Punctuality
TEC: bus operator in wallonia, with lines into brussels
- Format
- Parquet
- Updates
- Daily (one file per Brussels-day, anchored to 00:00 Europe/Brussels)
- Latest snapshot
- Snapshots stored
- about 687
curl -H "Authorization: Bearer $MOBILITYTWIN_TOKEN" -o punctuality.parquet "https://api.mobilitytwin.brussels/tec/punctuality"Daily punctuality table derived from TEC's GTFS-RT trip-update stream over a full Brussels-day window.
Each row represents one (trip, stop) observation with the latest known arrival/departure times, delays, and schedule_relationship — plus a synthetic row per cancelled trip when the feed signals cancellation without stop-time entries.
Single Parquet file (zstd-compressed). Read with Polars, DuckDB or PyArrow. The data is consistent with the GTFS schedule: scheduled = actual_time − delay matches the static GTFS arrival_time for TEC feeds that publish absolute times.
A single Parquet file. Read it with Polars, DuckDB or PyArrow.
Which days have data
Data from 21 Aug 2024 to 1 Nov 2026, on 798 days. 30 of the last 30 days have data. A typical day holds 1 snapshot. Days are in UTC.
- Full day
- Partial, 50–90% of a typical day
- Low, under 50%
- No snapshot
- Not expected
Click a day above, or pick a date, to get its requests.
Parameters
| Name | Meaning |
|---|---|
timestamp | Return the last snapshot at or before this moment, in Unix seconds or ISO 8601 (e.g. 2026-05-06T12:00:00Z). Without it, the latest snapshot is returned. |
Python example
# pip install polars requests
import io
import requests
import polars as pl
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()}
url = "https://api.mobilitytwin.brussels/tec/punctuality"
data = requests.get(url, headers={
"Authorization": "Bearer [MY_API_KEY]"
}, params=params).content
df = pl.read_parquet(io.BytesIO(data))
print(f"rows: {df.height:,}")
# Note: TEC's GTFS-RT feed publishes delays only — no absolute arrival_time /
# departure_time. Join to static GTFS on (trip_id, stop_id) to recover times.
not_cancelled = df.filter(~pl.col("cancelled") & pl.col("arrival_delay").is_not_null())
# Distribution of arrival delays.
print("\nArrival delay distribution (seconds):")
print(not_cancelled.select([
pl.col("arrival_delay").quantile(0.05).alias("p05"),
pl.col("arrival_delay").median().alias("median"),
pl.col("arrival_delay").mean().round(1).alias("mean"),
pl.col("arrival_delay").quantile(0.95).alias("p95"),
pl.col("arrival_delay").max().alias("max"),
(pl.col("arrival_delay") <= 60).mean().round(3).alias("share_within_60s"),
]))
# Where do the worst delays accumulate?
print("\nMean arrival delay per route (top 10):")
print(not_cancelled.group_by("route_id")
.agg([pl.col("arrival_delay").mean().round(1).alias("mean_delay_s"),
pl.len().alias("n")])
.filter(pl.col("n") >= 50)
.sort("mean_delay_s", descending=True)
.head(10))
# Cancellations.
cancelled = df.filter(pl.col("cancelled"))
print(f"\ncancelled rows: {cancelled.height:,}, unique trip-instances: "
f"{cancelled.select(['trip_id', 'start_date', 'start_time']).unique().height}")