Modeled levers clear both planning thresholds, while continuity remains unproven
Planning aim: by 31 December 2027, test a public/demo-data portfolio with at least 100,000 modeled active units and at least 95% modeled affected-fleet…
Across five independent modeling layers, the 100,000-unit portfolio target and the 95%-within-24-hours recovery target are both reachable only after authored parameter changes: a…
Full framing
Planning aim: by 31 December 2027, test a public/demo-data portfolio with at least 100,000 modeled active units and at least 95% modeled affected-fleet restoration within 24 hours.
Across five independent modeling layers, the 100,000-unit portfolio target and the 95%-within-24-hours recovery target are both reachable only after authored parameter changes: a stronger 104-week Product candidate and a Comdyn recovery intervention. The result is a planning consistency check, not proof of Vodacom service continuity, field performance, customer demand, an operator commitment, or an integrated end-to-end model.
Public-news source spine
The public record behind the subject
- Read 2026-07-28. Public Vodacom Business support context for M2M/IoT SIMs and managed services. vodacom.co.za · open source →
- Read 2026-07-28. Public operator context on IoT use cases and dedicated APNs. now.vodacom.co.za · open source →
- Read 2026-07-28. Public operator description of NB-IoT use cases; not field-performance evidence. now.vodacom.co.za · open source →
Review 4 more public sources
- Read 2026-07-28. Current public operator reporting context. vodacom.com · open source →
- Read 2026-07-28. Official Free State severe-weather classification context; unrelated to the synthetic power_outage fixture unless separately proved. gov.za · open source →
- Read 2026-07-28. Public provincial severe-weather response context; not fixture calibration. gov.za · open source →
- Read 2026-07-28. Public communications-sector context. icasa.org.za · open source →
Instrument evidence
One exact state for each modeled layer
Chapter 1 · incident-sim
1. Incident — the bounded demo begins with 16 of 33 cells affected
The exact Cell Analysis replay shows 33 modeled cells: 17 active after impact, 11 degraded and 5 offline. This is an immutable, stale Free State synthetic/demo power_outage…
Boundary Decision markers in this state — Affected cells in immutable Free State demo: 16 of 33; Active cells after modeled…
The exact Cell Analysis replay shows 33 modeled cells: 17 active after impact, 11 degraded and 5 offline. This is an immutable, stale Free State synthetic/demo power_outage fixture—not a real Vodacom incident, current condition, infrastructure record, or SLA.
Decision markers in this state — Affected cells in immutable Free State demo: 16 of 33; Active cells after modeled impact: 17 of 33.
Open the analyzed state to inspect the map, raster and pressure assumptions. A changed exposure state must be rerun and registered as a new scene.
The next chapter narrows the same decision through 2. Link — the selected Band-8 operating point clears its one-second radio-link deadline.
Chapter 2 · link-sim
2. Link — the selected Band-8 operating point clears its one-second radio-link deadline
At 3.0 km across 2,000 seeded simulation trials, 99.8% of modeled payloads were delivered and 99.3% arrived within one second (exact result 99.75% and 99.25%). The counter-view…
Boundary Decision markers in this state — Modeled Band-8 payloads delivered: 99.75%; Modeled Band-8 payloads within 1 second:…
At 3.0 km across 2,000 seeded simulation trials, 99.8% of modeled payloads were delivered and 99.3% arrived within one second (exact result 99.75% and 99.25%). The counter-view is strict: this does not establish site availability, fleet recovery, end-to-end NB-IoT continuity, operator performance, or an SLA.
Decision markers in this state — Modeled Band-8 payloads delivered: 99.75%; Modeled Band-8 payloads within 1 second: 99.25%.
Open the analyzed state to inspect the link budget and scenario assumptions; use a new run for any configuration that changes the decision.
The next chapter narrows the same decision through 3. Recovery dynamics — the baseline misses narrowly and the intervention crosses 95%.
Chapter 3 · comdyn
3. Recovery dynamics — the baseline misses narrowly and the intervention crosses 95%
At modeled hour 24, baseline restoration is 94,942 units. A +50% restoration-capacity marker from hour 0 produces 95,323 restored units; the corrected strict optimizer separately…
Boundary Decision markers in this state — Modeled baseline restored at hour 24: 94,942 units; Modeled restored at hour 24 with…
At modeled hour 24, baseline restoration is 94,942 units. A +50% restoration-capacity marker from hour 0 produces 95,323 restored units; the corrected strict optimizer separately finds a triage-rate proxy of 0.182138 for exactly 95,000. These are authored dynamics, not observed recovery performance.
Decision markers in this state — Modeled baseline restored at hour 24: 94,942 units; Modeled restored at hour 24 with +50% capacity marker: 95,323 units.
Open the live model to inspect compartments, timing and transition assumptions. Use a rerun—not a visual edit alone—before changing the conclusion.
With the trajectory bounded, the next chapter tests the product plan that would have to produce it.
Chapter 4 · product-launch-sim
4. Product trajectory — the stronger authored candidate reaches 117,798 units
The unchanged focused catalog baseline reaches only 6,277 modeled active units at week 104; a first authored candidate reaches 79,933 and misses. The visible second candidate…
Boundary Decision markers in this state — Unchanged Product baseline at week 104: 6,277 modeled active units; Target-reaching…
The unchanged focused catalog baseline reaches only 6,277 modeled active units at week 104; a first authored candidate reaches 79,933 and misses. The visible second candidate reaches 117,798, but is a sensitivity scenario—not a Vodacom commitment, forecast, customer count, or observed plan.
Decision markers in this state — Unchanged Product baseline at week 104: 6,277 modeled active units; Target-reaching Product candidate at week 104: 117,798 modeled active units. This is the commercial benchmark the adjacent product option must improve or qualify.
Open the live model to adjust product economics, launch timing, portfolio choices or design levers. Return the changed state for a fresh run and a new scene.
The next chapter narrows the same decision through 5. Readiness index — utilities leads, but only robustness is complete.
Chapter 5 · cidx
5. Readiness index — utilities leads, but only robustness is complete
Under equal authored weights, utilities/smart metering ranks first at 0.716. The robustness workflow resolves a composite score of 0.470 with a modeled 90% interval of…
Boundary Decision markers in this state — Utilities/smart-metering authored readiness score: 0.716; CIDX modeled…
Under equal authored weights, utilities/smart metering ranks first at 0.716. The robustness workflow resolves a composite score of 0.470 with a modeled 90% interval of 0.450–0.491; detect-changes and policy-impact need input, and the six-layer audit is partial. This ranks authored scenarios, not observed customers.
Decision markers in this state — Utilities/smart-metering authored readiness score: 0.716; CIDX modeled composite-score 90% interval: 0.450–0.491.
Open the analyzed state to inspect the dimensions, weights and sensitivity behind the comparison. A changed weighting should produce a new computed state.
With the instrument evidence assembled, the article now moves to the claim-bound conclusions and executive readout.
Claim-bound findings What the evidence supports—and how strongly 8 bound claims
- supported
Vodacom public materials describe Business IoT/M2M, managed connectivity and NB-IoT use cases; those materials establish subject context, not field performance or the private…
Full claim wording
Vodacom public materials describe Business IoT/M2M, managed connectivity and NB-IoT use cases; those materials establish subject context, not field performance or the private size of any fleet.
- supported
In the immutable Free State synthetic/demo power_outage fixture, the exact Cell Analysis state contains 33 modeled cells, with 17 active after impact, 11 degraded and 5…
Full claim wording
In the immutable Free State synthetic/demo power_outage fixture, the exact Cell Analysis state contains 33 modeled cells, with 17 active after impact, 11 degraded and 5 offline—16 affected.
- supported
At the selected modeled Vodacom ZA Band-8 NB-IoT operating point, LinkSim returned 99.75% payload delivery and 99.25% within one second across 2,000 seeded trials; this is…
Full claim wording
At the selected modeled Vodacom ZA Band-8 NB-IoT operating point, LinkSim returned 99.75% payload delivery and 99.25% within one second across 2,000 seeded trials; this is link-level simulation evidence only.
- supported
The authored Comdyn baseline restores 94,942 of 100,000 affected units by modeled hour 24, while a +50% restoration-capacity marker from hour 0 restores 95,323.
- supported
After the strict optimizer repair, the baseline shortfall is not accepted as achieved; the modeled optimizer reaches exactly 95,000 restored units at hour 24 with triage_rate…
Full claim wording
After the strict optimizer repair, the baseline shortfall is not accepted as achieved; the modeled optimizer reaches exactly 95,000 restored units at hour 24 with triage_rate 0.182138 under the stated remaining parameters.
- supported
The explicitly scheduled second Product sensitivity candidate reaches 117,798 modeled active portfolio units at week 104, after the unchanged baseline and first candidate remain…
Full claim wording
The explicitly scheduled second Product sensitivity candidate reaches 117,798 modeled active portfolio units at week 104, after the unchanged baseline and first candidate remain below 100,000.
- supported
Under equal authored weights, the CIDX ranks utilities/smart metering first at 0.716; its robustness result supports only modeled weight stability, while change detection and…
Full claim wording
Under equal authored weights, the CIDX ranks utilities/smart metering first at 0.716; its robustness result supports only modeled weight stability, while change detection and policy impact need input and the six-layer audit remains partial.
- projected
Taken as independent planning layers, the Product and Comdyn scenarios can each cross their proxy threshold after authored parameter changes; they do not jointly prove Vodacom…
Full claim wording
Taken as independent planning layers, the Product and Comdyn scenarios can each cross their proxy threshold after authored parameter changes; they do not jointly prove Vodacom service continuity or a validated end-to-end operating plan.
Executive readout
Decision summary
Supported finding
- Within the declared models, both numeric planning thresholds are reachable, but neither is met by the unchanged baseline: the Product target needs a stronger authored candidate,…
Review 1 full note
- Within the declared models, both numeric planning thresholds are reachable, but neither is met by the unchanged baseline: the Product target needs a stronger authored candidate, and the recovery target needs a modeled operational lever.
Decision boundary
- Every operational number is modeled or taken from a stale immutable demo fixture. The five layers are independent; no live Vodacom telemetry, customer data, field calibration,…
Review 1 full note
- Every operational number is modeled or taken from a stale immutable demo fixture. The five layers are independent; no live Vodacom telemetry, customer data, field calibration, causal design, operator commitment or SLA evidence was used.
Question to resolve
- Before acting, test the assumptions with observed, consented fleet-recovery logs; field measurements at the declared Band-8 operating point; use-case-specific Product inputs; and…
Review 1 full note
- Before acting, test the assumptions with observed, consented fleet-recovery logs; field measurements at the declared Band-8 operating point; use-case-specific Product inputs; and enough periods plus a genuine event design to complete CIDX change and policy verification.
How this was governed
97% autonomous · 2 human checkpoints (gate_a & gate_b)
Verification
- Supported claims7
- Projected1
- Illustrative0
Evidence
- Cited (source)7
- Modeled5
- Observational0