List of aggregations
This article describes the aggregations you can use in custom metrics.
AVG([context],[value])
Calculates the average of values grouped according to the [context].
[context]: The context in which the operation is calculated
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, DATE, TIME
Output data type: FLOAT, DATE, TIME
AVGIF([context],[condition],[value],[default])
Calculates the average of values that meet the [condition], grouped according to the [context].
[context]: The context in which the operation is calculated (supported: process, case, event, edge)
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, DATE, TIME
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT, FLOAT, DATE, TIME
Output data type: INT, FLOAT, DATE, TIME
Example: AVGIF(ViewEvents, userName == "Laura", eventCost)
Custom metric can be used anywhere in the application. Returns the average event cost for events associated with the user "Laura".
COUNT([context])
Counts the number of values grouped according to the [context].
- [context]: The context in which the operation is calculated
Output data type: INT
COUNTIF([context],[condition],[default])
Counts the number of values fulfilling the [condition], grouped according to the [context].
[context]: The context in which the operation is calculated
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT
Output data type: INT
COUNTUNIQUE([context],[value])
Counts the number of unique values, grouped according to the [context].
[context]: The context in which the operation is calculated
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, BOOL, STRING, DATE, TIME
Output data type: INT
COUNTUNIQUEIF([context],[condition],[value],[default])
Counts the number of unique [value] that meet the [condition], grouped according to the [context].
[context]: The context in which the operation is calculated
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT
Output data type: FLOAT, TIME
MAX([context],[value])
Returns the maximum of [value], grouped according to the [context].
[context]: The context in which the operation is calculated
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, DATE, TIME
Output data type: INT, FLOAT, DATE, TIME
MAXIF([context],[condition],[value],[default])
Returns the maximum of [values] that meet the [condition], grouped according to the [context].
[context]: The context in which the operation is calculated
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT, FLOAT, DATE, TIME
Output data type: FLOAT, TIME
MAXVAL([context],[value1],[value2])
Selects item with maximum value from [value1] and returns its calculated value defined by [value2], grouped according to the [context].
[context]: The context in which the operation is calculated (supported: process, case, event, edge)
[value1]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME, DATE
[value2]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, STRING, TIME, DATE
Output data type: INT, FLOAT, STRING, TIME, DATE
Example: MAXVAL(EventsPerAttribute, Duration(), userName)
Can be used on the process map Returns the username of a user who worked on the longest event per activity.
MAXVALIF([context],[condition],[value1],[value2],[default])
Selects item with the maximum value defined by [value1] that meets the [condition] and returns its calculated value defined by [value2], grouped according to the [context].
[context]: The context in which the operation is calculated (supported: process, case, event, edge)
[condition]: The condition under which the [value1] is included in the calculation
Data type: BOOL
[value1]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, DATE, TIME
[value2]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, STRING, TIME, DATE
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT, FLOAT, DATE, TIME
Output data type: INT, FLOAT, STRING, TIME, DATE
MIN([context],[value])
Returns the minimum of [value], grouped according to the [context].
[context]: The context in which the operation is calculated
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, DATE, TIME
Output data type: INT, FLOAT, DATE, TIME
MINIF([context],[condition],[value],[default])
Returns the minimum of [value] that meets the [condition], grouped according to the [context].
[context]: The context in which the operation is calculated
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT, FLOAT, DATE, TIME
Output data type: FLOAT, TIME
MINVAL([context],[value1],[value2])
Selects item with the minimum value defined by [value1] and returns its calculated value defined by [value2], grouped according to the [context].
[context]: The context in which the operation is calculated (supported: process, case, event, edge)
[value1]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, DATE, TIME
[value2]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, STRING, TIME, DATE
Output data type: INT, FLOAT, STRING, TIME, DATE
MINVALIF([context],[condition],[value1],[value2],[default])
Selects item with the minimum value defined by [value1] that meets the [condition] and returns its calculated value defined by [value2], grouped according to the [context].
[context]: The context in which the operation is calculated (supported: process, case, event, edge)
[condition]: The condition under which the [value1] is included in the calculation
Data type: BOOL
[value1]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, DATE, TIME
[value2]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, STRING, TIME, DATE
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT, FLOAT, DATE, TIME
Output data type: INT, FLOAT, STRING, TIME, DATE
RANGE([context],[value])
Returns the range (maximum-minimum) of [value], grouped according to the [context].
[context]: Defines the context in which the operation is calculated
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, DATE, TIME
Output data type: INT, FLOAT, TIME
RANGEIF([context],[condition],[value],[default])
Returns the range (maximum-minimum) of [value] that meets the [condition], grouped according to the [context].
[context]: The context in which the operation is calculated
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
[value] - An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT, FLOAT, DATE, TIME
Output data type: FLOAT, TIME
STDEV([context],[value])
Calculates the standard deviation of [value], grouped according to the [context].
[context]: The context in which the operation is calculated
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, DATE, TIME
Output data type: FLOAT, TIME
STDEVIF([context],[condition],[value])
Calculates the standard deviation of [value] that meets the [condition], grouped according to the [context].
[context]: The context in which the operation is calculated
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT, FLOAT, DATE, TIME
Output data type: FLOAT, TIME
SUM([context],[value])
Calculates the sum of [value], grouped according to the [context].
[context]: The context in which the operation is calculated
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME
Output data type: FLOAT, TIME
SUMIF([context],[condition],[value],[default])
Calculates the sum of [value] that meets the [condition], grouped according to the [context].
[context]: The context in which the operation is calculated
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT, FLOAT, DATE, TIME
Output data type: FLOAT, TIME
FIRST([context],[value])
Returns the first [value], grouped according to the [context].
[context]: The context in which the operation is calculated
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME
Output data type: FLOAT, TIME
FIRSTIF([context],[condition],[value],[default])
Returns the first [value] that meets the [condition], grouped according to the [context].
[context]: The context in which the operation is calculated
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME, STRING
[default]: Value to be returned, when condition is not met
Data type: BOOL, INT, FLOAT, STRING, DATE, TIME
Output data type: BOOL, INT, FLOAT, STRING, DATE, TIME
LAST([context],[value])
Returns the last [value], grouped according to the [context].
[context]: The context in which the operation is calculated
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME
Output data type: FLOAT, TIME
LASTIF([context],[condition],[value],[default])
Returns the last value that meets the [condition], grouped according to the [context].
[context]: The context in which the operation is calculated
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
[value]: An attribute name, nested operation, or expression
Data type: INT, FLOAT, TIME
[default]: Value to be returned, when condition is not met
Data type: BOOL, INT, FLOAT, STRING, DATE, TIME
Output data type: BOOL, INT, FLOAT, STRING, DATE, TIME
SELFLOOP([context],[attributeName])
Calculates a count of self-loop events, grouped according to the [context]; short version of COUNTIF(ISSELFLOOP()) expression.
[context]: The context in which the operation is calculated
[attributeName]: An attribute name; if no attribute is defined, the current activity (mining) attribute is used
Data type: STRING
Output data type: INT
LOOP([context],[attributeName])
Calculates a count of loop events, grouped according to the [context]; short version of COUNTIF(ISLOOP()) expression.
[context]: The context in which the operation is calculated
[attributeName]: An attribute name; if no attribute is defined, the current activity (mining) attribute is used
Data type: STRING
Output data type: INT
REWORK([context],[attributeName])
Calculates a count of rework events, grouped according to the [context]; short version of COUNTIF(ISREWORK()) expression.
[context]: The context in which the operation is calculated
[attributeName]: An attribute name; if no attribute is defined, the current activity (mining) attribute is used
Data type: STRING
Output data type: INT
LOOPINFLOW([context],[attributeName])
Calculates a count of loop inflows, grouped according to the [context]; short version of COUNTIF(ISLOOPINFLOW()) expression.
[context]: The context in which the operation is calculated
[attributeName]: An attribute name; if no attribute is defined, the current activity (mining) attribute is used
Data type: STRING
Output data type: INT
LOOPOUTFLOW([context],[attributeName])
Calculates a count of loop outflows, grouped according to the [context]; short version of COUNTIF(ISLOOPOUTFLOW()) expression.
[context]: The context in which the operation is calculated
[attributeName]: An attribute name; if no attribute is defined, the current activity (mining) attribute is used
Data type: STRING
Output data type: INT
NETLOOPGAIN([context],[attributeName])
Calculates the sum of loop gains, grouped according to the [context]; short version of SUM(LOOPGAIN()) expression.
[context]: The context in which the operation is calculated
[attributeName]: An attribute name; if no attribute is defined, the current activity (mining) attribute is used
Data type: STRING
Output data type: INT
MODE([context],[value])
Returns the most common [value] from [context].
[context]: The context in which the operation is calculated
[value]: An attribute name, nested operation, or expression
Data type: STRING, INT, FLOAT, DATE, TIME, BOOL
Output data type: STRING, INT, FLOAT, DATE, TIME, BOOL
MODEIF([context],[condition],[value],[default])
Returns the most common [value] from data elements that meet the [condition] in the [context].
[context]: The context in which the operation is calculated
[condition]: The condition under which the [value] is included in the calculation
[value]: An attribute name, nested operation, or expression
Data type: STRING, INT, FLOAT, DATE, TIME, BOOL
[default]: Default value returned by operator when no element in defined [context] meets the [condition]
Data type: INT, FLOAT, DATE, TIME
Output data type: STRING, INT, FLOAT, DATE, TIME, BOOL
ALL([context],[condition])
Returns true if all values, grouped according to the [context], meet the [condition].
[context]: The context in which the operation is calculated (supported: process, case, event, edge)
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
Output data type: BOOL
Example ALL(CaseEvents, eventCost > 0)
Can be used on case metrics filters or case overview statistics. Returns true if all events in the case have an event cost greater than zero.
ANY([context],[condition])
Returns true if any of the values, grouped according to the [context], meet the [condition].
[context]: The context in which the operation is calculated (supported: case [implicit])
[condition]: The condition under which the [value] is included in the calculation
Data type: BOOL
Output data type: BOOL
Example: ANY(CaseEvents, eventCost > 0)
Can be used on case metrics filters or case overview statistics. Returns true if any event in the case has an event cost greater than zero.