Network operations and assurance teams

Fix the incident that hurts the most customers, through the change process

Forty alarms fire in ten minutes across three regions. The queue is ordered by arrival time, and the cell serving a hospital is fourth.

Request a working session The product that runs it: Network Operations & Autonomous Networks

Read the journey; a working session shows it running.

Seen by the customer · Operations console

From the record, not a run: what would reach them, on Operations console, SMS.

The case

Annotated from the record

What happened

Forty alarms fire in ten minutes across three regions and the queue is ordered by the time each arrived.

Forty alarms fire in ten minutes across three regions. The queue is ordered by arrival time, and the cell serving a hospital is fourth.

What it reads

The alarm set and its correlation
Topology and the customers behind each cell
Maintenance freezes and the safety policy
The vendor controller's response

Checks before anything is sent

Telemetry quality, topology match, maintenance freeze, safety policy and vendor controller checks pass.

Every option considered, including doing nothing

The incident with the most customers and revenue behind it, firstConsidered
A tune or a traffic rehome the controller can apply on approvalConsidered
A dispatch, where the fault needs a visitConsidered
No action, where the alarm is a known conditionConsidered

Doing nothing is one of the outcomes, and it is recorded with its reason. The options are the record's own; a run decides between them.

Evidence record

Preview

Reference: The reference of the saved run.
Decision: One of the options, including no action, with the reason.
Checks: Telemetry quality, topology match, maintenance freeze, safety policy and vendor controller checks pass.
Evidence kept: Alarm set
correlation
customer impact
recommendation
approval or change ticket
controller response
before/after KPIs
The measures: Time to restore for the incidents that touched the most customers, and repeat incidents on the same cells.
Connection: Written on every record: the systems the run read, and the sample it ran on.

The shape of the record, with no values. A reference appears only after a run is saved.

The improvement, and how it is measured

Hypothesis

An incident worked in order of the customers it hurts restores the most service per hour; one worked in order of arrival restores the oldest alarm.

Modelled

The mechanism that could produce the result, stated; the figure is measured in the pilot.

Observed

Measured in a pilot’s validate stage, on your systems, against a comparison agreed first.

The result, the mechanism, the measurement, and the record’s five answers

Desired result

Improved service and reliability; lower cost/energy; better CAPEX prioritisation; reduced MTTR

Mechanism

Collect → validate topology → detect/forecast → quantify impact → recommend → human/controller approve → execute → verify service/economic result → learn

Measurement

Matched site/cell comparison, stepped-wedge rollout or before/after with seasonality controls; technical and economic acceptance criteria

Measured on: Service experience; utilisation; incidents; MTTR; repeat fault; avoided congestion; cost; energy; affected customers; incremental/retained value

01
Desired result
The result wanted.
Improved service and reliability; lower cost/energy; better CAPEX prioritisation; reduced MTTR
02
Mechanism
The mechanism that could produce it.
Collect → validate topology → detect/forecast → quantify impact → recommend → human/controller approve → execute → verify service/economic result → learn
03
Evidence required
The records kept to show it.
Telemetry snapshot; topology/version; recommendation; approval/change ticket; controller response; before/after KPI; finance validation
04
Costs and risks
The cost, and the ways it can go wrong.
Costs: Implementation + annual licence by network scale/use case; optional managed analytics; verified savings only with agreed finance baseline
If it fails: Default to observe/recommend; never make unapproved network change; open incident; preserve last-known-good config and audit trail
05
Measurement approach
The measure that shows it helped.
Matched site/cell comparison, stepped-wedge rollout or before/after with seasonality controls; technical and economic acceptance criteria
Measured on: Service experience; utilisation; incidents; MTTR; repeat fault; avoided congestion; cost; energy; affected customers; incremental/retained value

Measured in the pilot's validate stage, on your systems, with the comparison agreed first. Every line above is the record's own design intent.

The business side

The change: Time to restore for the incidents that touched the most customers, and repeat incidents on the same cells.
Running cost: Implementation + annual licence by network scale/use case; optional managed analytics; verified savings only with agreed finance baseline
What evidence supports it: Alarm set, correlation, customer impact, recommendation, approval or change ticket, controller response and before/after KPIs.

The journey in detail

Step by step, from both sides

Step	What the customer experiences	What the operator does
The momentThe moment that started it.	Forty alarms fire in ten minutes across three regions and the queue is ordered by the time each arrived.	Reads the alarm set and its correlation, topology and the customers behind each cell, maintenance freezes and the safety policy, the vendor controller's response.
The decisionThe decision to be made.	The customer's screen is unchanged.	Correlates the alarms into incidents, ranks them by the customers and revenue behind each, and recommends the remedy the change policy allows.
The safeguardsThe conditions that stop it.	Still unchanged: the action waits until every check has passed.	Telemetry quality, topology match, maintenance freeze, safety policy and vendor controller checks pass.
The actionThe action that reaches the customer.	Dispatch, tune or rehome through the controller on approval, and tell the affected customers what is being done. Reaches them on Operations console, SMS.	Recommends. The system that holds the right confirms, charges or provisions.
When it goes wrongThe failure, and the recovery.	The recommended rehome moves traffic onto a cell already at capacity, and the remedy causes a second incident.	The capacity check refuses the rehome, the recommendation falls back to a dispatch, and the before-and-after KPI for the first remedy is kept so the rule is corrected.
The resultThe change it made.	What changed for them is what is counted; nothing else is claimed.	Time to restore for the incidents that touched the most customers, and repeat incidents on the same cells.
The proofThe proof anyone can check.	Can be answered for, later, from the record.	Alarm set, correlation, customer impact, recommendation, approval or change ticket, controller response and before/after KPIs.

Operator systems remain authoritative. HeuriTel reads these to propose an action; it does not decide on their behalf.

The failure, and the recovery

What goes wrong

The recommended rehome moves traffic onto a cell already at capacity, and the remedy causes a second incident.

What happens then

The capacity check refuses the rehome, the recommendation falls back to a dispatch, and the before-and-after KPI for the first remedy is kept so the rule is corrected.

The products in this journey

Runs the decision
Network Operations & Autonomous Networks
Anomalies, correlated alarms and predicted faults are ranked by the customers and revenue behind them, and each remedy goes to the controller as a recommendation with its evidence.
Contributes
Network Investment & Intelligence
the same telemetry, read for capital rather than for faults
Contributes
Contact Centre & Customer Care
where a customer affected by a fault is told what is being done
The definition it demonstratesHeuriTel Network Incident Prioritisation
Results it movesTurn network capabilities into enterprise services · Reduce fraud and identity risk

Available to see today

Read the journey; a working session shows it running.

Request a working session