Enterprise AI Administration
Congestion Risk Model
Model Details · Telecom AI Model Registry
Model Details
Congestion Risk Model
Algorithm: Gradient Boosted TreesREGISTERED
- Code
MDL-CONGESTION- Version
- v0.7
- Owner
- Head of Network Planning
- Governance owner
- Network Risk Board
Governance configuration shown for review and validation. Read-only.
Business purpose
Predict cell congestion ahead of capacity breach
Production-recommended design specification. This record does not mean the model has been trained, fitted, deployed or validated.
Technical specification
- Model family
- —
- Algorithm
- Gradient Boosted Trees
- Implementation library
- —
- Model type
- —
- Explainability method
- —
Implementation governance
- Governance status
- REGISTERED
- Implementation status
- NOT IMPLEMENTED
- Validation status
- VALIDATION REQUIRED
- Lifecycle stage
- Design
- Current performance
- Not yet measured
- Refresh cadence
- Weekly
- Last reviewed
- 20 Aug 2026
- Next review due
- 05 Oct 2026
- Historical sample reference
- Recall 0.74 — earlier sample figure, kept for provenance only; not a validated model result.
Business
- Mission area
- Network Performance & Risk
- Decision journey
- Network Performance & Risk
- Target variable
- Congestion probability (14 days)
- Output
- Cells needing capacity action
Input features
Governed variables, in feature-rank order
- 1.Site congestion indexPeak utilisation score per site
- 2.capacity_utilization_pctBusy-hour capacity utilisation of a site.
Related Intelligence
Linked governed records
KPIs (1)
Business rules (1)
Formulas (0)
None recorded.
Source mappings (1)
Simulation models (1)