TolyqPassenger analytics
Dispatcher
Demo build · synthetic data
DDaulet · Dispatcher
Pilot route · 12 stops
Nurly Zhol station
Astana-1 station
Service dayThu 18 Feb 2027
Trip view

Trip T-0412 · Bus 1 · Thu 18 Feb 2027 · departs 08:10

Nurly Zhol station to Astana-1 station · 12 stops · arrived 09:04 · counts from front and rear door cameras
Peak occupancy
78/ 9087%
Baiterek to House of Ministries
Boardings
143passengers
11 of 12 stops with boardings
Alightings
143passengers
Balanced trip · 0 on board at the terminus
Counting latency
47 s95th percentile
Stop to dashboard · target 60 s
Occupancy along the route
On board per stop segment · boardings and alightings per stop
On boardCapacity 90Above 81BoardingsAlightings
Device health
Bus 1 · edge unit E-01
All systems OK
  • Front cameraOK
  • Rear cameraOK
  • GPSOK
  • LTEOK
Last message
14 s ago
Uptime
96.8%
Frames older than 24 h
0
Recent trips
Bus 1 · click a trip to load it
    Edge fleet · pilotBus 1 onlineBus 2 onlineServer v0.9.3 · edge 1.4.0
    Route load

    Mean occupancy by stop and hour · 25 Jan to 19 Feb 2027 · weekdays

    Mean passengers on board when leaving each stop, bus 1, inbound direction
    Route load heatmap
    Rows are stops in route order, columns are departure hours
    Mean occupancyPeak cellfewer than 3 trips
    Busiest segment
    79/ 9088%
    Baiterek to House of Ministries, 08:00
    mean on board
    Trips in sample20 weekdays · bus 1 · 06:00 to 22:59
    118
    Daily peaksMorning · 08:00
    79Baiterek
    Evening · 18:00
    74Khan Shatyr
    Legend
    04081 passengers
    Above 81 (90% of capacity) Above 86 (95% of capacity) fewer than 3 trips, not reliable
    No weekday mean crosses 81 in these four weeks. The 08:00 Baiterek cell is the closest at 79, two passengers under the threshold.
    Fare comparison

    Boardings against validations · week of 1 to 5 Mar 2027

    Camera boardings compared with fare validations per stop, both pilot buses, weekdays
    Boardings and validations per stop
    Week totals · lower panel: unpaid share per stop
    BoardingsValidationsShare ± 95% CIRoute 7.8%
    Estimated unpaid share
    7.8%
    95% CI 5.1% to 10.5%
    Boardings countedDoor cameras · both pilot buses
    16,420
    ValidationsDifference 1,281 boardings
    15,139
    Validation source
    Observer sheets and beep counter
    No link to card numbers
    Where a controller would find the most unpaid trips
    Ranked stop and hour windows, weekday average
    #StopHourEst. unpaid per day
    Unpaid per day, route
    256
    In top 4 windows
    120
    Share of daily unpaid
    47%
    Windows are ranked by estimated unpaid boardings, not by share, so a controller visit catches the most trips per hour on duty. Interval shown on the left is the 95% CI per stop.
    Estimate per stop and hour. No person is identified. No video leaves the bus.
    Alerts

    Safety alerts · week of 29 Mar to 2 Apr 2027

    Overcrowding and person-on-floor alerts from both pilot buses, delivered to the dispatcher console
    False alerts
    0.7per bus-dayOn target
    20 false alerts over 28 bus-days · target at most 1
    Alerts delivered within 30 s
    31 of 31
    Slowest delivery 22 s · median 11 s
    Fall detection precision
    80%staged test
    36 of 45 staged alerts were real falls
    Driver mute
    30 minlogged
    Each mute is stored with time and bus
    Recent alerts
    Mark an open alert as false if the dispatcher sees no problem on the bus
    ConfirmedOpenFalse alert3 open
    TimeBusTypeStop segmentDeliveryStatus
    Showing 9 of 31 alerts this week · all delivered within 30 sView all alerts
    False alerts per bus-day
    Live since 20 Mar 2027
    target at most 1
    Pilot log, both buses: 0.7 on average
    False alerts this week
    Updates when an alert is marked as false
    Planning

    Planning · what-if headway

    Weekday demand on the pilot route and the load each departure would carry at a chosen headway
    Demand profile · weekdays · eleven weeks of counts
    Passengers per hour at the peak load point, counts from 12 Jan to 2 Apr 2027 (eleven weeks)
    Passengers per hour
    What-if headway
    Morning peak · 08:00 · 464 passengers per hour
    Near threshold
    Headway
    10min
    6789101112131415
    At 10 min · morning peak
    77 passengers per departure · 86% of capacity
    Comparison · at 8 min
    At 8 min: 62 per departure · 69% of capacity
    Departures / hour
    6.0
    Buses needed
    7
    Load at 18:00
    72 / 90
    Per departure = demand per hour × headway / 60 · 64 min round trip
    Expected peak occupancy per departure by hour
    At a 10 min headway
    Below 90% of capacityAt or above 90%90% threshold (81)