≈ $212M/yr — realistic 2030 scenario, all-in
Range across scenarios: $174M/yr at the low end, $272M/yr at the high end. About 2.4% of the construction cost — a normal, even prudent, ratio for this kind of infrastructure. This total now includes every item long left out: immersive décor (≈ $15M), optimized shuttles ($15M instead of $10M), drainage, geothermal and phosphorescence.
Basis of calculation: amounts in constant dollars, without inflation. This budget covers annual operation and maintenance — it does not include amortization of the construction capital nor loan interest, which depend on the financing model chosen (public loan, P3, subsidy) and are calculated separately.
What this calculation adds. Five items long invisible or underestimated are now explicit. Drainage (≈ $3.8M) and geothermal (≈ $1.0M) were already counted, folded into other lines — isolating them does not change the total. However, maintenance of the immersive décor (≈ $15M, of which ~$11M is new), optimized shuttles (from ~$10M to $15M, i.e. +$5M) and phosphorescence (≈ $0.6M) add real cost: the realistic budget rises from $194M to ≈ $212M/yr. No figure is inflated — we simply make visible what was missing.
1. Staff
The network runs like a genuine public-transport service. The costs include loaded salaries (benefits and payroll charges, i.e. salary + ~30%). We count about 1,700 productive hours per full-time employee per year.
| Category | FTE | Annual cost ($M) |
|---|---|---|
| Bike patrollers (50 at peak, 20 off-peak) | 150 | 11.2 |
| Control centre (10 stations, 24/7) | 52 | 4.3 |
| Shuttle crews + terminals (optimized fleet ≈ 26 trucks + 4 boats) | 90 | 7.0 |
| Maintenance technicians (electromechanical, civil, cleaning) | 120 | 10.2 |
| Bike-fleet mechanics | 60 | 3.8 |
| Station staff and customer service | 60 | 3.5 |
| Management, administration, IT, finance, HR, marketing | 70 | 7.0 |
| Total staff | 602 | ≈ 47 |
How coverage is calculated
Three teams rotate according to ridership. Converting « people simultaneously on duty » into annual FTEs is done through person-hours (a post staffed 24/7 all year requires about 5.15 FTE):
Peak: 50 × 7 h = 350 h/day. Off-peak: 20 × 17 h = 340 h/day. Total 690 h/day × 365 ÷ 1,700.
10 stations × 24 h = 240 h/day × 365 ÷ 1,700. Camera monitoring and coordination, around the clock.
Fleet ≈ 26 trucks + 4 boats, a cycle of one every 1.5 min at peak. Drivers across 3 shifts (24/7), boat crews in summer and ground staff at the terminals.
Québec–Lévis shuttle frequency: departures every 2 minutes at peak at launch, tightened to ~1.5 minutes at maturity (fleet of ≈ 26 trucks), more spaced out off-peak. Monitoring 1,500 cameras from 10 stations assumes solid AI assistance, which surfaces only the relevant alerts to operators.
2. Maintenance, energy, shuttles and rents
The maintenance labour is already counted above; this section covers materials, parts and contracts. The concrete structures are durable and almost free of major maintenance for 30 years — it is the electromechanical systems, and above all the bike fleet, that dominate.
The energy breakdown: 94,000 MWh/yr
The « energy » item breaks down into six uses. At the Hydro-Québec rate assumed (≈ 8.5¢/kWh), total electricity comes to ≈ $8.0M, to which is added ≈ $1.5M of backup and test generators. Pumping and drainage is the 4th item — it appears here explicitly for the first time.
| Energy use | MWh/yr | Cost ($M) |
|---|---|---|
| Ventilation | 44,000 | 3.74 |
| Stations (heating, cooling, equipment) | 20,000 | 1.70 |
| LED lighting + gobos | 12,000 | 1.02 |
| Pumping and drainage | 9,000 | 0.77 |
| Electric-bike charging | 5,000 | 0.43 |
| Control, telecom, cameras | 4,000 | 0.34 |
| Electricity subtotal | 94,000 | 8.00 |
| Generators (backup + tests) | — | 1.50 |
| Total — « Energy » item | — | 9.50 |
The maintenance and operating items
The maintenance labour is already counted in staff; this section covers materials, parts and contracts. Drainage and geothermal, until now folded into « Tunnel + station maintenance », are now isolated on their own lines. Each combines its share of electricity (out of the « Energy » item) and its share of maintenance.
| Item | Annual cost ($M) |
|---|---|
| Energy (electricity + generators, excluding drainage, geothermal and décor) | 7.4 |
| Drainage and pumping (pump electricity 0.77 + maintenance: pumps, sumps, drain clearing 3.0) | 3.8 |
| Station geothermal (heat-pump electricity 0.3 + maintenance 0.7) | 1.0 |
| Immersive décor (the « simulating nature » set: LEDs, gobos, sound, station screens, sacrificial films, vandalism) | 15.0 |
| Tunnel + station maintenance (structure, pathway, ventilation, fire, cleaning — excluding drainage, geothermal and décor) | 25.3 |
| Phosphorescence (evacuation marking: photometric inspection + reapplication provision) | 0.6 |
| Safety maintenance (cameras, SOS posts, drones, AI) + telecom/IT | 18.0 |
| Bike-fleet maintenance (76,000 vehicles) | 45.0 |
| Québec–Lévis shuttles (fuel, maintenance, terminals — excluding salaries, optimized fleet) | 8.0 |
| Rents and leases (premises not owned, depots, offices, terminals) | 4.0 |
| Insurance and miscellaneous (liability, snow clearing of entrances, water-sewer, ticketing) | 18.0 |
Drainage: a load that never sleeps. In places the tunnel passes below the water table: water seeps in continuously and must be pumped 24/7 across 150 km, on top of summer condensation. Unlike ventilation, which eases off at night, the pumps never stop — hence electricity of ≈ $0.77M/yr (9,000 MWh, i.e. 10% of all the network's electricity) and maintenance of ≈ $3M/yr for the dozens of pumping stations (submersible pumps replaced in rotation, drain clearing, controllers), i.e. ≈ $3.8M/yr in total.
Geothermal: a net-positive item. Heating and cooling the 150 entrances with geothermal heat pumps (COP 3.5) costs ≈ $0.3M/yr of electricity and ≈ $0.7M/yr of maintenance — i.e. ≈ $1.0M/yr. But by replacing direct electric heating (10–14 GWh) with only 3–4 GWh, geothermal saves $0.7 to $0.9M/yr: it is the rare item that pays for itself in operation. Details on the Geothermal page.
Shuttles: the optimized version. Until a sub-river tunnel is bored, the Québec–Lévis link relies on a fleet of ≈ 26 trucks and 4 boats (doubled capacity per vehicle: 50 bikes per truck). It absorbs the peak and raises the item from ~$10M to ≈ $15M/yr — ≈ $7M of salaries (90 FTE, counted in staff) and ≈ $8M of operations. Details on the Shuttles page.
Phosphorescence: ≈ $0.6M/yr, and zero electricity. The photoluminescent evacuation marking recharges passively from the light of the LEDs already installed: its electricity consumption is nil. Operation is limited to photometric inspection (~$0.1M) and a reapplication provision amortized over ~12 years (~$0.5M). Details on the Phosphorescence page.
The key item: the bike fleet. With 50,000 electric bikes, 25,000 regular bikes and 1,000 self-service quadricycles, wear is intense. Bike-share systems typically cost $500 to $1,500 per bike per year. The realistic scenario takes ≈ $600/bike ($45M) — more prudent than the $30M in the original dossier (≈ $400/bike, optimistic). It is by far the most uncertain item in the budget.
Maintaining the immersive décor: ≈ $15M/yr
This is the item that was missing. The décor that turns a concrete tube into an underground forest — LED sky, printed surfaces, gobos, sound and station screens — must be maintained: replacing LEDs, recalibrating, cleaning the optics, and above all replacing sacrificial films and repairing vandalism. The dedicated analysis puts this maintenance at 2 to 4% of the $500M CAPEX.
| Scenario | Annual rate | Décor OPEX |
|---|---|---|
| Optimistic | 2% | $10M |
| Realistic (retained) | 3% | $15M |
| Prudent | 4% | $20M |
$15M is already the optimized figure. If the network relied on dynamic displays everywhere (screens and video mapping over 150 km), maintenance would climb to ~$45M/yr. It is because ~70% of the décor is passive — printed surfaces, enamelled porcelain, an LED sky that lasts 30 to 50 years — and the costly electronics are confined to the stations that the item holds at $15M. This design choice already saves ~$30M/yr. Details on the Simulating Nature page.
Of those $15M, about $4M was already counted (the LED sky's energy in the « Energy » line, and part of the LED replacement and cleaning in « Maintenance »). The ~$11M remaining is genuinely new — the décor's own maintenance, the sound, the gobos, the station electronics, the films and vandalism. It is the item that weighs most in moving the budget from $194M to ≈ $212M/yr (the rest coming from shuttles and phosphorescence).
3. The annual budget — realistic 2030
| Budget item | $M / yr | Share |
|---|---|---|
| Staff (602 FTE) | 47.0 | 22% |
| Energy (excluding drainage, geothermal and décor) | 7.4 | 4% |
| Drainage and pumping | 3.8 | 2% |
| Station geothermal | 1.0 | <1% |
| Immersive décor | 15.0 | 7% |
| Tunnel + station maintenance | 25.3 | 12% |
| Phosphorescence | 0.6 | <1% |
| Safety + telecom maintenance | 18.0 | 8% |
| Bike-fleet maintenance | 45.0 | 21% |
| Shuttles (excluding salaries) | 8.0 | 4% |
| Rents and leases | 4.0 | 2% |
| Insurance and miscellaneous | 18.0 | 8% |
| Subtotal | 193.1 | 91% |
| Operational contingency (10%) | 19.3 | 9% |
| TOTAL — realistic 2030 | ≈ 212 | 100% |
The biggest items are the bike fleet ($45M), staff ($47M), tunnel and station maintenance ($25.3M) and maintenance of the immersive décor ($15M). The shuttles weigh $15M in total ($8M here + $7M of salaries in staff). All the network's electricity — ventilation, stations, lighting, drainage and charging combined — weighs only ≈ $9.5M thanks to Hydro-Québec rates, which confirms the energy-independence argument. Drainage ($3.8M), geothermal ($1.0M) and phosphorescence ($0.6M) remain modest but very real items.
4. Minimum and maximum by choices
Staff is almost fixed (the headcount is frozen); what really moves the total is the bike fleet, the ambition of the immersive décor (2 to 4% of its cost) and the decision to buy or lease the premises.
| Lever | Minimum | Realistic 2030 | Maximum |
|---|---|---|---|
| Staff (headcount fixed) | 46 | 47 | 49 |
| Energy (excluding drainage, geothermal and décor) | 6 | 7.4 | 10 |
| Drainage and pumping | 3 | 3.8 | 5 |
| Station geothermal | 0.8 | 1.0 | 1.3 |
| Immersive décor | 10 | 15 | 20 |
| Tunnel + station maintenance | 23 | 25.3 | 28 |
| Phosphorescence | 0.5 | 0.6 | 0.8 |
| Safety + telecom maintenance | 16 | 18 | 20 |
| Bike fleet (the key item) | 30 | 45 | 75 |
| Shuttles (excluding salaries) | 6 | 8 | 10 |
| Rents (all owned ↔ much leased) | 1 | 4 | 8 |
| Insurance and miscellaneous | 16 | 18 | 20 |
| Subtotal | 158.3 | 193.1 | 247.1 |
| Operational contingency (10%) | 15.8 | 19.3 | 24.7 |
| TOTAL | ≈ $174M/yr | ≈ $212M/yr | ≈ $272M/yr |
| Per user (200,000 users) | ≈ $870 | ≈ $1,060 | ≈ $1,360 |
The cost per user is calculated on 200,000 regular users (100,000 intensive + 100,000 occasional) and covers operation only: amortization of the construction capital adds on top, depending on the financing model chosen.
5. Buy or lease — the link with construction
The « rents » line depends directly on a choice made on the construction side: if the sites are bought ($250M in capital), the annual rent is minimal (~$1M, mostly offices and depots). If you lease or reuse existing buildings, you save that capital, but the rent climbs to ~$6–8M/yr.
The capital ↔ operation trade-off. Over about thirty years, $250M of bought land is roughly equivalent to $6–8M of rent per year: near break-even, which mostly changes when the money goes out and who bears the charge. To be decided together with the financing structure — which is why we exclude inflation and interest here.
6. Lowering the net cost: revenue
This budget is gross, before any revenue. The network would generate several sources of income to subtract for the real net cost:
- Self-service rental of electric bikes and scooters at every station
- Network access pricing (monthly or annual subscription)
- Advertising in the tunnels and stations
- Concessions, shops and secure parking at the entrances
Even modest pricing — on the order of $30 to $50 per month per regular user — would cover a significant part, or even all, of the annual operating budget.