Without data
2 days
of overflow before a complaint
- MonCrews follow the fixed round, bin after bin
- TueThe container by the market overflows
- WedResidents complain to the council
IoT Solutions
Digitize lighting, parking, air quality and transport on one platform that scales with the city
Applications
The projects teams in the sector deliver most often once their devices are connected.
Dim, schedule and monitor public lighting per street, and detect failed luminaires without patrol rounds.
Publish live occupancy of parking spots to guide drivers and cut traffic circling the block.
Build a dense network of low-cost stations and map pollution levels neighborhood by neighborhood.
Measure container fill levels and route collection trucks only where they are needed.
Track buses, bikes and shared vehicles in real time to improve service and inform citizens.
Automate park watering and monitor fountains, tanks and coastal water from the same platform.
Why it matters
Fixed rounds empty half empty bins and miss the full ones. Fill sensors let each day's route follow what the street actually needs.
2 days
of overflow before a complaint
Same day
from full container to emptied
In the field
Parking sensors, waste containers, street lights and noise and air quality stations report to a single account: each service sees its own data, the city sees all of it.
Street instrumentation
What urban service teams actually deploy across real streets, with the ranges and radio interfaces you already work with.
Tells the city which bays are free right now, so drivers stop circling the block and enforcement goes where it matters.
Reports how full every container is, so trucks visit the ones that need a pass and skip the ones that do not.
Puts a number on the neighbourhood soundtrack: continuous LAeq that turns noise complaints into ordinance evidence.
Maps pollution street by street with gas cells and a particle counter small enough to hang on any lamppost.
Turns every luminaire into a managed asset: dim it by schedule, meter its energy and catch failures without night patrols.
Counts people, not faces: anonymous footfall that shows how streets, parks and events are really used.
The city's radio backbone: one rooftop box that hears thousands of battery sensors and hands their data to the platform.

STREET BY STREET, CONNECTED
Live dashboard
A live app built with the dashboard SDK, running on a Thinger.io demo server. It is read-only, so explore it freely.
ConnectingRead-only demo account
The platform
Field devices connect to Thinger.io over standard protocols. There you manage every device, store its data and trigger alarms.
A smart city is dozens of projects from different vendors, and they all land on one common data layer.
City projects are delivered by integrators and operated by public teams, and the platform serves both.
Success stories
Good to know
What teams ask before a pilot, answered with the numbers they need.
Each bay gets a flush-mounted or surface-mounted sensor. The base technology is a magnetometer that senses the distortion a car produces in the local magnetic field, and the best devices add radar as a second, independent detection principle to filter out interference from passing traffic, power lines or trams. With that dual approach, Bosch reports over 96% accurate detection of parking state changes in field tests covering more than 2,000 sensors and 46 car types.
Each sensor sends occupancy events over LoRaWAN or NB-IoT and runs 5 to 10 years on a battery. The platform aggregates events per street or zone, so live availability can feed guidance panels, enforcement teams and citizen apps, and the same data quantifies real occupancy for parking policy.
Continuous monitors use sound level meters built to IEC 61672, usually class 2 for permanent urban networks, which allows a tolerance of about 1.4 dB at the 1 kHz reference. The core indicator is LAeq, the A-weighted equivalent continuous level over an interval, complemented by long-term indicators such as Lden and Lnight used in EU noise mapping.
As a reference for limits, the EU Environmental Noise Directive sets reporting thresholds at 55 dB Lden and 50 dB Lnight, while the WHO 2018 guidelines recommend staying below 53 dB Lden and 45 dB Lnight for road traffic noise. Municipal ordinances translate this into per-zone day and night limits, and a network of connected monitors turns neighbour complaints into evidence: the platform stores every interval, compares it against the zone limit and alerts the moment a nightlife area or works site goes over.
An ultrasonic sensor under the container lid measures the distance to the waste surface a few times a day and reports fill level over LoRaWAN or NB-IoT. Instead of fixed routes, trucks are dispatched only to containers above a threshold, typically around 70 to 80% full, with routes recalculated from the latest readings.
The savings are documented because the baseline is usually poor: in one municipality analyzed by Sensoneo, bins averaged only 24% full at collection time, and adjusted routes and frequencies produced up to 63% cost savings. The city of Nitra reports saving up to 30% on collection costs, and Prague has deployed more than 3,000 fill sensors to optimize frequencies. A realistic expectation for most cities is a 20 to 30% reduction in collection cost, plus fewer overflowing containers between passes.
Both work, and many cities end up with a mix. A municipal LoRaWAN network is city-owned infrastructure: a few dozen gateways on rooftops or masts cover a mid-size city, with extra density where sensors sit underground or flush in the pavement, there is no per-device subscription, and the same network serves parking, waste, noise and metering devices from any vendor. The tradeoff is that the city or its integrator operates the gateways and network server.
NB-IoT rides on licensed carrier networks, so there is no radio infrastructure to maintain, just a SIM and data plan per device, with good penetration into underground containers and chambers. It fits scattered deployments and cities that prefer opex over owning infrastructure. Thinger.io ingests both: LoRaWAN through The Things Stack or ChirpStack integrations with payload decoding, and NB-IoT devices directly over MQTT or lightweight HTTP.
Less than most procurement teams expect, because the platform is licensed by instance rather than by a per-project development. As a reference, Private Cloud plans are published on the pricing page and start at 49 euros per month for a dedicated instance, which is enough to run a pilot such as one street of parking sensors, a waste collection district or a small noise network. Hardware is the main variable: city sensors typically cost tens to a few hundred euros per point plus installation.
From there, growth is incremental: the same instance absorbs more devices and services, and enterprise or on-premise deployments are sized per project when a tender requires dedicated infrastructure.
Yes, and that is the point of a horizontal platform. Parking, waste, lighting, noise and air quality devices coexist in the same instance, each with its own dashboards, data buckets and alert rules, so every new service adds devices instead of adding another vendor silo. Per-user permissions give each municipal department exactly the visibility it needs, from a full operations view to a single public dashboard.
For integrators and utilities serving several municipalities, licenses include white label branding and multi-tenant projects: each town gets its own users, devices and dashboards under the integrator's brand and domain, kept separate from the rest, while the integrator operates everything from one place.
Next step
Tell us about your devices and your deployment, and we will map them to the platform. Or start on a free account and see it first.