Defining Aggregation Schemas
An aggregation schema is a series of steps enabling consolidation of assessment results according to specified assessment rules.
Two different aggregation schemas can give different global results from the same assessment. An aggregation schema can be applied before effective start of the session. It will be used in reports and dashboards.
Aggregation Schema Principle
After execution of an assessment session, the value of each assessed characteristic is calculated based on a scoring rule defined for the assessment template.
An aggregation schema enables calculation of the value of an assessed characteristic for a root object from calculated values for child objects.
Aggregation Schema Example
Let's take the example of assessment of the impact of risks connected to entities: impact value should be calculated for each assessment node.
The value of the "Impact" of the "Fraud" risk in the context of the "United States" entity is an example of an assessment node.
If the “United States” entity comprises several sub-entities (States or Cities), themselves exposed to the "Fraud" risk, each of these sub-entities is associated with an assessment node with its own impact value.
For example, you can calculate the impact value of the "Fraud" risk for the "United States" entity from assessed characteristics values for each sub-entity (State or City).
Creating an Aggregation Schema
An aggregation schema consists of a series of aggregation steps enabling consolidation results according to specified aggregation rules.
To create an aggregation schema associated with an assessment template:
2. Open the assessment template properties.
3. Select Assessment Template Definition > Aggregation Schemas.
4. Select the Aggregation Schema folder and click New.
5. Indicate the name of the schema and click OK.
A weighting factor can be defined in the aggregation schema.
Creating an aggregation including aggregation steps
To create an aggregation including aggregation rules:
1. In the properties of an assessment template, select Assessment Template Definition > Aggregation Schemas.
2. Expand the folder of the aggregation schema that interests you.
3. Right-click the Aggregation folder and select New.
4. Select the method for grouping results for this aggregation step.
• By substitution: a set of assessment nodes is replaced by another set
Aggregation by substitution replaces Context objects by other objects. • By Reduction: an assessment node is removed from the context collection.
Aggregation by reduction enables elimination of Context objects defined in the assessment session context collection. • By Distribution: a set of nodes is redistributed on new nodes
Distribution enables redeployment of the values of another aggregation step on new contexts. • By Group of Aggregation Steps: regroups several aggregation steps.
5. Click Next.
You must now create an aggregation rule corresponding to the aggregation step.
6. In the Aggregation Rule section, click New.
7. Specify the Aggregation Rule Type.
• MIN, MAX, SUM, Average
• Macro
For more details, see Aggregation rule types.8. Select the Characteristic that carries the result of aggregation.
9. Connect the Aggregated Characteristics corresponding to potential assessed characteristics to which the aggregation step relates.
10. Click OK.
Aggregation rule types
In "Macro" type aggregation rules, the macro should have the following signature:
Sub aggregationRuleCompute(mgobjAggregationRule As MegaObject,mgobjOutAssessedValue As MegaObject)
End Sub
where:
• mgobjAggregationRule = the current aggregation
• mgobjAssessedValue: assessed characteristic value to be computed.
Skipping the creation of aggregation nodes
You may specify that the aggregation step should not generate aggregation nodes. This way aggregation computation is lighter and performance is improved. Skipping aggregation nodes may have an impact on the data used in reports.
To avoid generating aggregation steps:
1. In the properties of an assessment template, select Assessment Template Definition > Aggregation Schemas.
2. Expand the "Aggregation" folder.
3. Open the properties of the aggregation step of interest.
4. Select the Skip node creation check box.