Aggregating Data 

Use

With the  Aggregate Data node, you are able to group the data according to the values in specific fields (grouping fields, database function group by) and can aggregate the data within this group into other fields (aggregation fields). You are also able to only group data using these nodes, by only selecting grouping fields and not aggregation fields.

The node can be integrated into any point in an analysis process.

Functions

Since the data is going to be sorted, first all of the data has to be read. After sorting has taken place, the data is aggregated and passed on to the next node in groups. Only those fields are passed on that were either aggregated or used as grouping fields.

Activities

       1.      Select at least one grouping field from the available fields and if necessary, one or more aggregation fields.

       2.      As necessary, change the sorting sequence from ascending to descending in the grouping fields.

       3.      Specify the aggregation behavior for each aggregation field. The following values are currently supported:

¡        SUM: All values in a group, that is those with identical values in all grouping fields, are added up and the total is passed on to the next node.

¡        MIN: Only the minimum of the values in the group is passed on.

¡        MAX: Only the maximum of the values in the group is passed on.

¡        AVG: The average of all values in a group is passed on.

¡        AV0: The average of all values in a group is passed on. But, zero values in the fields are not considered in this aggregation.

¡        NOP: No aggregation.

Depending on the type of aggregation field, not all of the listed aggregation behaviors are supported. For example, for a field of type Character, only the values MIN and MAX are useful.

Example

If the characteristics Material (0MATERIAL), Material Group (0MATL_GROUP) and Customer (0CUSTOMER) along with the key figures Costs (0COSTS) and Revenue (0REVENUE) are available as input fields for the node, but you only transfer Customer and Material Group as grouping fields and the key figures as aggregation fields, then the data records look like this:

Aggregation on the customer level, material group with aggregation SUM for both key figures:

Aggregation on customer level with aggregation SUM for revenue and AVG for costs: