Deep Dive: Average Time Calculation

A precise understanding of how average times are calculated is essential for accurate process performance analysis. In Noreja, duration metrics are clearly separated into two analytical layers: the Case Instance Level, where individual process executions are evaluated, and the Model Level, where data is aggregated across multiple instances. This structured distinction ensures transparency in how timestamps, transition durations, and cycle times are computed—allowing analysts to drill down into specific cases while simultaneously identifying broader performance patterns and outliers across the entire process landscape.

We distinguish between two levels of duration analysis:

Case Instance Level – where each individual process instance is analyzed.

Model Level – where information is aggregated across multiple instances, e.g., in a process model, variant model, or case model.

1. Case Instance Level

At this level, we calculate:

Activity Timestamp
Each activity within an instance receives a start timestamp, representing the moment the activity begins.

Transition Duration (Relationship)
For each relationship between two activities (e.g., "A leads to B"), we calculate the difference between their start timestamps:
Transition Duration = Source Activity Start Timestamp → Target Activity Start Timestamp

Example: Transition Duration (Relationship)

Case Instance Cycle Time
The total runtime of a case instance is calculated as the time span between the earliest and latest start or change timestamp within that instance:
Cycle Time = MIN(Start/Change) → MAX(Start/Change)

2. Model Level (Process / Variant / Case Model)

At the model level, we aggregate data from multiple instances and derive the following metrics:

Activities
For each activity, we aggregate the start timestamps across all included instances and display:
MIN / MAX / AVG of all start timestamps per activity

Relationships (Transition Durations)
Similar to the instance level, but aggregated across all cases:
MIN / MAX / AVG of all transition durations between activities

Example: Relationships (Transition Durations)

Model-Level Cycle Time
We also aggregate the individual cycle times of all case instances:
MIN / MAX / AVG of all case instance cycle times

This logic enables precise analysis of individual process flows while also revealing broader patterns and outliers at an aggregated level. By clearly separating both views, we ensure full transparency throughout the analysis.

Deep Dive: Average Time Calculation

1. Case Instance Level

2. Model Level (Process / Variant / Case Model)

Was this article helpful?