It is an online high-available database (running on Cassandra) capable of archiving multiple terabytes of AIS-data with an insert rate of > 100.000 messages/minute. Besides the archiving mechanism is features an export tools, simple analytical capabilities based on time, geography and identity. In also supports simple real-time (less than 1 second) queries based on small time intervals, limited geographical areas, or single ships.
Besides AisStore there is a lot of other AIS tools in https://github.com/dma-ais Unfortunately, the department I was consulting for has been closed. So the projects are dormant.
There are a lot of really impressive technology projects out there that focus on crowdsourcing data like this. ADS-B (for aircraft), $this->AIS, ATCS (railroads), radio communications etc.
What's even more amazing is how these project inspire new uses of the data for media, intelligence, politics, investment analysis etc.
Would be interesting to see this shipping data joined w/ other macroeconomic indicators like foreign investment, trade deficit/surplus, and +/- GDP...could be a good way to spot emerging markets
I see a lot of startups popping up in this industry again. Gathering and processing AIS data is a lot of fun.
I was first and, for some time, lead coder for one of them, Weft, until acquisition.
Drop me a note if you are seeking for help in this area, I'd be pleased to get my hands on that again.
It's pretty costly to collect globally in the first place, so I'd be surprised if there was a good API for that.
The good news is that it's relatively cheap to start collecting yourself, get a dAISy[0] and a VHF antenna and you can collect the data directly from any ships in range.
Marinetraffic.com also has a program where they'll send you a receiver for free if you're in a low-coverage area. I got one at home.
I've thought about it - the Puget Sound is ever-so-slightly within line of sight (about half mile to the shore and about 3-4 miles from the ship lane). I figured this was not quite clear enough. May be wrong though.
Go for it. Right now I'm receiving a ship on the river 12.5 miles upstream from my house, through trees and buildings, and my antenna isn't even mounted properly. I'll probably receive from much farther out if I mount it on the roof like I've been meaning to.
If you can see the ships, you'll definitely receive them. 3-4 miles is nothing.
I am not aware of any to any extent exhaustive open databases of MMSI or AIS data.
Sea register MMSI data is fragmented and generally not open enough.
Only commercial data providers can load you up with good historical AIS data and registers, so you can create MMSI table or whatever.
This again depends of what do you want to achieve. Something like 3 months of amateur worldwide AIS feed can give you a pretty good snapshot with MMSIs, names etc.
Sort of off-topic, but I just wanted to point out the environmental impact that these ships have one the planet.
According to Wikipedia, "maritime transport accounts for 3.5 to 4 percent of all climate change emissions, primarily carbon dioxide." [1] Oil spills, ballast water, sewage and other issues that they cause are NOT counted in that 3.5-4% figure.
There are plenty of other jaw-dropping statistics about maritime trade, but I can't find bulletproof sources for them. For example, it is estimated that a big cargo ship can produce as much pollution as 50 millions cars and it is said that the biggest 15 ships in the world produce more pollution than all the cars in the world combined.
Shipping is both the cheapest and most carbon-efficient method of cargo transportation [1].
They do let out an awful lot of Sulphur though. Which causes a lot of premature deaths. However, from next year an annex [2] to the MARPOL Convention is coming into effect which hopefully will lead to a huge reduction.
I wish there was a community or even governmental service of this data with zero limitations. Some countries already provide data for their waters but the rest is a weird mixture of commercial exploitation of community hosted receivers.
I seem to recall an idea to have ships listen for aircraft transponders and relay that information by satellite to improve coverage in the middle of the ocean for tracking aircraft.
You should be able to predict the price of shipping things if you know where all the ships are... or even slow down or speed up vessels to obtain a maximum price at the next port. But if everyone starts doing this it will probably just make everything worse.
Major shipping lanes have strict and well known schedules. Of course not without accidents of various nature.
Most dull problems actually happen in ports (customs etc)
I'm surprised no-one has automated shipping yet, I was pretty surprised how huge the margins were (usually 100% profit) and how big the shipping industry is.
The right startup in this area could be very effective I think.
They actually kind of have... the crew on a big ship do very little “shopping” and a hell of a lot of maintenance. The open sea is a very hostile place and most of the work involved is keeping the ship healthy in that environment.
Agreed. Nowadays the majority of the crew's time is spent either doing maintenance or filling out paperwork to satisfy regulatory requirements - very little involvement with cargo operations (w/ the exception of some specialty liquid cargoes)
I've never heard any one classify maritime shipping as high margin. Having spent over a decade in both ocean and inland waterways operations, my experience has suggested more on the order of 5-10% net margin at best
Margins are high because there are A LOT of risks, starting from superexpensive vehicles and equipment which is never evenly loaded, and ending with bureaucracy, incredibly punishing contracts and weather.
I'd like to see a timelapse of this map / the data behind it, just see the flow of transportation is quite satisfying (like in certain games, Transport Tycoon and Cities Skylines come to mind).
At the default setting, there is a noticeable absence of ships in the Atlantic to the west of the British isles - where is the traffic between Northern Europe and the Americas? I wondered if it was due to really bad weather (no - it appears to be good), but if you zoom in, the traffic appears.
As far as I can tell, this is not a consequence of the filter settings - no exclusions seem to be set.
I can see the Atlantic from my window, great when I found this web site so I can put name / destination to those huge ships on the horizon off the coast of Cornwall, UK.
Reminds me of a customer of ours, https://econavis.io
We built an entire API to fetch marine life data. If you're interested in bundling / including this data in your api, I'd suggest to contact them!
I would guess a popular fishing spot since most of the boats in the area now states "fishing". The location does not update very often, too far from land and recievers most likely. The popup says location recieved by satelite somewhere between 3 hours and 20 minutes ago.
It's probably just logging AIS data ( https://en.wikipedia.org/wiki/Automatic_Identification_Syste... ). Large ships are required to broadcast (like, on the radio) their position, heading, and so on. Anyone with a radio receiver can collect this data for ships within radio range. Anyone with a large number of internet-connected radio receivers can collect this data over a large area.
I've been debating this for years. Partly because it sounds like a fun life, partly because the only way I'm getting affordable housing in London is if it's floating in the Thames.
> I think they do a good job for the free version too.
They did but, imo, no longer do. Features like passage details are now restricted with a link to paid plans (which are absurdly expensive for occasional recreation usage).
- If I click a vessel and then "Past track", a line appears with the voyage but I cannot see details like e.g. speed. There is only a message: "Try the voyage data service...". For pure curiosity the prices are much to high: https://www.marinetraffic.com/en/online-services/plans.
- If I wanted to see where a vessel has come frome, I would need last 30 days minimum as recreational boats often do pause at marinas/anchor.
Prices are much too high for occasional 'curiosity' looking. Maybe I'd pay up to 80/y, but not 140 and certainly not 470. The latter would be needed for voyage data older than 3 days. Nautical map overlay would be great, maybe I'd pay 50/y for this in addition.
Wonder if the service could distinguish between commercial ships and recreational motor/sailboats? Or maybe use a traffic limit to differentiate between professional/hobby usage?
Personal information/pure curiosity. Not really important.
E.g. about a couple of sailing friends, how fast did they sail in comparison to others? Or maybe in which anchorages or marinas someone was? The issue is, that sometime you want to check this out one (or even a couple of) months later.
Potentially also great for passage planing, e.g. density report and/or nautical map layers. (Maybe one-off months could work for this but so far I didn't try).
Satellite data would rarely be needed. For individual tracking 'YB Tracking' seems better suited than AIS.
It is an online high-available database (running on Cassandra) capable of archiving multiple terabytes of AIS-data with an insert rate of > 100.000 messages/minute. Besides the archiving mechanism is features an export tools, simple analytical capabilities based on time, geography and identity. In also supports simple real-time (less than 1 second) queries based on small time intervals, limited geographical areas, or single ships.
Besides AisStore there is a lot of other AIS tools in https://github.com/dma-ais Unfortunately, the department I was consulting for has been closed. So the projects are dormant.
Also happy to answer any questions.