CraftWorld · Access Domain

Goal-driven FTTx,
PON & last mile.

Service delivery is the SLA. Declare provisioning speed, fibre-fault MTTR and per-subscriber availability across FTTx and fixed wireless — and let Ritam (our AIOps engine) hold them, predicting fibre cuts before the subscriber calls.

One splitter fault resolves to one root cause — not a thousand blind tickets.

FTTx / PONFixed WirelessBusiness AccessOTDR
Fibre MTTR · LiveHOLDING
9 cuts predicted / wk1 branch at risk · SPL-B
The problem

The last mile is where SLAs quietly break.

Millions of ONTs, thousands of fibre runs, and a truck roll for every fault. Provisioning is slow and manual, fibre degradation is invisible until the link drops, and one splitter fault becomes a mass-outage guessing game.

Days
to provision
Manual enterprise-circuit turn-up — an SLA liability from day one.
Truck roll
per fault
Field dispatched to the OLT port, not the actual break point.
Blind
fibre health
ONT Rx power degrades unseen until the link finally drops.
Mass outage
guesswork
One PON/splitter fault, thousands of tickets, no shared root cause.
Service delivery is the goal; the truck roll can’t guarantee it. Provisioning must be automated and fibre faults predicted — so MTTR and availability SLAs hold before the subscriber ever notices.
The thesis

The last mile, managed by goals — not truck rolls.

You declare the provisioning, MTTR and availability goals the access network must hold. Ritam automates turn-up, predicts fibre degradation, and localises faults to the exact ODN segment — so field goes to the break, not the guess.

Dispatch-driven — reactive
Goal-driven — pre-emptive
Provisioning is manual, measured in days
Zero-touch turn-up — service live in minutes
Fibre degradation invisible until the drop
ONT Rx-power slope detected 30–60 min ahead
Field dispatched to the OLT, not the break
Fault localised to the exact ODN segment
Mass outage = thousands of blind tickets
One splitter root cause; tickets auto-correlated
“With you in command.” Zero-touch provisioning and ticket correlation run autonomously; field dispatch and network change are approved.
The scope

Five layers, one control plane — each with its SLA.

CraftWorld covers the full access stack from the OLT to the customer premises — monitoring and controlling every layer, and holding a declared service SLA at each.

LayerCraftWorld monitorsRitam controlsSLA / goal
Business accessLeased-line · E-LINE/E-LAN · CoSBandwidth profile, CoS enforceleased-line SLA
Service / BNGPPPoE/IPoE · RADIUS · QoSZero-touch provisioning, resetprovision < 2 hr
FTTx / PONGPON/XGS-PON · ONT Rx · BERONT re-provision, ranging resetONT > 99.9%
OLTPort / PON-card health · uplinkCard failover, traffic re-homeport availability
Fibre plant / ODNOptical loss · splitter · reflectanceFault localisation, field dispatchMTTR < 2 hr

An ONT Rx-power fade, the PON tree it sits on and the OLT port all resolve to one root cause — pinpointing the exact ODN segment, not just the port.

The access tree, as Ritam sees it

From OLT to ONT — the whole PON tree, live.

A correlated view of the PON — OLT, splitters and every ONT, each with its optical health. When a branch’s loss climbs, Ritam localises the fault to that ODN segment and predicts the cut before the link drops.

SPL-B branch 2 · Live

Loss climbing

Rx power−27.4 dBm (budget −28) ▼
Affected ONTs38 on this segment
Predictioncut likely in ~45 min
Ritamfield dispatched to splice point
The capstone

Declare the service goals. Own the MTTR.

The access goals you own — provisioning speed, fibre MTTR, per-ONT availability and mass-outage triage — tracked continuously, with Ritam pre-empting faults before the subscriber calls.

GoalTargetNowStatus
Fibre-fault MTTR< 2 hrs1h 42mOn track
Enterprise provisioning time< 2 hrs38 minOn track
Per-ONT availability> 99.9%99.87%At risk
Mass-outage triage time< 15 min9 minOn track
Fibre cuts predicted (vs reactive)> 80%71%At risk
1h 42m
Fibre-fault MTTR
target < 2 hrs · localised, not searched
1 goal at risk — SPL-B branch 2 loss climbing; Ritam has field dispatched to the splice point, 38 ONTs protected. 9 fibre cuts predicted pre-emptively this week.
Worked example · fibre plant

A fibre cut in 45 minutes — field is already rolling.

TSLAM4b learns each ODN branch’s optical-loss signature. A creeping Rx-power fade means a splice or bend is failing — Ritam localises it and dispatches to the break point before the link drops.

T–45 min

TSLAM4b detects an Rx-power slope on SPL-B branch 2 — 38 ONTs downstream.

T–43 min

ARGUS correlates reflectance & loss profile — a degrading splice, not a config issue.

T–42 min

Ritam localises to the ODN segment (not just the OLT port) and raises a proactive work order.

T–40 min

Field dispatched to the splice point on approval — repaired before any subscriber drops.

30–60 min
Prediction lead before the fibre actually cuts.
38
ONTs protected — repaired on a planned window.
0
Subscribers who lost service.
The splice still failed — but it was repaired on a planned window, so none of the 38 subscribers ever lost service.
How it works

Autonomous across the last mile.

Ritam operates the access network end-to-end — fibre plant, PON, service edge and business circuits — with autonomous, standards-based actions on approval.

Fibre & PON

GPON/XGS-PON OMCI, ONT Rx power & BER
ODN optical-loss & reflectance profiling
Splitter / feeder fault localisation
Ritam: fault-to-segment, field dispatch

Service Edge

BNG PPPoE/IPoE sessions, RADIUS, QoS
Zero-touch ONT discovery & provisioning
Per-subscriber bearer & CoS health
Ritam: auto-provision, session reset

Business Access

Leased-line & E-LINE/E-LAN SLA
Bandwidth-profile & CoS conformance
Per-account circuit availability
Ritam: profile enforce, re-home
Multi-vendor

Native across every access vendor.

One control plane over a multi-vendor access network — GPON, XGS-PON and fixed wireless — correlating fibre plant, OLT and service edge regardless of vendor.

OLT / PON

Nokia · Huawei · ZTE · Calix · ADTRAN · DZS · Ribbon

ONT / CPE

Nokia · Huawei · ZTE · Calix · Sercomm · multi-vendor OMCI

Fixed wireless / BNG

Cambium · Tarana · Cisco BNG · Juniper BNG · RADIUS / TR-069

Interfaces & standards
OMCI (G.988) · TR-069 / TR-369 (USP) · NETCONF / YANG · SNMP · gRPC / gNMI · RADIUS · TR-101 / TR-156 · OTDR integration
Autonomous execution — zero-touch provisioning, ONT re-ranging, fault localisation, field dispatch — all standards-based, on approval.
Product tour

See CraftWorld Access in action.

The PON tree, live — OLT to ONT, with fault localisation to the exact ODN segment.

CraftWorld Access — console screenshot
The measurable delta

What goal-driven access delivers.

From truck-roll-driven operations to pre-emptive, automated service delivery — measurable from the first quarter.

< 2 hrs
Fibre-fault MTTR
fault localised to the segment, not searched
30–60 min
Fibre-cut lead time
degradation caught before the link drops
< 40 min
Provisioning time
zero-touch turn-up vs days, no truck roll
< 15 min
Mass-outage triage
thousands of tickets to one root cause
> 99.9%
Per-ONT availability
held as a goal across the estate
Fewer
truck rolls
field sent to the break, not the guess
CraftWorld · the carrier suite

Three domains, one autonomous engine.

Mobility, transport and access run the same goal-driven model on one engine — Ritam, the carrier hub. Prove one, extend across the network.

See it live

The access console.

PON tree with a degrading branch, fault localised to the ODN segment, field WO raised.

The access console
See it on your estate

See Ritam predict a fibre cut.

A performance-guaranteed PoC on your fibre plant & PON — baselined, KPI-gated, reversible.