Usually the inductor is designed based on the energy need to be stored in the inductor. P=0.5*L*I^2

I is the maximum current flowing through the inductor. If the current exceed the limit, the inductor will get saturated. A saturated inductor will behave just like wires.

If the inductor saturates or overloaded extremely, it will behave as a short, the current will increase very rapidly and will break your MOSFET and possible other components....

The primary concern regarding the inductor is saturation: as already stated, beyond the point of saturation the current will increase rapidly as the effective inductivity decreases.

Another concern will be about overheating the inductor - up to the point that it is damaged permanently.

yes its affect its performance ,dont overload it .

You may wish to use the Webench tool for designing - http://www.ti.com/lsds/ti/analog/webench/power.page

It will most probably affect it after a few milliseconds due to the fact that the inductor saturates and then leakage takes place. 

The answer might not be quite so black and white.  

First many inductors have a soft saturation characteristic, so the inductance doesn't drop to zero immediately.  The slope of the graph of current vs. time gets gradually steeper with time.  Some inductors such as those from CoilCraft have two or three saturation current ratings, representing 10%, 20% and 30% reduction in inductance.

Second, the RMS current rating will also be given at different levels, such as the current required for 20C, 40C and 60C temperature rise.

It is important to understand that the RMS current rating will be between 60% and 100% of the peak current for a buck converter operating in continuous mode.

So I think the more complicated answer is that the overall efficiency of the converter will be reduced for modest overload values, but will eventually fail entirely for greater overload values.  If the inductor current waveform appears roughly triangular and you can place a finger on the inductor while it is running, the converter is not overloaded.

As a final comment, going for the highest possible efficiency may require more extreme EMI filtering measures, so for a buck converter anything above 85% is probably sufficient if the temperature rise is acceptable.

The real question you should answer is how much overload.

Inductors are designed on the basis of Ampere turns. So with the increase of current beyond max rated current the inductor will become non linear and after certain current limit core will saturate causing heating due to core loss as well as conductor loss.

          Because of gross non linearity your equation for BUK converter will no longer remain valid. So design with lower flux density is much more important   than considering low current density of conductor as momentary over current hardly affects the heating performance of the conductor but even momentary over loading with considerable non linear core characteristics your BUK converter performance will get affected badly.