Hypothesis
In hybrid teams, human review capacity becomes the dominant constraint once agent throughput exceeds it, so delivery time is better predicted by approval-queue utilization than by agent count.
Proposed methodology
- Observe hybrid projects with recorded review and approval events.
- Model approvals as a queue and estimate utilization over time.
- Compare predictors: agent count, review policy, queue utilization.
Required data
- Approval and review events with timestamps and outcomes.
- Work-item lifecycle events.
- Team composition over time.
Evaluation metrics
- Explained variance in delivery time per predictor.
- Escaped-defect rate vs. review policy.
- Calibration of predicted delivery ranges.
Current status
Planned. No data has been collected and no experiment has been registered for this track.
Known uncertainties
- Approval behaviour is organization-specific.
- Observed teams may not be representative.
Source references
- Little, J. D. C. (1961). A proof for the queuing formula L = λW. Operations Research 9(3).