Route data resolves this. Running a fleet's route portfolio through an EV routing simulation produces per-vehicle energy demand data: how much charge each vehicle will need at departure tomorrow, given its assigned route, its consumption model, the forecast temperature, and the expected payload. Aggregating this across the fleet gives the actual overnight charging requirement — not what a ratio suggests, but what the routes demand.

From that data, infrastructure sizing becomes a real calculation. How many bays need to support 100+ km of charge per night? What electrical connection capacity is needed if charging is staggered to avoid simultaneous peak demand? Which vehicles can charge from a standard 7.4 kW AC connection and which need faster infrastructure because their routes demand departure at high SoC after returning at low SoC?

The analysis also reveals which routes require mid-route charging — infrastructure that cannot be solved at the depot. For these routes, the routing simulation identifies where along the route a stop is needed and what charge speed is required to fit the stop into the schedule. This is the data that supports either a case for public infrastructure investment at a specific corridor location or a negotiation with a CPO about installing destination charging at a regular delivery point.