A Tier-1 fiber operator's inventory was scattered across the systems of record it inherited from acquisition after acquisition. In a proof of concept run entirely on the operator's own data, NetSingularity correlated those silos into a single digital twin, and surfaced real, monetizable gaps hiding in the network.
The Challenges
Each acquisition brought its own network, and its own systems, formats, and record-keeping habits. The result was an inventory that no single tool could see end to end, and that the operator itself knew was inaccurate.
Passive plant lived in GIS platforms (GE Smallworld / IQGeo), active inventory in an M6-class system, faults and tickets in ServiceNow, plus spreadsheets carried over from each acquired business. No layer joined passive to active.
Circuit chains had missing cables, equipment and segments. Capacity, run-rate and exhaust dates couldn't be trusted, so planning and sales decisions were made on incomplete information.
Fiber strands and card/port capacity freed up by churn or decommissioning weren't reliably visible, so idle assets that could be resold were effectively lost on the books.
Building an A-to-Z picture of a single circuit meant stitching records across systems by hand: expensive, error-prone, and impossible to do at network scale.
The Approach
Rather than a canned demo, the POC ingested the operator's actual passive inventory and enriched it with public data, then built a live digital twin on the NetSingularity platform.
Structures, routes, cables, connectivity and user/order data were loaded through the ETL layer, then enriched with public census, demographic and competitor data.
A GIS-based twin with layered aerial and underground views, A-to-Z circuit views, and building → rack → shelf → card → port drill-down, with passive and active in one model.
The correlation engine joined records across the silos and ran automated analysis for missing inventory, free capacity, anomalies and market opportunity.
Aerial and underground plant (sites, cables, conduit, structures, spans and equipment) renders together on a live GIS twin, so any customer circuit can be traced end to end across the network.
Endpoints, fiber and strand counts, the structures a circuit connects and its install history are correlated from the underlying systems of record, giving the trusted view that makes missing segments and anomalies impossible to miss.
Results in the Field
Because the platform ran on production data, the findings weren't hypothetical. Each one is either a data-quality issue to fix or a real inventory opportunity to act on.
End-to-end circuit views exposed chains with missing cable, equipment or segment records, flagging where the data (or the network) has a real gap.
Cable-level views revealed available fiber with no segment consumption: idle strands that can be reclaimed and resold instead of trenching new build.
By correlating card and circuit data, the platform inferred which ports are truly in use versus available, turning guesswork into a utilization view.
Single-circuit anomaly views isolated records that indicate either bad data or a genuine inventory issue: a prioritized cleanup list, not a spreadsheet audit.
Inside the Platform
Beyond the analytics findings, the POC exercised the day-to-day workflows an operations team lives in.
Passive-to-active mapping with populate, modify and delete across the model.
Create cables, manholes, splice enclosures, conduit, panels, ports and poles, on a tablet in the field.
Design a new customer fiber span that consumes fibers in a cable, with diversity and shortest-path logic.
Aerial↔underground route changes, structure-type changes, and relocation with configurable workflow.
Cable specs, glass type per strand, and locating the single strand a customer uses.
Consumption, run-rate and estimated exhaust date at the cable level.
From a building down to a specific card for inventory, utilization and customer association.
Who changed what and when, on any object in the model.
Cable data plus public demographics and competitor data to target new and upsell opportunities.
The Impact
Confidence that the platform performs against messy, acquisition-era production data, the hardest test there is.
Passive and active inventory joined across previously siloed systems, viewable end to end.
Stranded fiber and free card/port capacity become sellable inventory instead of invisible sunk cost.
The same correlation engine drives closed-loop automation for auto-ticketing, MOPs and reconciliation as a next phase.
The Platform
A single, cloud-native platform spanning OSS, BSS and service management, with a correlation engine, GIS digital twin, ML/analytics studio and closed-loop automation built in, so inventory intelligence and action live in one place.
GE Smallworld / IQGeo
Passive GIS inventory
M6-class OSS
Active inventory
ServiceNow
Fault & ticketing
Public / FCC data
Demographic enrichment
The Path Forward
Stand up the platform in the operator's own cloud environment.
Connect remaining systems of record for a fully aggregated, reconciled inventory view.
Add functionality and analytics against the growing data set as priorities emerge.
Move toward closed-loop automation for reconciliation, ticketing and MOPs.
If mergers, multiple systems, or years of manual record-keeping have blurred your view of the network, a NetSingularity proof of concept will show you the truth, using your own data.
Every IMSI, ICCID and number range the operator owned lived in shared spreadsheets, maintained by hand and reconciled by memory. One platform became the database of record for every SIM in the country.
A network of hundreds of thousands of sites generating ~1.25 crore alarms a day, handled by static correlation rules and manual RCA. Six coordinated AI capabilities turned that flood into unique actionable incidents, without a single byte leaving the operator's estate.
Thousands of change requests a month across hundreds of change categories, assessed by checklist and delivered by hand. Seven coordinated flows turn each CR into a risk-scored, simulation-backed, auditable operational outcome.
Case study based on a completed proof of concept. Customer identity anonymized.
