State-Aware OCPM Evaluation Logs
Description
This dataset provides two synthetic, paper-scale object-centric event logs for evaluating state-aware object-centric process mining. The scenarios cover inventory management, with stock-policy states along item-location lifecycles, and manufacturing with predictive maintenance, with operational states along machine lifecycles.
State-Aware Object-Centric Process Mining: Two Synthetic OCEL 2.0 Evaluation Logs
The logs support the evaluation of deterministic state expressions, operational-state validity, state-aware patterns and transitions, state-aware object-centric directly-follows graphs, transition KPIs, and unsupervised execution-state detection. Both logs contain synthetic data only.
Logs
| Scenario | Events | Objects | Event types | Object types | Leading object type |
|---|---|---|---|---|---|
| Inventory management | 46,256 | 18,506 | 24 | 10 | ItemLocation |
| Manufacturing and predictive maintenance | 43,458 | 3,578 | 30 | 11 | Machine |
The inventory log covers stock, demand, replenishment, receipt, transfer, adjustment, policy-change, and data-gap behavior. The manufacturing log combines production, telemetry, alarms, quality holds, failures, maintenance response, repair, inspection, testing, restart, and recovery.
Files
Each scenario is available in all five OCEL serialization formats:
- XML (
.ocel.xml) - JSON (
.ocel.json) - SQLite (
.ocel.sqlite) - CSV (
.ocel.csv) - Bundled ZIP (
.ocel.zip)
The files include typed event and object attributes, event-to-object relationships, object-to-object relationships, and UTC timestamps. The deposited logs contain the observable attributes required by the state expressions, but no precomputed state or hidden-regime labels.
The dataset is designed for use with any OCEL 2.0-compatible reader. In FlowVault, use ItemLocation as the leading object type for the inventory scenario and Machine for the manufacturing scenario.