The Network Estate — 6 archetypes · 47,000 elements
Network elements clustered by stable behavioural patterns. Updated weekly.
Well-configured, modern-SW elements across all four vendors. Performance KPIs consistently green. CM fingerprint conformant to golden baseline. HW age median 3.2 years.
- →NHI stable at 86 (+1 vs last week)
- →No significant alarm activity
- →1,200 elements eligible for SW upgrade to n+1 release
- →Benchmark reference cohort for golden baseline calculations
High-traffic cells in dense urban areas where DL throughput utilisation exceeds 80%. Performance KPIs still acceptable but headroom is gone. Customer-facing risk if traffic grows 5–10% further.
- →NHI declined to 52 (-3 vs last week)
- →DL utilisation >85% on 847 elements this week
- →38 London South Ericsson cells at 81–84% — threshold breach projected 3–4 weeks
- →No config drift — performance issue is traffic volume, not configuration
Nokia BBUs in the North cluster that received a SW upgrade 6 weeks ago. Performance KPIs still mostly green, but CM audit shows 23% have non-default srPeriodicity values and 18% have prachConfigIndex deviating from cluster median. NHI declining 4pts/week — the classic pre-degradation pattern.
- →NHI declined to 67 (-4 vs last week, -16 over 6 weeks)
- →CM audit: 194 elements with srPeriodicity = 40ms (golden baseline: 20ms)
- →Alarm rate increasing: avg 0.8 vs 0.3 pre-upgrade
- →3 elements have migrated to Persistent Degrader this week — escalation risk
- →No field engineer visits yet — purely config-driven
Mature Huawei BTS units installed 2013–2016. HW approaching or past EoL. SW on n-2 or older releases. Performance KPIs are manageable now but the combination of age, EoL exposure, and rising alarm frequency signals a lifecycle cliff.
- →NHI at 44 (-1 vs last week, -8 over 12 weeks)
- →214 elements: HW EoL date within 67 days — no support contract extension confirmed
- →SW upgrade backlog: 712 elements on version released Q3 2021
- →Alarm burden highest of all archetypes — rising 3% week-on-week
Nokia cells that completed SW upgrade 2–4 weeks ago (earlier cohort than Config Drifters). Initial KPI instability normalising. CM fingerprint being actively remediated by the optimisation team. Watch-and-wait status.
- →NHI recovered to 61 (+1 vs last week)
- →CM remediation in progress: 112 elements normalised this week
- →Alarm rate declining: 1.1 avg vs 1.8 peak post-upgrade
- →Projected full recovery to High Performer archetype: 3–4 weeks
Chronic underperformers across multiple vendors with multi-week amber/red PM KPIs, no correlation with open maintenance tickets, and no recent config changes. These elements are degrading for reasons the current model cannot fully explain — likely HW issues or untracked environmental factors.
- →NHI at 38 (-2 vs last week)
- →3 new arrivals from Config Drifter archetype this week — migration escalation
- →12 elements have been in this archetype for >8 consecutive weeks
- →Model confidence on root cause: LOW — recommend field investigation