I don't understand what you are looking for.  Are you trying to find a method to demagnetize a magnet or are you looking for a standard (or regulation) for determining when something is demagnetized?

The International Magnetics Association has a standard specification for (among other things) demagnetization level, but it is based on the number of particles a magnet picks up.  They basically state a magnet is considered demagnetized when it picks up 3 or less particles of soft iron particles within size -5, +10 mesh.

It is sometimes difficult to use a magnetic reading to determine demagnetization.  When demagnetizing a magnet via magnetic fields, the face of the magnet has regions that are still magnetized.  Sometimes those regions alternate polarity, making it more difficult.

The "best means of measurement" depends on the application.  What is the purpose of demagnetizing the magnet?

What kind of item are you measuring?
In long product (bar and tube) often direct readings are used.  In other geometery attraction of particles may be the only meaningful test.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection
http://www.trent-tube.com/contact/Tech_Assist.cfm

MagMike : Concerning our research, we'd like to know when our parts made of bearing steels are "correctly" demagnetized and how measure them (pb of local magnetization in complex geometry despite global demagnetization). The demagnetization is important for us for machining, non destructive controls and in service reasons.

The IMA specification is very interesting. Could we associate the emperical test with a residual magnetization value?

EdStainless : Today, we measure macroscopic residual magnetization in the air (<8 Gauss) but we have also transveral magnetic induction measurement device. Could we link these 2 measurements?

Regarding the IMA specification: It is not possible to associate it with a residual magnetization value.  A number of factors are involved in determining how many particles a magnetic material will pick up.  Residual magnetization is only one of those factors.

If you are looking for an "official" definition of demagnetization level, the IMA spec is a close as you're going to find.  It may not be practical in every application.  You might need to work with your end customer on a mutually agreeable level or test methodology.  

You'll probably find that a part that holds more than 3 particles can still be machined without any problems.  The same could be true for the other situations you mentioned.

Many thanks for your information.

Could you give me the reference of the IMA specification concerning the demagnetization level?I do not find it on the web.

Here is the link:
http://www.intl-magnetics.org/pdfs/0100-00.pdf

The demagnetization level is discussed on page 3.

The reason that I asked about geometery is that it has a major impact.
For example, heavy walled tubing, measured individually with reading 3-6G taken at the ends of 20' long pieces.
When these are in a bundle of 36 pcs the reading is 20G.
When a single piece is cut into 12' lengths the readings range from 1G to 10G.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection
http://www.trent-tube.com/contact/Tech_Assist.cfm

We have principally flat products made of bearing steels from 6 to 12 mm thick. The parts are fine blanked and heat treated to obtained a hardness of 700-800 HV.

The geometries of the different parts are rarely symetric.