ETA accuracy benchmark · 🇳🇱 Rotterdam

How Port Flow's predicted ETA compares to each vessel's own broadcast ETA at Rotterdam, measured on closed voyages over the last 30 days. We publish the number — good or bad. Lower error is better.

Port Flow predicted ETA (RMSE)

3.25 h

MAE 3.54 h · 978 voyages

Broadcast ETA (RMSE)

62.50 h

crew-declared, 434 voyages (same set)

Our advantage

94.8 %

lower error than the broadcast ETA

Broadcast ETAs more than 7 days from the actual arrival are excluded as AIS sentinel / placeholder values (252 excluded). Both figures are computed on the same remaining voyage set — apples to apples.

Closed voyages

error = predicted/broadcast minus actual arrival (hours)


244721416	product	10/12, 04:40 AM UTC	-2.31 h	6.95 h
244690044	container	10/12, 04:38 AM UTC	-2.01 h	3389.52 h
244660830	container	10/12, 04:38 AM UTC	-1.73 h	—
205587990	lng	10/12, 04:36 AM UTC	-0.53 h	-343.60 h
244100001	container	10/12, 04:33 AM UTC	3.48 h	-80.60 h
244078754	container	10/12, 04:30 AM UTC	-0.43 h	—
205568390	product	10/12, 04:27 AM UTC	6.09 h	0.54 h
244660522	container	10/12, 04:24 AM UTC	-1.12 h	-81.82 h
244650874	container	10/12, 04:22 AM UTC	-0.23 h	—
244670494	container	10/12, 04:21 AM UTC	-0.65 h	40.59 h
244700987	container	10/12, 04:17 AM UTC	-1.55 h	—
244690528	product	10/12, 04:16 AM UTC	-1.17 h	-12.20 h
275525000	product	10/12, 04:16 AM UTC	-0.19 h	-204.28 h
211517600	product	10/12, 04:15 AM UTC	-1.39 h	-27.77 h
244137310	container	10/12, 04:13 AM UTC	-1.09 h	2.41 h
244218293	container	10/12, 04:11 AM UTC	-1.21 h	—
244071771	container	10/12, 04:09 AM UTC	7.67 h	-82.90 h
244630814	product	10/12, 04:09 AM UTC	-1.65 h	-367.47 h
244650928	container	10/12, 04:08 AM UTC	3.23 h	-78.80 h
244006037	product	10/12, 04:07 AM UTC	-1.83 h	-41.77 h
244650874	container	10/12, 04:06 AM UTC	-1.41 h	—
205393290	container	10/12, 04:06 AM UTC	-1.52 h	7.86 h
244013087	product	10/12, 04:02 AM UTC	-0.86 h	-413.21 h
244995397	container	10/12, 04:01 AM UTC	-1.57 h	-3.13 h
244890894	container	10/12, 03:57 AM UTC	2.53 h	—
253242498	product	10/12, 03:56 AM UTC	-1.74 h	23.05 h
211169910	container	10/12, 03:56 AM UTC	-1.85 h	4522.05 h
244650935	container	10/12, 03:53 AM UTC	10.56 h	—
244710378	container	10/12, 03:53 AM UTC	-1.78 h	—
244620698	container	10/12, 03:50 AM UTC	-0.46 h	—
244650575	container	10/12, 03:41 AM UTC	8.53 h	—
205508590	product	10/12, 03:38 AM UTC	-0.75 h	-203.65 h
244218293	container	10/12, 03:30 AM UTC	0.62 h	-2.76 h
244650709	product	10/12, 03:28 AM UTC	4.53 h	-1.52 h
257562000	product	10/12, 03:28 AM UTC	5.75 h	-126.48 h
244650874	container	10/12, 03:09 AM UTC	1.34 h	-1198.16 h
205581790	product	10/12, 03:03 AM UTC	0.09 h	—
205310590	product	10/12, 03:03 AM UTC	2.99 h	-180.30 h
205578990	product	10/12, 02:59 AM UTC	2.99 h	8.26 h
244650537	product	10/12, 02:57 AM UTC	2.31 h	-10.07 h
244822424	product	10/12, 02:57 AM UTC	-0.15 h	34.55 h
244700714	product	10/12, 02:53 AM UTC	5.28 h	—
244850962	container	10/12, 02:51 AM UTC	4.84 h	—
205505590	product	10/12, 02:49 AM UTC	1.38 h	13.45 h
215345000	product	10/12, 02:48 AM UTC	4.72 h	-204.81 h
269057536	product	10/12, 02:42 AM UTC	-3.25 h	2.56 h
563052100	container	10/12, 02:37 AM UTC	1.54 h	-691.12 h
244620945	container	10/12, 02:35 AM UTC	-0.17 h	—
244710133	container	10/12, 02:25 AM UTC	1.53 h	-35.25 h
244750019	container	10/12, 02:21 AM UTC	-0.17 h	-23.39 h

See it on a live desk

Predicted ETAs, congestion and sanctions screening across 51 ports — try it in one click, no signup.

Open dashboard See pricing →

How it's measured

Predicted ETA: distance-to-berth over speed-made-good, refreshed on every AIS position, with a seasonal correction from rolling medians.
Broadcast ETA: the crew-entered ETA in the vessel's AIS message — the de-facto industry default.
Both are compared to the actual arrival timestamp on the same closed voyages. We report RMSE (penalises large misses) and MAE (typical error).
Broadcast ETAs more than 7 days off are dropped as AIS sentinel / placeholder values, and our model is scored on that same set — so the comparison is honest, not cherry-picked.
Coverage is public AIS only; ports with weak coverage show fewer closed voyages. We'd rather show a small honest sample than a big misleading one.