Smart Water

The proven approach is sectorization. The network is divided into district metered areas, or DMAs: hydraulically isolated zones of roughly 500 to 3,000 service connections, following IWA guidance. A flow meter at each DMA inlet measures everything entering the zone, and pressure sensors cover critical points. The platform compares inflow against expected demand around the clock, so a new burst shows up as a step change in the balance long before water surfaces on the street.

Continuous telemetry turns leak detection from periodic acoustic surveys into permanent monitoring. When a zone starts drifting, crews go out with correlators or ground microphones to pinpoint the leak, spending field time on a confirmed anomaly instead of sweeping the whole network.

Minimum night flow, or MNF, is the lowest inflow into a DMA, normally recorded between 2:00 and 4:00 in the morning, when legitimate consumption is at its lowest and network pressure is at its highest. Subtract an estimate of genuine night use, such as households and permanent commercial loads, and most of what remains is leakage.

Because the measurement repeats every night, the trend matters more than any single value. A rising MNF in a zone that used to be stable is one of the earliest and cheapest signals of a new burst or growing background leakage. An IoT platform automates the analysis by storing every reading and alerting when tonight's minimum exceeds the zone baseline.

Low power wide area networks are the default choice. LoRaWAN and NB-IoT both reach battery devices in meter pits, chambers and remote sites, and devices transmitting a few readings per day routinely reach battery lives of 10 years or more. LoRaWAN lets a utility run its own private radio network with gateways it controls, which helps in rural areas with weak cellular coverage. NB-IoT rides on licensed carrier networks, so there is no radio infrastructure to maintain, just a SIM and data plan per device.

Thinger.io ingests both: NB-IoT devices usually speak MQTT or lightweight HTTP directly, while LoRaWAN traffic arrives through The Things Stack or ChirpStack integrations with payload decoding.

Yes, and in practice that is how most projects start. Installed instruments usually expose pulse outputs, 4-20 mA signals or Modbus registers. A low cost gateway or RTU reads them locally and forwards the data over MQTT or HTTP, so the meters keep working exactly as before. Smart meters already on LoRaWAN or NB-IoT connect through their network servers without touching the hardware.

Integration also runs in the other direction. Thinger.io exposes every stored reading through a REST API and can push events from rules, so validated telemetry can feed an existing SCADA historian, GIS or billing system instead of creating another data silo.

For deeper control room integration, the built-in Node-RED plugin bridges industrial protocols such as OPC UA and others available in its ecosystem, so field data flows into the platform and curated data flows back to your SCADA, GIS or billing systems.

Both models are common, and the decision usually comes down to procurement and data governance rather than technology. Cloud instances start faster, need no server administration and scale with the device fleet, which fits pilots and small and mid size utilities. On premise or private cloud deployments keep telemetry inside the utility's own infrastructure, which some public tenders and critical infrastructure policies require.

Thinger.io runs identically in either mode, and licenses include white label branding, so an operator or engineering firm can serve several municipalities under its own name and domain, with each customer's devices and dashboards kept separate.

With off the shelf hardware, a small pilot takes weeks, not months. A typical scope is one or two DMAs: flow meters at the inlets, a few pressure points and a level sensor on the feeding tank. On the platform side, device provisioning, dashboards and night flow alerts take days, so the schedule is driven by fieldwork such as chamber access, meter installation and radio coverage checks.

A good pilot covers at least one billing cycle, so metered inflow can be compared against billed consumption and the zone's non-revenue water quantified with real data. That number justifies scaling out: studies published by the IWA put global non-revenue water at around 126 billion cubic meters per year.

Yes. Thinger.io is an ISO 27001 certified company and every device connects through authenticated, encrypted channels, with per-device certificates over mutual TLS where the hardware supports it. Critical deployments can run on an isolated private cloud or fully on premise, with EU data residency and managed backups. Cloud, private cloud and on premise follow the same security model, so utilities meet their compliance requirements without changing how the solution works.

It depends on the number of sensors, the sites covered, the technical requirements and the integration work, so most projects are quoted individually. As a reference, our Private Cloud plans are published on the pricing page and start at 49 euros per month for a dedicated instance, which is usually enough to run a first pilot. Enterprise and on premise deployments are sized per project.