab initio calculations in order to say something.
There are some specific functional groups whose wavenumber (or frequency) is mostly decoupled from the surrounding and for which there are simple rules whether its surrounding is electron withdrawing or donating; a classic in this are carbonyls in infrared spectroscopy. You can look whether your material group has such a functionality.
Depending on a material, If the major Raman peak move towards higher wave number then from the references it mean that there might be decrease in the structure thickness. Mean the outer layer of the structure is peeling off.
Saud Asif Ahmed : A correlation to thickness and peel-offs is material-specific and should not be generalized; often such effects just affect the peak width, not the position. The question doesn't even specify that it is about solids.
Raman peak blue shift or red shift could be due to many reasons, but most likely linked to change in structure and state of the studied samples. I have seen people using raman peak shift as a gauge to measure pressure change. Or, in some raman studies by Pemberton, peak ratios and peak shifts in c-h stretching, bending, twisting modes gradually change when melting a solid sample.