Three hundred alarms are usually one fault.
Telecommunications operators manage distributed infrastructure: transmission networks, base stations, fibre, compute nodes and customer services. Mantikon is designed to model the network and connect technical events to their consequences for the services customers actually use.
- Next waveDesigned as an application area; a demonstrated world model in this domain comes first.
- Who decidesNOC, network operations, infrastructure management
- Typical sourcesPermitted network metrics, incidents, asset registers, service tickets
- 01
Separate cause from echo
Mantikon establishes probable causal links between events, distinguishes the primary fault from secondary alarms and identifies the services affected.
Causality - 02
Check the evidence
The council examines the technical hypotheses, tests whether the evidence is sufficient and produces an investigation plan rather than an assertion.
Hypotheses - 03
Dispatch with priorities
Repair tasks are distributed with service priority, geography and specialist availability in view.
Plan
A worked
example
Illustrative scenario, not a customer case.
An incident produces several hundred technical notifications. The platform analyses the event stream, the topology and the history of previous incidents, and proposes that one physical break explains most of them.
The critic marks which of the supporting signals are independent and which are repeats from the same element, so confidence is not inflated by duplication.
At the strategic level the same model analyses network load, traffic dynamics and the need for expansion, and helps determine which investments do most for resilience and quality of service.
Scope and
measures
Example workflows
Incident correlation · root-cause investigation · crew dispatch · capacity planning.
Evaluation metrics
MTTD and MTTR · SLA breaches · false-positive rate · share of incidents with a confirmed single cause.
Disclosure
For infrastructure scenarios aggregated technical telemetry is sufficient; analysis of the content of user communications is not part of the platform.
Discuss a network operations use case