Every capacity on this page starts from a deliberately conservative assumption: an ordinary surface bike path. Yet the tunnel has no pedestrians, no red lights, no intersections; it welcomes electric bikes and enforces a minimum speed. Each of these factors only increases the figures below. These are floors, not ceilings.
Two questions, two scales. How many bikes pass through one segment per hour? And how many people can the entire 150 km network move in one hour? We answer both with the formulas, then test them against what is actually measured in the world's cycling cities.
At equal width, the bike moves far more people than the car. And the tunnel, freed from surface friction, sits at the top of the global range — where an ordinary path stays capped.
1. The throughput of a segment
It all starts from the usable width of a cyclist. A bike is 60 to 75 cm wide; adding the balance margin, we count about 1.0 m per file. The tunnel path is 3.20 m, i.e. 1.6 m per direction: enough to fit a comfortable file with its passing margin.
The global reference, the Highway Capacity Manual, uses 1,700 to 2,500 bikes/h per lane (lanes of 0.9 to 1.2 m). From this we derive the throughput of a segment:
Throughput of a segment
per direction (1.6 m) = ~1.3 files × ~1,500 bikes/h = 1,500 – 2,000 bikes/h
two-way (× 2 directions) = 3,000 – 4,000 bikes/h (up to ~5,000 at peak)
| Regime | Per direction (1.6 m) | Total (3.2 m) |
|---|---|---|
| Comfortable (to plan for) | ~1,000–1,500 bikes/h | ~2,000–3,000 bikes/h |
| Sustained peak | ~1,500–2,000 bikes/h | ~3,000–4,000 bikes/h |
| Theoretical maximum (saturated) | ~2,500 bikes/h | ~5,000 bikes/h |
Real-world anchor. In Copenhagen, a two-file path of only 2.35 m already carries ≈ 3,000 bikes/h, with measured peaks of 20 bikes in 10 seconds at 21 km/h. In London, the Blackfriars Bridge path moves ≈ 2,000 people/h in a single direction over barely 2 m of width — five times more than the adjacent car lane.
2. The capacity of the entire network
At the scale of the 150 km, we no longer reason segment by segment but by renewal. The logic comes in two steps: how many cyclists ride at the same time, and how often they renew.
Network capacity (at 20 km/h, average trip 10 km)
cyclists present N = density × lane length
= (50 to 75 /km/direction) × (150 km × 2 directions) = 15,000 – 22,000 cyclists
renewal = speed ÷ distance = 20 ÷ 10 = 2 times/hour
network throughput = N × renewal = 30,000 – 45,000 people/hour
Compact form: Throughput = 2 × Q × L ÷ d = 30 × Q, where Q is the per-direction throughput of a segment, L = 150 km and d = 10 km. With Q ≈ 1,000–1,500, we land exactly on 30,000 to 45,000/h. Two methods, same result.
What this means against demand: the market study targets ≈ 57.5 million trips per year, i.e. ~157,000 per day. At a typical peak of ~11–12% of the daily total, that gives ~17,000 to 19,000 trips in the peak hour for the whole network. Capacity (30,000–45,000/h) therefore represents 2 to 2.5 times the projected peak demand: the network would run at 40–60% of its capacity, even at its busiest.
The path is never the bottleneck. The limiting factor is not the open tube, but the access points — elevators, ramps, station turnstiles. That's where you size, not on the path, which keeps a huge margin.
3. Why the tunnel exceeds these figures
Here is the decisive point. The published capacities — HCM, Copenhagen, London — are practical capacities, measured on real surface paths, degraded by pedestrians, traffic lights, weather and the speed gap between cyclists. The tunnel removes each of these degradations. Its real capacity therefore sits at the top of these ranges — or even above.
Ordinary surface path
- Pedestrians and crossings that cut the flow
- Red lights and intersections: the real ceiling
- Rain, wind, ice: speed collapses
- Hills: large gap between slow and fast cyclists
- We stay at practical capacity (~1,280/lane)
Dedicated bike tunnel
- No pedestrians: 100% of the space for riding
- No stops: continuous flow, separated by level
- ~10 °C, no wind or precipitation, 365 days/year
- Flat: tighter speeds, few passing manoeuvres
- We aim for theoretical capacity (~2,000/lane)
Four factors, four concrete gains:
On a shared path, throughput collapses: guides advise against it beyond ~300 bikes/h. A tube reserved for bikes has no such friction — the whole width is used for riding.
On the surface, it's the lights and intersections that cap throughput, not the path. With no crossings, the tunnel releases the real capacity of the open segment.
At equal spacing, throughput follows speed: going from 20 to 25 km/h adds ~25%. And with no hills, the gap between cyclists tightens — fewer passing manoeuvres, more fluidity.
As on a motorway, a floor speed evens out the flow. That's what shifts from practical to theoretical capacity — the most underestimated gain.
4. Concrete examples around the world
These figures are not theoretical: they are measured where cycling is massive. The last row is the most telling for our case.
| Source / place | Context | Observed throughput | What it reveals |
|---|---|---|---|
| Highway Capacity Manual global reference |
Bike lane 0.9–1.2 m | 1,700–2,500 bikes/h per lane | The basis of any cycling capacity calculation |
| Copenhagen Danish study |
Two-file path, 2.35 m | ≈ 3,000 bikes/h; 20 bikes/10 s at 21 km/h | A narrow path already absorbs huge peaks |
| Blackfriars Bridge London |
~2 m, single direction, surface | ≈ 2,000 ppl/h — 5× the adjacent car lane | Even capped by lights, the path beats the car |
| Bike lanes measurements in China |
Typical lane, heavy traffic | theoretical ≈ 2,000/h · practical ≈ 1,280/h | The gap = exactly what the tunnel recovers |
| Car lane reference |
3.5 m wide | ≈ 2,000 people/h | The bike moves 1.5 to 2.5× more, at equal width |
Floors, not ceilings.
Every figure on this page is calculated on an ordinary surface path — the most conservative assumption possible. The tunnel has no pedestrians, no lights, no bad weather, welcomes electric bikes and enforces a minimum speed. Its real capacity therefore sits at the top of the global ranges: ~3,000 to 5,000 bikes/h per segment, and on the order of 30,000 to 45,000 people per hour across the whole network — far beyond projected demand.
Sources: Highway Capacity Manual (bike lane capacity); empirical study of Copenhagen's one-way paths (measured widths and throughputs); Blackfriars Bridge cycle path, London; bike lane capacity measurements (theoretical vs practical); car lane / bike lane throughput comparison at equal width. Project assumptions: 3.20 m path (1.6 m/direction), 10 km average trip, 20 km/h average speed, 150 km network. Order-of-magnitude figures, to be refined in detailed engineering.