The sampling area of an inventory sampling usually contains a large number of stock management units with extremely differing values. The larger the value range between the smallest value and the largest value within the sampling area, the more stock management units must be counted, in order to obtain statistically correct results.

To reduce the number of elements to be counted, the system divides the sampling area into classes, with each class containing elements of "approximately" the same value. Later, the system uses this classification as the basis for generating strata (stratification). For each stratum, the random selection and extrapolation is carried out separately.

Consequently, a class is the quantity of stock management units (in the sampling area) whose values lie within a certain interval. These intervals (possibly with the exception of the last one) are of equal size, but generally contain a variable number of stock management units.

The classification of the sampling area is carried out automatically when the stock population is formed. The parameters set during creation of the inventory sampling are used as the basis for the classification. The system attempts to generate 1000 classes. It calculates the class interval on the basis of the upper value limit specified, as follows:

class interval = upper value limit / 1000 (the result is rounded up to the nearest integer).

On the basis of the class interval, the system then generates classes until the upper value limit is reached. Due to the rounding up process, usually less than 1000 classes are generated. The last class ends with the upper value limit.

The upper value limit determines that stock management units with a value exceeding this limit are allocated to the complete-count area. For this reason, the system generates only 957 classes. If the classes just below class 957 contain only few elements, it may be useful to specify a smaller upper value limit, in order to obtain a smaller class interval. For example, if classes 924 to 957 contain only few elements, the upper value limit can be set to 11999. In that case, the elements as from class 924 and beyond would then be moved to the complete-count area. However, through a finer classification a more favorable stratification may be possible that requires less counting.