GeistBench category
Process modelling
Process models, process-mining event logs and the analyses built on them, graded only where the outcome can be checked by code.
This page defines a category, not the evaluated scope of a release. See the release coverage for capabilities actually assessed. All categories
Systems and records
- Process models and diagrams
- Process-mining event logs
- Charts and analyses built from process data
What the tasks involve
- Change a process model so it contains the required elements and connections
- Derive figures from an event log and report them
- Create an analysis or chart from process data with the required content
Roles and handoffs
Model changes follow the same rules as other writes: the acting role decides whether a change is saved directly or proposed for review.
What a correct outcome means
The saved model or analysis has the required structure and values, reported figures match the event log, and nothing else changed.
How it is checked
The checker inspects the stored model and derived records for the required elements, connections and values, and recomputes figures from the log.
What this category does not test
- Diagram layout, naming style or overall modelling quality
- Whether a model is the best representation of a real process
- Process-mining conclusions that need interpretation
FAQ
Frequently asked questions
- What does the GeistBench Process modelling category test?
- Process models, process-mining event logs and the analyses built on them, graded only where the outcome can be checked by code.
- How are Process modelling tasks graded?
- The checker inspects the stored model and derived records for the required elements, connections and values, and recomputes figures from the log.
- What does the Process modelling category not test?
- Diagram layout, naming style or overall modelling quality. Whether a model is the best representation of a real process. Process-mining conclusions that need interpretation.
- Where are the Process modelling scores?
- Category scores are published per release on the GeistBench leaderboard, with 95% intervals. This page does not repeat them. A release may not include every category; then the category has no score in that release.