&CAUSE&
With the context structures for the attributes in the planninghierarchy, you define the characteristics on which the characteristicvalues of the attribute are dependent. Through this dependency ofattributes, a sequence is determined, in which the characteristicvalues of the attributes are defined.
In principle, it is possible that the characterisitc value of anattribute is dependent on the characteristic value of anotherattribute. In certain cases, this leads to the problem of circulardependencies.
Example:
The levels material group (MATKL), country (LAND) and branch (WERKS)are contained in the planning hierarchy. The planning level materialgroup is a component of the specified master data hierarchy. Thecountry and branch are not contained in the master data hierarchy. Theyare, therefore, attributes.
The planning level WERKS must contain all branches that have thecorresponding material groups within the particular country (LAND) intheir assortment. For the attribute WERKS, the context structuretherefore contains the planning levels MATKL (component of the masterdata hierarchy) and LAND (attribute)
You still have to determine how the attribute LAND is to be filled. Ifyou want to see the country independently of the branches (perhaps todetermine the country of the branch from the adress data), thefollowing occurs: The context structure for the attribute LAND containsthe attribute WERKS. This is in accordance with its context structurebut in turn it is dependent on the attribute LAND!
This is a circular dependency: Between attributes there is a chain ofmutual dependencies, without a "start" in the chain:
Attr1 -> Attr2 -> .... -> AttrN -> Attr1.
In the case of circular dependencies, the system cannot determine theattributes to be defined or in what order.

System Response
The system could not create the planning hierarchy.

Procedure
Check the relevant context structures for the attributes. Remove thecircular dependencies and begin constructing the planning hierarchyfrom master data again.