A busy-hour counter is not a congestion verdict

Planning rooms love a single number because it fits on a slide. A PRB utilisation sampled at a vendor’s default busy hour — often a weekday mid-morning — will tell you something true about offices and nothing useful about a Black Sea boulevard on a Friday in August.

In Adjara the load moves. The port approaches fill at different hours from the old town. A ridge site that looks idle at 11:00 can be the only usable server when the waterfront sectors are full of visitors who will not walk inland. If we draw a congestion map from one default hour, we will accuse the wrong rooftops and miss the ones the freeze actually has to fund.

When a radio-access study starts, we ask the planning lead to name three windows: the hour they already believe is busy, a tourist-season evening, and a quiet midweek slice. The third is not decoration. It is how you see whether a cell is always high or only high when the boulevard is. A cell that is high on Wednesday at 10:00 and Friday at 20:00 is a different object from a cell that spikes only with the weekend.

Outage hours have to leave the export before any of this is drawn. A planned microwave swap that lasted ninety minutes will manufacture a “congested” sample if nobody strips it. That is why the first week of a study is a quality note, not a choropleth. We would rather send an awkward query about a missing calendar than hang a red polygon on a site that was honestly dark.

None of this requires a new counter. Operators already have the files. What they rarely have is a sitting in which someone is allowed to say that the default busy hour is the wrong hour for this market. That sentence belongs in the pack, in ordinary language, so the freeze meeting does not reopen the argument from memory.