ETA accuracy benchmark · 🇳🇱 Amsterdam

How Port Flow's predicted ETA compares to each vessel's own broadcast ETA at Amsterdam, 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)

1.34 h

MAE 13.24 h · 987 voyages

Broadcast ETA (RMSE)

59.91 h

crew-declared, 282 voyages (same set)

Our advantage

97.8 %

lower error than the broadcast ETA

Broadcast ETAs more than 7 days from the actual arrival are excluded as AIS sentinel / placeholder values (184 excluded). Both figures are computed on the same remaining voyage set — apples to apples.

Closed voyages

error = predicted/broadcast minus actual arrival (hours)


244690653	container	10/12, 04:39 AM UTC	-0.22 h	4814.35 h
244750687	container	10/12, 04:37 AM UTC	-0.13 h	—
244690137	container	10/12, 04:36 AM UTC	2.63 h	—
244373397	container	10/12, 04:35 AM UTC	0.11 h	—
244750687	container	10/12, 04:25 AM UTC	3.24 h	—
218027120	product	10/12, 04:25 AM UTC	0.10 h	-37.87 h
269057594	container	10/12, 04:23 AM UTC	-0.14 h	—
244660196	container	10/12, 04:22 AM UTC	-1042.66 h	-1045.14 h
244770588	container	10/12, 04:21 AM UTC	2.07 h	6811.65 h
244690137	container	10/12, 04:14 AM UTC	0.23 h	—
244001448	product	10/12, 04:10 AM UTC	0.60 h	3.13 h
205566890	container	10/12, 04:07 AM UTC	0.10 h	—
244690137	container	10/12, 04:06 AM UTC	2.41 h	—
211395020	product	10/12, 04:05 AM UTC	0.95 h	-148.10 h
211540960	product	10/12, 04:03 AM UTC	1.29 h	—
244660117	container	10/12, 04:03 AM UTC	0.04 h	—
244102434	container	10/12, 03:58 AM UTC	-0.33 h	-54.47 h
205571490	container	10/12, 03:54 AM UTC	0.51 h	—
244690352	container	10/12, 03:53 AM UTC	-0.23 h	—
563319100	product	10/12, 03:50 AM UTC	1.17 h	-141.84 h
244770588	container	10/12, 03:40 AM UTC	0.15 h	6812.33 h
244690358	container	10/12, 03:38 AM UTC	0.12 h	—
636018752	product	10/12, 03:38 AM UTC	1.00 h	-1.63 h
205585590	container	10/12, 03:28 AM UTC	3.36 h	-85.59 h
244130285	container	10/12, 03:22 AM UTC	3.22 h	-55.38 h
244670266	product	10/12, 03:07 AM UTC	1.51 h	-11.13 h
244020558	container	10/12, 02:59 AM UTC	-0.09 h	-78.35 h
244140757	container	10/12, 02:54 AM UTC	-0.01 h	-67.57 h
244371956	container	10/12, 02:47 AM UTC	0.12 h	2.38 h
244690352	container	10/12, 02:45 AM UTC	0.15 h	-6.63 h
211476600	container	10/12, 02:31 AM UTC	0.51 h	3201.73 h
244650971	container	10/12, 02:23 AM UTC	-0.03 h	—
636022979	product	10/12, 02:21 AM UTC	0.01 h	-405.35 h
211417530	container	10/12, 02:14 AM UTC	-1.10 h	-661.36 h
244371956	container	10/12, 02:12 AM UTC	0.02 h	2.96 h
244710165	container	10/12, 02:05 AM UTC	-475.85 h	-465.59 h
305002300	container	10/12, 01:55 AM UTC	0.64 h	-134.93 h
211500430	product	10/12, 01:44 AM UTC	0.11 h	-29.75 h
205592790	product	10/12, 01:31 AM UTC	0.71 h	3.26 h
244030466	container	10/12, 01:29 AM UTC	0.34 h	-7.36 h
211494230	product	10/12, 01:13 AM UTC	1.59 h	—
244790153	container	10/12, 01:08 AM UTC	0.65 h	—
244790153	container	10/12, 12:42 AM UTC	-1.11 h	-398.66 h
205571490	container	10/12, 12:42 AM UTC	1.13 h	—
244670612	product	10/12, 12:41 AM UTC	1.05 h	-50.68 h
244730412	container	10/12, 12:39 AM UTC	2.25 h	—
244371956	container	10/12, 12:38 AM UTC	2.25 h	—
244730412	container	10/12, 12:33 AM UTC	3.24 h	—
244660933	container	10/12, 12:32 AM UTC	1.35 h	—
244730412	container	10/12, 12:31 AM UTC	1.31 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.