That is certainly possible, because the fully magnetized state is most likely not the ground state, but the system can lower its energy when breaking into a multi-domain state. The time scale on which that happens depends on temperature and the energy barriers that separate the local minima in the free energy from each other. As an example, if you hit a ferromagnet with a hammer, it can demagnetize as well by finding a lower energy state. 

thank you dear Christian

but this is not my problem. please consider an N pole of a magnet which is always at a distance from N pole of another magnet. I wanna know if the strength of magnets decreases through time?

That is what I tried to answer. Yes, it can happen. What I described is the fact that the fully magnetized ferromagnet is not in its ground state and thus can decay into a multi-domain state. If you have another magnet next to it in the way you describe you expose the ferromagnet  to a field that even further supports the tendency of demagnetization. Note that what you call strength of a magnet is determined by its magnetization. If this magnetization is reduced when the magnet breaks into a multi-domain state "the strength" is reduced. A more precise wording for strength would be "magnitude of the dipolar stray-field". 

Dear Professor

I really appreciate you for your good answer.

Thanks

Dear M Christian Binek,

Machines which are using permanent magnets have performances which can vary in time. The behaviour of the magnet is important. Its strength can decrease.

In permanent magnets machines, the temperature is the first parameter about the decrease of the strength.

You will find with this message, two articles. The first one is quite old (1991) and the last one is new (2014). They look at the same problem. It means that such a behaviour is always an issue.

Best regards

Hi Massoud,

The answers given here above are quite clear about the external demagnetizing influence from other magnets or magnetic fields created in electromagnetic machines. The (negative) impact concerns with the magnetic domain distribution within the microstructure.  However, since the development of coercivity mechanisms (resistance to demagnetization) are quite different from a type of magnet to another one (e.g. different in NdFeB magnets from the sintered, the melt spun, the spring type magnets), different from the SmFeCoCuZr magnets series (from type 1-5 to type 2-17 where it can depend on the ternary phase microstructure), different from the AlNiCo family of magnets where the initial texture plays a major role on domain distribution, different from the ferrites family of magnets where intergranular phases; texture and anisotropic structure play important roles.

Correspondingly there are as well different demagnetization processes that could be considered, the first ones being more temperature sensitive, the other ones more external field sensitive. Important parameters to consider are the Curie temperature of the used magnet (where both magnetization and magneto-crystalline anisotropy collapse) and the strenght of external fields imposed by the environment. Do not forget as well to consider the frequency of stimulation of your magnet by external fields that can as well induce eddy currents in the metal body of a magnet.

Often, from the catalogue of permanent magnet producers, you can find many various formula; more or less adapted to one or another one use.

Best regards

My large magnet for Hall measurements had a 'KEEPER' to limit field degradation. Magnetic circuits are just as real as electric circuits.

In the absence of other magnetic fields, the strength of an ordinary "permanent" magnet typically decrease very slowly with time.  At low temperatures, the rate of decrease is extremely slow (e-folding times of 10**9 or 10**10 years); however, if the temperature increases, the rate of decrease increases.  As the temperature approaches the Curie point (or the blocking temperature see Neel), the rate of decrease increases dramatically and the magnetization is lost.

All else being equal, a fair rule of thumb is that the "harder" the magnetization, the longer it lasts.  This is why folks moved from soft iron, to Alnico, to rare earth alloys, and beyond to make ever more permanent magnts.

The presence of an external magnetic field changes things (see hysteresis), but in your case, one must ask how "permanent" is the applied field? 

Dear Dr Voorhies

I really do appreciate for your good answer. for the field, assume the same as strength and same pole, distance= 1 cm, in room temperature, how much does it take to be decreased?

How much does what take to be decreased how much? Christian has the right of it: the opposing B fields that make like poles repel should also tend to accelerate the very, very slow natural thermal de-magnetization. But the effect should only be notable for a "strong" field -- a coercive force approaching, well, a fair fraction of the coercivity. Are these bar magnets?  If so, are they held in parallel -- side by side with N-N and S-S adjacent?  Or are they held end to end, so that N-N are adjacent, but the S poles are separated by a large distance?

Magnets do degrade with time as evidenced by the need for a 'keeper' in very big electromagnets

According to manufacturers' specifications, AlNiCo magnets do deteriorate, even with no strong extra magnetic field present. Two such magnets, in disputed configuration, will degrade noticeably within months, and should be remagnetized to restore their original properties.  On the other hand, 50 years old loudspeakers with AlNiCo magnets still perform satisfactorily (extra field from voice coil is pretty weak and the magnetic circuit is otherwise nearly closed).  NdFeB magnets will behave similarly, but the time scale is much, much longer in this case.  NdFeB magnet will spoil AlNiCo magnet within days, while itself remaining practically intact.

certainly, magnets will get demagnetized if it is applied a external magnetic field in disputed configuration. For modern magnets, elevated temperature can cause significant demagnetization as well.  

Dear Masoud Hasany.

You will find information about this topic in "Handbook of Magnetic Materials", K.H.J. Buschow. Such a topic is still in interest, have a look to the attached file.