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)

5.26 h

MAE 3.62 h · 986 voyages

Broadcast ETA (RMSE)

60.38 h

crew-declared, 440 voyages (same set)

Our advantage

91.3 %

lower error than the broadcast ETA

Broadcast ETAs more than 7 days from the actual arrival are excluded as AIS sentinel / placeholder values (276 excluded). Both figures are computed on the same remaining voyage set — apples to apples.

Closed voyages

error = predicted/broadcast minus actual arrival (hours)


244730042	container	11/14, 02:25 PM UTC	-0.18 h	—
244620511	container	11/14, 02:25 PM UTC	0.35 h	-566.43 h
211572680	product	11/14, 02:25 PM UTC	4.72 h	-1009.93 h
244710710	product	11/14, 02:23 PM UTC	1.31 h	-1397.74 h
244780450	container	11/14, 02:22 PM UTC	15.23 h	—
244660419	product	11/14, 02:19 PM UTC	1.38 h	-20.33 h
205569190	container	11/14, 02:19 PM UTC	1.06 h	6987.67 h
244020144	container	11/14, 02:18 PM UTC	1.45 h	—
244810900	container	11/14, 02:17 PM UTC	2.45 h	647.03 h
244087055	product	11/14, 02:17 PM UTC	1.86 h	-106.45 h
244650900	container	11/14, 02:17 PM UTC	1.20 h	-34.06 h
244050207	container	11/14, 02:16 PM UTC	0.77 h	-12.76 h
255806226	product	11/14, 02:13 PM UTC	1.74 h	-171.23 h
244397268	product	11/14, 02:13 PM UTC	1.02 h	-128.21 h
244012151	container	11/14, 02:10 PM UTC	1.41 h	—
244125138	container	11/14, 02:09 PM UTC	5.75 h	—
244660532	container	11/14, 02:09 PM UTC	1.49 h	-32.61 h
244690697	container	11/14, 02:08 PM UTC	0.84 h	-376.01 h
244005969	product	11/14, 02:07 PM UTC	4.62 h	-22.74 h
244780024	container	11/14, 02:06 PM UTC	5.47 h	—
244078754	container	11/14, 02:06 PM UTC	4.28 h	—
636026417	container	11/14, 02:04 PM UTC	3.20 h	-154.58 h
244630697	container	11/14, 02:03 PM UTC	0.18 h	-20.68 h
244078754	container	11/14, 01:55 PM UTC	6.43 h	-600.06 h
244690223	product	11/14, 01:54 PM UTC	1.34 h	-108.26 h
244012151	container	11/14, 01:53 PM UTC	3.10 h	5224.10 h
244190323	product	11/14, 01:53 PM UTC	1.63 h	-908.48 h
244730485	product	11/14, 01:53 PM UTC	1.05 h	-23.80 h
205549290	lng	11/14, 01:51 PM UTC	5.65 h	-670.55 h
244710549	product	11/14, 01:51 PM UTC	0.91 h	-170.73 h
244660111	product	11/14, 01:51 PM UTC	2.06 h	-251.95 h
244750786	container	11/14, 01:40 PM UTC	3.14 h	-193.51 h
244690168	container	11/14, 01:40 PM UTC	0.59 h	-32.71 h
244010927	container	11/14, 01:40 PM UTC	0.26 h	-11.39 h
244125138	container	11/14, 01:39 PM UTC	5.34 h	—
244700498	product	11/14, 01:38 PM UTC	5.78 h	-130.52 h
244020518	container	11/14, 01:37 PM UTC	0.68 h	—
244710971	container	11/14, 01:37 PM UTC	0.60 h	6050.84 h
244670114	container	11/14, 01:37 PM UTC	6.07 h	-260.92 h
244740814	container	11/14, 01:36 PM UTC	2.23 h	3504.39 h
244010910	container	11/14, 01:30 PM UTC	2.23 h	—
244620511	container	11/14, 01:29 PM UTC	0.59 h	-565.49 h
244670113	container	11/14, 01:29 PM UTC	2.09 h	3826.51 h
244078754	container	11/14, 01:23 PM UTC	5.47 h	—
205595190	container	11/14, 01:21 PM UTC	3.79 h	-199.36 h
244630027	product	11/14, 01:19 PM UTC	1.05 h	—
244620953	product	11/14, 01:17 PM UTC	1.37 h	-979.30 h
244003138	product	11/14, 01:13 PM UTC	1.00 h	-56.23 h
244700039	container	11/14, 01:12 PM UTC	3.78 h	2570.54 h
211689390	container	11/14, 01:11 PM UTC	5.89 h	3774.81 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.