Infrabel

Line sections

/infrabel/line-sections

GEOJSON APPLICATION/JSON

Segmentation of the rail network according to the characteristics of the infrastructure and the possibilities of operation. This list includes the route and technical characteristics of the line sections. Important note:

These data are to be associated with the data set of the network's Operational points. It geolocates the start and end points of the sections whose identifiers and abbreviations are given in this dataset.
These data are intended to allow schematic representations of the Belgian rail network (i.e. for a maximum scale of 1/25 000).

(source: https://opendata.infrabel.be/explore/dataset/lijnsecties/information)

Refresh: Daily

From

2024-08-21 14:52:32

To

2026-10-10 04:20:05

Records

707

import requests
import geopandas as gpd

from datetime import datetime, timedelta, timezone

yesterday_noon = (datetime.now(timezone.utc) - timedelta(days=1)).replace(
    hour=12, minute=0, second=0, microsecond=0
)
params = {
    "timestamp": yesterday_noon.isoformat(),   # ISO accepted; epoch also works
}

url = "https://api.mobilitytwin.brussels/infrabel/line-sections"

data = requests.get(url, headers={
        'Authorization': 'Bearer [MY_API_KEY]'
}, params=params).json()

gdf = gpd.GeoDataFrame.from_features(data["features"], crs=4326)

# Plot the Belgian rail network
gdf.plot(figsize=(12, 10), linewidth=0.5, color="darkblue")

# Compute total network length
gdf['length_km'] = gdf.to_crs(31370).geometry.length / 1000   # metres need a projected CRS (Belgian Lambert 72)
print(f"Total rail network: {gdf['length_km'].sum():.0f} km")

Operational points

/infrabel/operational-points

GEOJSON APPLICATION/JSON

The operational points of the railway network in Belgium

Refresh: Daily

From

2024-08-21 14:52:54

To

2026-10-10 03:15:01

Records

707

import requests
import geopandas as gpd

from datetime import datetime, timedelta, timezone

yesterday_noon = (datetime.now(timezone.utc) - timedelta(days=1)).replace(
    hour=12, minute=0, second=0, microsecond=0
)
params = {
    "timestamp": yesterday_noon.isoformat(),   # ISO accepted; epoch also works
}

url = "https://api.mobilitytwin.brussels/infrabel/operational-points"

data = requests.get(url, headers={
        'Authorization': 'Bearer [MY_API_KEY]'
}, params=params).json()

gdf = gpd.GeoDataFrame.from_features(data["features"])

# Filter only stations (class_fr "Gare")
stations = gdf[gdf['class_fr'] == 'Gare']
print(f"Total stations: {len(stations)}")
for _, s in stations.head(10).iterrows():
    print(f"  {s['commerciallongnamefrench']} ({s['symbolicname']})")

# Plot all operational points
gdf.plot(figsize=(10, 8), markersize=5, color="navy")

Punctuality

/infrabel/punctuality

JSON APPLICATION/JSON

Infrabel's raw punctuality data: every stop of every train on one day, with its planned and actual arrival and departure. A file holds the trains of one day, given by datdep, and is stored at 06:00 UTC the next day: for day D, request any timestamp on D+1 after 06:00 UTC. Some days are missing (Infrabel outages, e.g. 2026-05-26 to 2026-06-23); a request for such a day returns an earlier file or an empty one, so check datdep.

Refresh: Daily

From

2024-08-21 08:00:00

To

2026-10-10 09:05:49

Records

759

import requests
import pandas as pd

from datetime import datetime, time, timedelta, timezone

day = datetime.now(timezone.utc).date() - timedelta(days=2)   # the day you want
# The trains of day D are stored at 06:00 UTC on D+1
params = {
    "timestamp": datetime.combine(day + timedelta(days=1), time(12), timezone.utc).isoformat(),   # ISO accepted; epoch also works
}

url = "https://api.mobilitytwin.brussels/infrabel/punctuality"

data = requests.get(url, headers={
        'Authorization': 'Bearer [MY_API_KEY]'
}, params=params).json()

df = pd.DataFrame(data)
assert df['datdep'].iloc[0] == day.isoformat()   # a missing day returns an earlier file

# Compute delay statistics
df['delay_min'] = df['delay_arr'] / 60   # seconds late at arrival
avg_delay = df['delay_min'].mean()
late_trains = df[df['delay_min'] > 5]
print(f"Average delay: {avg_delay:.1f} min")
print(f"Trains >5min late: {len(late_trains)} / {len(df)}")

# Most delayed stations
worst = df.groupby('ptcar_lg_nm_nl')['delay_min'].mean().nlargest(10)
print("\nMost delayed stations:")
print(worst)

Segments

/infrabel/segments

GEOJSON APPLICATION/JSON

Geolocalised route of all railway tracks on the Infrabel network (not only the main tracks with a line number).

Refresh: Daily

From

2024-08-21 14:52:55

To

2026-10-10 03:45:04

Records

705

import requests
import geopandas as gpd

from datetime import datetime, timedelta, timezone

yesterday_noon = (datetime.now(timezone.utc) - timedelta(days=1)).replace(
    hour=12, minute=0, second=0, microsecond=0
)
params = {
    "timestamp": yesterday_noon.isoformat(),   # ISO accepted; epoch also works
}

url = "https://api.mobilitytwin.brussels/infrabel/segments"

data = requests.get(url, headers={
        'Authorization': 'Bearer [MY_API_KEY]'
}, params=params).json()

gdf = gpd.GeoDataFrame.from_features(data["features"])

# Segment lengths are given in km
gdf['segment'] = gdf['stationfrom_name'] + ' – ' + gdf['stationto_name']

# Show longest segments
longest = gdf.nlargest(5, 'length')[['segment', 'length']]
print("Longest rail segments:")
print(longest)

# Plot the full railway network
gdf.plot(figsize=(12, 10), linewidth=0.5, color="gray")

Endpoints

Line sections

Operational points

Punctuality

Segments