Could autonomous shuttles carry every bus ride inside Yishun?
A simulation study in two parts, on LTA's fare-card counts of the 163,745 bus rides that start and end inside the town on a school-term weekday, and its live bus data.
Singapore is starting driverless shuttles inside new towns. This study asks what it would take in a mature one. It sends a simulated weekday of Yishun's internal bus rides, drawn from LTA fare-card counts between every pair of stops, through an on-demand shuttle fleet and compares every ride with today's buses. Road times are calibrated to LTA's live bus data and checked against bus GPS movement.
Four-seat vehicles would need 700, twenty-seat ones 375. The dispatcher matters as much as the vehicles: one that considers four times as many shuttles per request needs only 375. Across 18 alternative assumptions the ten-seat fleet ranges 375–550. Several assumptions favour the shuttles: unlimited kerb space, no traffic added by the fleet itself, no charging, travel times known exactly, and road speeds measured outside the morning rush. Two favour the bus: crowding is ignored, and the dispatcher is a simple one.
Every internal ride of a simulated weekday goes through a fleet of pooled, on-demand shuttles and is compared, ride by ride, with today's buses. Press play to watch the day, or scroll for what it showed.
Keys: W A S D fly the camera (Q E turn, R F up and down, Shift for speed, H returns to Yishun), space plays or pauses, arrow keys step 5 minutes, G rides along behind one vehicle (Esc to leave), M switches between the Streetmap and Cartoon styles; clicking a vehicle rides along with it. The flying trucks are decoration and play no part in the study. Shuttle movements are the simulator's own output for one demand replicate. Bus movements are modelled from LTA's published headways and calibrated running times (LTA does not publish timetables or vehicle traces). Building heights for HDB blocks come from HDB's storey counts; other buildings are context massing from OpenStreetMap.
Car ownership in Singapore is capped by the vehicle quota, so the autonomous future being tested here is fleets serving towns. Punggol's autonomous shuttles opened to the public on 1 April 2026 to "enhance connectivity to key amenities and transport nodes for residents within Punggol" (LTA, 4 March 2026). In September the Transport Minister said Singapore had "only 16 passenger vehicles, all in Punggol" and about 20,000 riders so far. Next comes a trial of point-to-point service "within the whole estate, without the need for fixed routes", then one to two more HDB towns, Tampines North and Tengah (MOT, 18 September 2026). In October LTA stopped a planned pilot of autonomous buses on two public bus services, to look at narrower uses first (Mothership, 13 October 2026, confirmed by LTA).
Yishun was not among the towns named. Three in five of its HDB blocks were completed in the 1980s (HDB property data), it has about 221,600 residents (Census 2020), and about 163,745 bus rides a day start and end inside the town. In Punggol, Grab is piloting on-demand rides "between existing pick-up and drop-off points", taking the most direct path (LTA, 29 July 2026). This study asks what that model would take at Yishun's scale. Pickups and drop-offs stay at today's bus stops, so walking is the same in both modes.
It does not forecast AV adoption, fares, costs or safety, and it is not an LTA plan. Demand is held at today's observed bus rides. It compares two ways of serving the same rides on the same roads, and measures what the comparison would cost in vehicles, kilometres and kerb space. The closest Singapore study, Mo et al. (2021), simulated shared autonomous vehicles competing with buses for first-mile trips in Tampines, with operators choosing fleets and fares. This one holds demand fixed and asks what replacing every internal bus ride would take.
An earlier plan for this project was to simulate 20–80% self-driving private cars in Yishun's traffic, with roadside radio units and new traffic-signal plans. That does not fit Singapore. Car numbers are capped, traffic signals already adapt to traffic, and simulated self-driving cars mostly reproduce the following distances they are given. Open junction counts are sparse one-day surveys, while fare-card data counts the bus rides inside the town.
163,745
bus rides on an average school-term weekday start and end inside Yishun.
—
in the — hour alone, the morning rush. The evening peak at 18:00 is —.
1.06 km
Half of all rides are shorter than this, as the crow flies: a few stops.
66%
start or end at an MRT station or the bus interchange. These are feeder trips, where waiting is a large share of the journey.
LTA publishes, for each month, how many fare-card rides went from every bus stop to every other, hour by hour. Keeping only rides that start and end inside the Yishun planning area leaves 163,745 rides on an average school-term weekday (July and August 2026), across 4,861 stop pairs. September, with a week of school holidays, has 8% fewer rides at 07:00.
Half of them cover less than 1.06 km as the crow flies, and 66% start or end at an MRT station (Yishun, Khatib or Springleaf) or the bus interchange. These are feeder trips: a few stops between an HDB block and the train. For trips this short, waiting is a large share of the journey. That is where an on-demand vehicle could help, and also where it has the least room for error.
LTA counts fare-card rides by boarding and alighting stop. Riders tap in and out, and LTA does not document how the few rides without a tap-out are handled. A trip that changes buses inside Yishun counts as two rides. Riders may live anywhere; what is counted is where they board and alight.
10.4 min
by today's bus, stop to stop (waiting plus riding), in the 08:00 hour.
450
ten-seat shuttles keep up with the bus in every hour of all three simulated weekdays: 6.9 min at 08:00. Read the curve right to left as the fleet shrinks.
12.2 min
with 425 shuttles, worse than the bus. The 07:00 rush leaves a queue that peaks around 08:00. The edge is sharp.
39.1 min
with 375 shuttles, and 1.1% of the day's riders are refused (no pickup within 45 minutes).
700 or 375
four-seat or twenty-seat vehicles instead. At the threshold a shuttle carries only 1.8 riders on average, so bigger vehicles help less than their seats suggest.
Each simulated day sends every observed ride through a fleet of pooled, on-demand shuttles that stop only where riders board or alight. Today's buses are modelled from LTA's routes and published headways. Running times, signal delay and dwell are calibrated on 171,838 matched pairs of LTA's live arrival estimates and checked against buses' reported GPS positions. Waiting times use headway regularity estimated from live predictions for GPS-monitored buses.
Above a threshold the shuttles are faster than the bus: riders wait less on average and ride direct. The margin is wide off-peak and narrower in the rush. Below it, the 07:00 rush leaves a queue that peaks around 08:00. The edge is sharp. With ten-seat shuttles, 450 vehicles keep up in every hour of all three simulated weekdays: 6.9 min at 08:00 against the bus's 10.4 min. With 425, the 08:00 hour averages 12.2 min, worse than the bus, before recovering by 09:00. With 375, it averages 39.1 min and 1.1% of the day's riders are refused.
Each point is the mean over three simulated weekdays (demand replicates); the band spans them. Both modes are measured on the same riders: the 0.4% of rides with no direct bus in LTA's route data are left out. Open markers show fleets that also refused riders (no pickup within 45 minutes), with a whisker to the mean if each refused rider counts as 45 minutes plus a direct ride. At the threshold a shuttle carries 1.8 riders on average, so bigger vehicles help less than their seats suggest: twenty seats instead of ten cut the fleet only to 375. A stricter rule, matching the bus's 90th-percentile time in every hour, needs 450 ten-seat shuttles.
450
ten-seat shuttles, running all day. One dot is one vehicle.
152
buses of all services are inside Yishun at 07:00. The shuttle fleet is 3.0 times as many vehicles, though each bus carries about ten times as many people.
16
autonomous passenger vehicles in all of Singapore in September 2026 (MOT).
4.7 times
the driving: 159,924 shuttle-km inside Yishun per weekday, against 33,871 km by all buses.
1.8
riders on board per shuttle-km, against at least 8 per bus-km. 20% of shuttle-km are driven empty.
9,450
vehicle-hours a day for a fixed fleet sized for the morning rush (thin amber line), while buses inside Yishun follow the timetable (teal).
200
shuttles would do around midday if the fleet could shrink after the rush: at least 5,250 vehicle-hours a day.
| 10-seat shuttles (simulated) | Buses today (modelled) | |
|---|---|---|
| Vehicles in service at 07:00 | 450 | 152 |
| Vehicle-hours in service per day | 9,450 | — |
| … if resized every hour (a lower bound) | 5,250 | — |
| Hours driving in Yishun per day | 6,632 | 1,463 |
| Vehicle-km in Yishun per day | 159,924 | 33,871 |
| Share of km driven empty | 20% | not known |
| Riders on board on average (rider-km per vehicle-km) | 1.8 | at least 8 |
| Mean wait, whole day | 1.2 min | 3.3 min |
| Mean wait, 07:00–09:00 | 2.1 min | 2.8 min |
| Mean time on board | 5.5 min | 8.5 min |
| 90th-percentile wait, 07:00–08:00 | 6.2 min | 6.1 min |
450 shuttles is 3.0 times the 152 buses of all services that are inside Yishun at 07:00, and 5.6 times the 81 feeder buses on the road then. Each bus carries about ten times as many people, so today's buses offer more places (about 16,200 seated and standing, against 4,500 shuttle seats), and they also carry riders to and from other towns. In September 2026 Singapore's entire autonomous passenger fleet was 16 vehicles.
The shuttle fleet drives 159,924 km inside Yishun per weekday, 4.7 times the 33,871 km that all buses drive on Yishun's roads (15,102 km by feeders). Counting only the rides inside Yishun, buses carry at least 8 riders per kilometre they drive there. The shuttles carry 1.8, and 20% of their kilometres are driven empty, repositioning or heading to the next pickup.
Feeder-service theory points the same way. In Li and Quadrifoglio's analysis (2010) of a single feeder zone, demand-responsive service beats fixed routes only below about 10–50 riders per square mile per hour. Averaged over the town, Yishun's internal demand at 07:00 is about 2,000 riders per square mile per hour.
If the fleet in service could shrink after the rush, with vehicles parked or charging, far fewer would be needed off-peak: about 200 around midday, a few dozen after midnight. That would be at least 5,250 vehicle-hours in service a day instead of 9,450. The fixed fleet spends about 6,632 of its hours driving, against about 1,463 hours of bus driving inside Yishun.
Fleets by hour come from separate one-hour simulations with a 45-minute warm-up and no backlog carried over from earlier hours, so they are lower bounds. Each is the smallest fleet whose mean stop-to-stop time in that hour is no worse than the bus, with no refused riders (larger of two demand replicates, 25-vehicle grid). Bus counts and bus-km come from published headways and modelled running times. 98.5% of shuttle-km fall inside the planning-area boundary, as bus-km do.
27
shuttles boarding or alighting at once at the Yishun MRT and interchange stops at the busiest moment. Kerb space is unlimited in the base case; the model counts what the fleet uses.
18
at Khatib. Two-thirds of rides touch a station, so a pooled fleet converges on the same few kerbs.
—
at 07:00 with at most — shuttles per station at once: still faster than the bus (—).
—
with a limit of —. Adding vehicles does not help: — shuttles take —, because the extra vehicles queue too.
Two-thirds of rides touch a station, so a pooled fleet converges on the stops around each MRT station. In the base case kerb space is unlimited, and the model counts what the fleet uses. With the fleet that matches the bus all day, up to 27 shuttles board or alight at once at the Yishun MRT and interchange stops, and 18 at Khatib, before counting idle shuttles waiting for work.
The model does not know how many vehicles each stop can hold. The kerb test asks how much it would need. The design question moves from "how many vehicles" to where they stop, and how riders walk to them.
In the kerb test, a shuttle that finds every berth at its station taken queues until one frees. The dispatcher does not anticipate these queues, so pickups there can slip beyond the usual five-minute limit.
92%
of rides reach their stop sooner by shuttle, with 450 ten-seat shuttles. The average saving is 5.2 min.
—
of rides over 1.5 km are faster: bus routes loop and stop more often than a direct shuttle.
—
of rides under 0.5 km are faster, the lowest share. Where the shuttle loses, it is mostly on the wait.
Gains by the subzone where rides start. The census share of residents aged 65+ is context only: the fare-card data has no rider ages.
| Subzone | Rides a day (with a bus option) | Residents 65+ | Faster by shuttle | Mean gain |
|---|---|---|---|---|
| — | ||||
Comparing each simulated ride with the bus it replaces shows where the gains fall. With 450 ten-seat shuttles, pooled over 3 simulated weekdays, 92% of rides reach their stop sooner than by bus, comparing each ride with one simulated bus wait (94% beat the bus's average time for that trip). The average saving is 5.2 minutes. Longer rides gain more: 78% of all rides under 0.5 km are faster, against 97% of all rides over 1.5 km, where bus routes loop and stop more often than a direct shuttle. After 20:00, when buses run less often, 98% are faster; in the 07:00–09:00 peak, 78%. The slowest tenth of shuttle rides take more than 10.5 min, against 19.7 min by bus (12.9 against 17.3 min in the 07:00–09:00 peak). Of the rides that are slower by shuttle, 18% are under 0.5 km; in 94% the shuttle wait is longer than the bus wait, and in 18% the shuttle ride is longer than the bus ride.
Same fleet as above. "Faster" compares each simulated shuttle ride with one simulated bus wait (random arrival, measured headway regularity) plus the bus's in-vehicle time for the same stops and minute.
Part 1 replaced every bus ride. Part 2 asks whether a combined service, keeping every bus and adding shuttles only where the timetable is slowest, saves riders' time at a cost worth paying.
Each option carries the same 163,069 internal rides of an average weekday, on the same simulated days, timed stop to stop: waiting plus riding, as in Part 1.
Every option with acceptable service costs more to run than today's buses and saves passenger time. With a safety operator on board, as today's pilots require, the four C designs save 1,609–10,111 passenger-hours a weekday for SGD 56,000–370,000 more a day: SGD 33–36 for each hour of riders' time saved. Full replacement costs SGD 43 an hour with every bus kept, and SGD 36 if the services that carry only internal riders were withdrawn (SGD 37 counting the bus operator's fee only), about level with C. Remote supervision with nobody on board, not permitted today, would bring C to SGD 20–21 an hour, and full replacement with those services withdrawn to SGD 17.
Is an hour of riders' time worth that much? A separate social estimate values it at SGD 19.7 in the vehicle and 1.5 times as much spent waiting. Time saved is not revenue: nobody is paid it.
There is no single acceptable cost, so the study shows what each daily budget buys. Each point is one simulated design; the line joins the designs that save more than any cheaper one. With someone on board, combined designs fill the line at small and medium budgets, and full replacement buys the most time at the highest price per hour. With remote supervision only, full replacement takes over at a smaller budget, and more cheaply.
| Budget per weekday | Best design | Passenger-hours saved | Per hour saved |
|---|---|---|---|
| — | |||
The time saved goes to riders whose bus trip is slow next to a direct drive: the feeders that loop through the estates, and the long waits early in the morning and late at night. Under the combined designs, riders who stay on the bus keep today's trip, and riders declined a shuttle take the bus as today.
Passenger-hours saved per weekday against today's buses in the same conditions, mean of the same three simulated days in every row. Full replacement (B) accepts a quoted pickup up to 45 minutes away and refuses beyond that, with no bus; the combined designs decline beyond 15 minutes and the rider takes the bus. The column "B, C's rule" is full replacement under the combined designs' rule, which separates that choice from replacement itself. Kerb limits of 10 and 7 shuttles at once per station are test values: the real capacity is unknown. Speed stresses scale road times; added shuttle traffic slows nothing, because congestion is not modelled.
| Condition |
|---|
| — |
observed rides by stop pair and hour (LTA fare-card OD), routes and headways, and all Singapore bus rides, which share Yishun's bus-km between internal and other riders (internal riders account for 53.4% of in-area bus passenger-km). simulated shuttle waits, rides, declines, refusals, vehicle-km and vehicle-hours; bus times modelled. assumed cost values without published evidence (insurance, software, staffing ratios, permits, control centre), the charging-reserve estimate (not a charging simulation), fixed demand, no congestion feedback.
Full method, sources and limits: the Phase 2 report, its independent review, the source register and the evidence page.
Every figure on this page comes from the project's own pipeline, and each was checked. The evidence page has the full scorecard, the charts behind it, the data files and the commands that rebuild them.