Instrumenting the systems a city runs on, so utilities, transport and infrastructure can be measured rather than guessed at.
Every sector fails in its own way. These are the conditions that decide the design here, long before anyone picks a framework.
Mains power drops out for hours at a time, and a meter that reboots into a blank state loses the billing period with it.
Battery-backed nodes with journaled storage, so a power cut costs you a gap in the graph and never a gap in the ledger.
A node in a culvert or a basement riser has no usable cellular signal, and the ones that do are on prepaid data.
Store-and-forward buffering over LoRaWAN or NB-IoT, with backhaul that batches uploads instead of chattering.
Field visits cost more than the node. A device that needs a technician twice a year has already failed economically.
Watchdogs, staged over-the-air updates with automatic rollback, and remote diagnostics before anyone is dispatched.
A starting point, not a template. The right choice depends on what your deployment actually has to survive.
Tell us what the deployment has to survive and we will come back with scope, timeline and a price. Your enquiry arrives tagged to this sector, so it lands with whoever knows the work.