Oslo's Municipal On-Street EV Charging Network
Bymiljøetaten promised 350 new public points in 2023 and delivered 122 open to ordinary drivers — the rest went to car-rental fleets. Even the city agrees it needs roughly 7,000 by 2030.
Charging is where electrification stops being a vehicle question and becomes a city one. A bus fleet needs a depot with enough power; a household without a driveway needs a socket on the street it parks in; and both need someone to decide who builds them, who is paid, and what a kilowatt-hour costs the person plugging in.
This guide reads the documented record on charging in the atlas. Fleet electrification itself is covered under public transport, the network behind the socket under smart grids, and the generation feeding it under rooftop solar. Definitions of a microgrid and a smart grid are in the glossary.
Depot charging is a procurement problem. A fleet comes home at night, so the question is contracts and grid capacity, not coverage. Shenzhen electrified more than 16,000 buses on a backbone of roughly 8,000 charge points across some 510 depots and stations — enough to top up about half the fleet at once — and the entry's own lesson is that this was a procurement and charging-infrastructure problem, not a technology one. Eindhoven charges 43 articulated buses by pantograph at a single depot, and had to build its own fleet-monitoring software because the existing diesel scheduling tools could not plan around battery range. Delhi and Mumbai both put the depot on someone else's balance sheet, procuring under gross-cost contracts in which the operator owns and maintains the vehicles and the city pays per kilometre.
Kerbside charging has no such owner. The people who need it are precisely those without a garage, and no fleet manager is going to build it for them. The recurring municipal answer is to hide it in street furniture that already has power: Dortmund led a three-city federal project putting charge points into existing streetlights, reaching 320 points at up to 11 kilowatts; Greenwich did the same in its EU demonstrator, and its entry notes the lamppost model was widely replicated for dense boroughs. Oslo is the outlier that simply built its own: about 2,600 municipal points across roughly 1,000 locations by February 2025, priced and load-balanced through a city app.
Cities announce charge points. Almost none publish how often they are used. The exception in this record is worth more than the announcements: in Dortmund the utility DEW21 describes 5.5 percent utilisation as a very good value for inner-city charging. That single figure is the honest yardstick for the whole genre — and it explains why the atlas treats a count of installed points as an input, not an outcome.
Delhi publishes the other useful number, the gap: 8,849 public charging points and 893 battery-swapping stations by the end of 2025 against an assessed requirement of about 36,150 points. Oslo's entry adds a subtler correction. Its network is routinely folded into Norway's "EV capital of the world" story, but that story is mostly national tax policy — no purchase tax, no VAT, toll and ferry exemptions. The city's contribution is the sockets, and in 2023 its network grew by a net 272 points of which only 122 were newly open to ordinary drivers.
The tender decides more than the technology does. Delhi Transco pooled about 100 public land parcels and awarded them on the lowest service charge to drivers rather than the highest revenue to the state — producing public charging at roughly 2 rupees a unit, and a subsidised home charger available through the electricity distributor. That is a deliberate inversion of the usual municipal instinct to treat a concession as a revenue line, and it is the clearest example in this atlas of a city using a procurement rule to set a consumer price.
Where the concession runs the other way, the city loses the lever. Bangkok's electric fleet is run by a private operator that took over routes the state operator gave up, which moved faster than the public undertaking could — and tied the city's flagship green fleet to one company's balance sheet and its supplier's. Lake Garda's network shows the municipal-utility variant: chargers around the lake powered by the utility's own photovoltaic plants on local schools, owned collectively by the municipalities it serves.
A charge point is a controllable load, and a small group of entries treats it as one. State Grid Shanghai aggregates data centres, building air conditioning and EV charging into a virtual power plant it can call on at peak. Amsterdam's stadium runs a 3 MW battery built partly from used electric-car packs and pilots 15 bidirectional chargers that let consenting visitors' cars feed power back into the building. Copenhagen's Nordhavn testbed paired DC fast charging with battery buffering so the fast chargers did not simply become a new peak. These are pilots, not services — but they are the only entries here where charging gives something back to the network rather than only drawing from it.
Electrification announcements outrun electrification. New York's MTA committed in 2018 to a zero-emission fleet by 2040 and planned 560 battery-electric buses in its first stage; by mid-2025 sixty were in service, averaging 2,111 miles between major failures against 13,000 for diesel and available for a full day of service 60 percent of the time against 91 percent. The retreat is documented only because the operator publishes those two measures as standing performance indicators — which is why this atlas grades the evidence behind an entry rather than the ambition in it. Paris's Autolib' is the older warning: roughly 4,000 electric cars, about 1,080 stations, some 150,000 subscribers, and an accumulated deficit that ended the contract in 2018.
They are different problems with different owners. Depot charging serves a fleet that returns to one place overnight, so it is a procurement question: Shenzhen electrified more than 16,000 buses on a backbone of roughly 8,000 charge points across some 510 depots and stations, and its own lesson records that this was a procurement and infrastructure problem rather than a technology one. Kerbside charging serves people who park on the street and has no single owner, which is why cities build it into existing lampposts — Dortmund's federally funded NOX-Block project and Greenwich's Sharing Cities demonstrator both did exactly that.
No city in this atlas has published a defensible answer, but Delhi has published the gap: 8,849 public charging points and 893 battery-swapping stations by 31 December 2025, against an assessed requirement of about 36,150 points. Oslo, which has run its own municipal network since 2007, had about 2,600 standard points across roughly 1,000 locations by February 2025 and is aiming for 7,000 by 2030.
This is the number that matters and the one operators rarely publish. In Dortmund the utility DEW21 frames 5.5 percent utilisation as a very good value for inner-city charging — which is a more useful description of what success looks like in this genre than any count of installed points.
Whoever writes the tender. Delhi's 2020 policy pooled about 100 public land parcels and awarded them on the lowest service charge to drivers rather than the highest revenue to the state, producing public charging at roughly 2 rupees a unit. Oslo prices and load-balances its own network through a municipal app. Where a private concession supplies the chargers, the city usually sets neither the price nor the expansion pace.
Ask a city three questions about its charging programme. Who does the kerbside network serve — the people with driveways or the people without? What utilisation is the operator willing to publish? And who was the tender written for: the driver, or the treasury? The answers separate the documented cases in this atlas far more sharply than any figure for installed capacity.
Where residents park on the street, charging has to go into the street furniture. All four did that. Only one of them says how often the plugs are actually used — which is the number the whole genre turns on.
| Project | Since | Status | Outcome | Evidence | Run by | Cost | Who built it, and what they admit |
|---|---|---|---|---|---|---|---|
| NOX-Block Lamppost EV ChargingDortmund, 🇩🇪 | 2019 | Live | Ongoing | Official | Utility or public operator, City government | not public | The utility built it into existing streetlights on a federal grant, and then published utilisation — calling five and a half per cent a very good inner-city value. |
| Sharing Cities — Greenwich Smart DistrictLondon, 🇬🇧 | 2016 | Completed | Succeeded | Official | City government | not public | Same idea, EU demonstrator money, and the lamppost approach was copied widely because it avoids new street furniture entirely. |
| Oslo's Municipal On-Street EV Charging NetworkOslo, 🇳🇴 | 2007 | Live | Scaled back | Official | City government, Private company | not public | The city's own agency has run its street network since 2007, and keeps missing its own annual targets — a modest municipal effort folded into a national tax-policy success story. |
| Delhi EV Policy & Discom-Tendered Public ChargingNew Delhi, 🇮🇳 | 2020 | Completed | Scaled back | Official | City government, Utility or public operator | not public | Tendered on the lowest price to the driver rather than the highest revenue to the state, which produced cheap public charging and nowhere near enough of it. |