Port Congestion Alerts & Vessel Arrival Notifications

Three people who use this

The chartering desk watching one number

Demurrage starts biting somewhere, and that somewhere is a number you already know. Set it — say 20 hours at Singapore — and you hear from us when the median passes it, not when some index decides the port got "worse". A port sitting steadily above your line keeps telling you, once a day, until it drops back.

The operator waiting on one ship

You know the vessel and you know the port. Add both, pick the stages you care about, and the rest of the voyage is silence. Most people switch on Moored and Departed and ignore everything in between.

The planner choosing between two ports

Follow both. Two independent streams on the same measurement, so when one moves and the other does not, you see it in your inbox rather than by going looking.

Port alerts — when conditions change at a port you follow

Each port is checked continuously against the same AIS-derived figures published on its port page, plus a marine weather feed. 7 conditions can fire: 6 fixed, and one you set yourself.

Congestion Surge

Triggered when congestion rises to High or Severe.

Wind Advisory

Sustained wind >20kn. Critical above 30kn — berthing delays likely.

Swell Warning

Wave height >2m. Critical above 3m — cargo operations may be affected.

Low Visibility

Visibility <2km. Critical below 1km — vessel movements may be restricted.

Berth Occupancy

Berth usage crosses 85% capacity.

Wait Time Spike

24h median >50% above the 7-day median, measured on the figure shown on the port page. Needs at least 5 completed stays in the day, so a thin sample cannot trigger it.

Wait Time Above Your Level

You pick 1–72 hours. Measured on the median shown on the port page, not the average, and sent at most once a day while the port stays above it.

Wind, swell and visibility come from the marine forecast at the port's own position, and say the same sentence as the advisory banner on its port page — fog closing a pilot station is a different problem from a swell stopping cargo work, so they are three alerts and not one.

The fixed conditions fire on a change. The one you set fires on the level — the difference between "Singapore got worse this week" and "Singapore is above 20 hours, which is where my cargo starts costing money". A port that sits high for a month never trips a change rule, because steady is not a change.

Vessel alerts — when a ship you are watching moves

A watch covers one port call. You add a vessel and the port you expect it at, and each stage of that call can alert you:

• Arriving — the vessel approaches the port you are watching
• Anchored — it drops anchor in a monitored anchorage
• Moored — it is alongside at a berth
• Departed — it leaves, which ends the port call
• Delayed — predicted delay passes a threshold you set, from 1 to 48 hours
• AIS lost — the signal goes stale and the vessel can no longer be tracked live

Each switch is per vessel, so you can watch one ship for its arrival only and another for every stage.

Why you will get fewer emails than you expect

AIS coverage is patchy and ships hover around the thresholds that define a status, so the raw signal flips constantly. We would rather be late than wrong:

• A status change has to hold for 30 minutes before it becomes an email — 90 minutes for tugs, pilot boats and fishing vessels, which flip more.
• The same arrival can be noticed by three separate code paths. Only the first one sends; the rest are suppressed for six hours. One event, one email.
• A threshold you set repeats at most once a day while the port stays above it, not every time we look.

The cost is up to one extra check of latency on a real change. That is deliberate: a wrong alert is worse than a late one, and an alerting path you stop trusting is worse than none at all.

What the alerts are measured on

The same figures as the port pages, with the same definitions. The port alert uses the median time at anchorage, not the average — at Singapore 12% of stays run past the 72-hour tracking cap, which drags the average 51% above the median. An alert that fires on a number the page never shows is an alert you cannot check.

Anchorage time covers ships that anchored. Ships that go straight to a berth are not in it, and the share that anchor at all differs enormously by port — 39% at Hamburg against 12% at Busan — so each port page publishes its own denominator next to the figure.

One caveat before you set a number at Los Angeles: container ships there are held offshore under a vessel queuing system and barely anchor at all (a 1.6-hour median, which is a pilot stop), while tankers anchor for about 20 hours. The port figure covers both.

Common questions

How do I get an alert when vessel waiting times exceed a certain threshold?

Follow the port, switch on "Wait Time Above Your Level" and pick a number between 1 and 72 hours. It is measured on that port's published median time at anchorage.

How do I monitor port congestion to anticipate delays?

Two ways, and they answer different questions. Port alerts tell you the conditions changed; a vessel watch tells you your ship did. Most people running cargo use both — the port alert for the week ahead, the vessel watch for today.

Can I get notified when a ship arrives at a port?

Yes. Add the vessel and the port you expect it at, and switch on Arriving. The watch covers that one port call and ends when the vessel departs.

What does it cost?

Nothing. A free account can follow all 7 ports and watch up to 10 vessels at a time. No trial and no card.

Turn them on

Create a free account, or read how PortInsight measures congestion first — the alerts are only as good as the measurement behind them.