Hello,. this cannot be prevented from happening - these are 'reserved' dimension rows to support late-arriving and 'Unknown' dimension processing.
For example lets say you have ORDER fact with a SHIPPER column that is the business key for the SHIPPER dimension.
If the ORDER fact record has not been shipped yet, and thus has NULL for the SHIPPER column, it will get related to the '0' record - which denotes an Unknown for the dimension. When the fact source gets updated to have a SHIPPER value (e.g USPS) - it would then update to the appropriate dimension id.
If you have a late-arriving dimension scenario for example the SHIPPER has a value of "NEWSHIP" for a new shipper and that business key does not exist in your DIM_SHIPPER dimension, the data mart processing will relate that fact to the -1 row. This denotes a missing dimension member in your dimension table. When the "NEWSHIP" business key arrives for teh DIM_SHIPPER dimension - its dimension row will be created, and the fact record updated to align to the previously missing dimension.
Hope this makes sense as to the processing characteristics and use of the -1 (late-arriving) and 0 (NULL business key) members in the dimension tables.
Once the 'E
