Network passengers this hour
Baku Metro Hackathon MVP
Ağıllı Sərnişin
Çox rahatlıq, az izdiham.
Scenario controls
Change date and hour to simulate metro congestion patterns
High-risk stations
—
Most loaded station
—
—
Selected time
—
Forecast by station
Hourly view for one selected station
No forecast loaded yet.
Data source: —
AI decision layer
MetroFlow analyzes all problem types
The system analyzes real metro data and selects the most relevant problem for a specific station, time and situation.
Problem pool
The system considers passenger flow, entrance density, escalators, platform distribution, train intervals, boarding/alighting and Digital Twin scenarios.
Dynamic selection
MetroFlow is not limited to one problem. It identifies the most critical task based on current data.
Decision and impact
For the selected problem, the system shows suitable actions and expected impact in the operator dashboard.
Metro network map
Color-coded station risk on a schematic metro view
Low
Medium
High
Top overloaded stations
Most critical stations for selected hour
No data loaded yet.
What this system shows
Operational meaning of the dashboard
Passenger count shows the predicted number of passengers at a station for the selected hour.
Congestion score is a normalized 0–100 crowding indicator.
Top overloaded stations highlight the highest-risk stations for the selected time slot.
How this helps solve the problem
From analytics to real metro actions
Operators can identify critical stations before crowding becomes severe and take preventive actions.
Examples: passenger redirection, staff allocation, entrance flow management, escalator mode optimization and display-based guidance.
The same data can also be shown on station displays or public-facing interfaces.
Operator recommendations
Action-oriented hints for metro operations team
No recommendations loaded yet.
Passenger-facing use case
How the same prediction can be communicated outside the control room
No passenger guidance generated yet.