Container Logistics(lesswrong.com)
lesswrong.com
Container Logistics
https://www.lesswrong.com/posts/PHmYhE4sKnwzYgvkh/a-brief-introduction-to-container-logistics
19 comments
I’d love to see a business flow diagram of how all this works and which parties are responsible at each step. The deeper I look into the shipping process the more complex and grift-oriented it seems to get.
It is incredibly interesting to me, yeah. I’m not going to lie, I have 25 pages of a book sitting around in my Google Drive, but I quickly realized two things:
1) this will be the most boring book in existence, and requires a much better writer than me to make for an interesting read. And 2) lots of my knowledge is specific to the European situation, which can sometimes be wildly different. The shipping industry can be surprisingly locally focused.
But I might draw out a few diagrams and write a blog post some day :)
But I might draw out a few diagrams and write a blog post some day :)
Actually, I had this write-up lying around which I wrote for somebody involved in building a container terminal simulation game: https://docs.google.com/document/d/e/2PACX-1vQYtBzhkogHUvn2z...
It's pretty specific to terminals, but it's a decent introduction.
It's pretty specific to terminals, but it's a decent introduction.
Thank you for posting this - it's exactly the level of explanation I was looking for. And if you want to write a book with lots of factual information that isn't completely boring, you might want to try doing it in the style of Cadillac Desert which explains California's water management system by following the history of the people and organizations that built it.
I noticed on your section on inland terminals that there's no mention of rail-based terminals, which is interesting because that is the dominant form of inland terminal I think about. Are they rare in Europe (or, equivalently, are they unusually common in the US), or is this just something that you're not familiar with?
Yes, this is definitely my personal experience shining through: my primary experience is with inland terminals which predominantly handle barges. In the Rhine-delta (the Rotterdam/Antwerp hinterland) this is much more common than rail terminals, although trimodal (barge/rail/truck) are on the rise. Most hinterlands however don't have fine mazed rivers and canals suitable for water transport like The Netherlands, Germany, and Belgium have, so in other areas of the world (notably the US) rail-based inland terminals are a lot more common.
But inland terminals are an entirely different beast from sea terminals due to the scaling I mentioned. Area scales quadratically, but quay side or rail length scales linearly. This means larger sea terminals have lots of trouble managing an efficient flow of containers onto and from the ships. Inland terminals, both barge and rail, are often not constrained by this in and out flow, but have more natural constraints like the total amount of containerizable goods transported in the geographic area they serve.
The biggest difference is probably what parts of the transport inland terminals are responsible for. In The Netherlands, Belgium, and Germany I've noticed that often inland terminals act as hinterland operators, also taking responsibility for the barge or train leg, and usually also last mile trucking to the customer. Sea side terminals tend to not do this nearly as much.
So TL;DR: yes, rail is the dominant form basically everywhere except the area where I happen to live and work. But no, both rail and barge inland terminals are very similar in many respects, but are quite different from sea side terminal operations.
EDIT: Actually, just to clarify, I have a lot more experience with inland terminals than sea side terminals. But because the typical container terminal problems (berthing, stacking, equipment assignment etc.) are so much less of an issue with the smaller scale of inland terminals, they only get a very short mention in my document. There really is no berthing problem if you handle 2 barges a week, there is no equipment assignment problem if you have a single crane and a single reachstacker, and there is no real stacking problem if you can ask a trucker to walk around and spot their assigned box between the ~300 boxes you currently have. Typically the biggest issue for inland terminal operations is the transport planning from and to the sea side terminals, and that is similar for both barge and rail terminals.
But inland terminals are an entirely different beast from sea terminals due to the scaling I mentioned. Area scales quadratically, but quay side or rail length scales linearly. This means larger sea terminals have lots of trouble managing an efficient flow of containers onto and from the ships. Inland terminals, both barge and rail, are often not constrained by this in and out flow, but have more natural constraints like the total amount of containerizable goods transported in the geographic area they serve.
The biggest difference is probably what parts of the transport inland terminals are responsible for. In The Netherlands, Belgium, and Germany I've noticed that often inland terminals act as hinterland operators, also taking responsibility for the barge or train leg, and usually also last mile trucking to the customer. Sea side terminals tend to not do this nearly as much.
So TL;DR: yes, rail is the dominant form basically everywhere except the area where I happen to live and work. But no, both rail and barge inland terminals are very similar in many respects, but are quite different from sea side terminal operations.
EDIT: Actually, just to clarify, I have a lot more experience with inland terminals than sea side terminals. But because the typical container terminal problems (berthing, stacking, equipment assignment etc.) are so much less of an issue with the smaller scale of inland terminals, they only get a very short mention in my document. There really is no berthing problem if you handle 2 barges a week, there is no equipment assignment problem if you have a single crane and a single reachstacker, and there is no real stacking problem if you can ask a trucker to walk around and spot their assigned box between the ~300 boxes you currently have. Typically the biggest issue for inland terminal operations is the transport planning from and to the sea side terminals, and that is similar for both barge and rail terminals.
Super interesting. Thank you for this link and your posts above!
> The deeper I look into the shipping process the more complex and grift-oriented it seems to get.
The more industries I start to get deeper into, the more I realize this applies to almost every single industry, no matter size or for how long it's been around. Industries seems get into "extract as much money from as many parties as possible" really quickly.
The more industries I start to get deeper into, the more I realize this applies to almost every single industry, no matter size or for how long it's been around. Industries seems get into "extract as much money from as many parties as possible" really quickly.
Isn't that the definition of capitalism? Part of my duties in IT is to minimize costs with vendors. Sometimes simply asking (or demanding) a reduction is all it takes. Try that with your car insurance... I bet you can get $10 - $20 a month knocked off.
> The deeper I look into the shipping process the more complex and grift-oriented it seems to get.
If you don't like what you see, then you shouldn't look into advertising and ad tech.
If you don't like what you see, then you shouldn't look into advertising and ad tech.
What happens when the fees due on collection get so high that the importer / final delivery company / final customer changes their mind about collecting the container? Who has to store and ultimately dispose of the contents? Do they get to auction it off and keep the proceeds?
I have no clue, never have heard of that happening, but then again, I never bothered to ask. I suspect that customs issues would be the more common use case to not actually want to pick up a container though.
The fees are not usually due on collection, but as part of the contract between shipping line and shipper. Not picking up the container probably does nothing to get you out from under that contract.
The fees are not usually due on collection, but as part of the contract between shipping line and shipper. Not picking up the container probably does nothing to get you out from under that contract.
Containers get auctioned off if not paid for eventually. There was program like Storage Wars but focused on shipping containers.
They may be more efficient for the shipping line but do the 20,000 TEU behemoths (ULVC) exacerbate the problem?
Yes and no. Yes because you can't run a ship carrying some containers on a certain line as often, which means that the wait after missing a ship is likely going to be bigger. But also no because fundamentally the problem of having 10 delayed 2000 TEU ships is not that different from 1 ship of 20,000 TEU. Not to mention that ships only got about 2x bigger over the past 20 years, so the numbers I just pulled out of my ass are quite arbitrary.
More smaller ships would give a bit more flexibility, but I don't think it's an easy fix for the current problems.
More smaller ships would give a bit more flexibility, but I don't think it's an easy fix for the current problems.
Were there not that many empty containers sitting on container chassis?
Yes, probably! This is directly related to the detention issue: many containers stay on the same chassis from pickup at the empty depot until they get delivered at the sea terminal. Equipment to handle containers properly is not cheap (think reachstackers, not regular forklifts). ETAs of ships getting pushed back means longer waiting times till the containers can be delivered and thus they spend longer on chassis. I’m not too sure who bears the financial impact of that, it might be that trucking companies include chassis rent in the transport, or they might similarly charge after X days of usage. Pretty sure the shipping line does never assume liability for this though, so either the shipper or the trucker is getting shafted.
> Similarly, I could stack the containers much more closely together, sacrificing the "random access" property in the process, to create extra space in the port that might increase average throughput. However, this would destroy any guarantee that a given container will ever get out of the port. Humanity as a whole might be better off, but the owner of that particular cargo will be severely affected.
Doesn't averaging already handle this exact problem?
Doesn't averaging already handle this exact problem?
If you book at a shipping line, there is really a tree of companies being engaged to work. There are two empty depots involved for empty pickup and delivery. The container is usually owned by the shipping line, but stored at various third party locations. Then there is the two sea side terminals, who need to do 4 billable operations usually called Terminal Handling Charges or THC for short. These are in order gate-in land, out onto ship, in from ship, gate-out landside. Then there is the temporary storage at the sea side terminals on either side. For you as the shipper you can usually deliver the full container a few days before the cargo closing for outbounds, and you get a few days to pick up the full container once it is discharged off the ship. Failing two meet these time windows will result either in your container getting rejected at the gate, which is expensive because you have to pay for the hinterland transport 3 times instead of once (try #1, return after fail, retry). Alternatively you pay the storage costs, these are often called demurrage charges. In addition to these all there are restrictions on the time you as the shipper can take to stuff or strip (fill/empty) the container, going over this you're going to get charged something called detention charges (a late fee or "renting" free for the container, you get X days included in the transport).
The tricky part comes in when the ETA of the ship shifts. Say you already picked up the container from the empty depot to stuff it, but then the shipping line notifies you the ship is delayed by a week. You now have to store it somewhere, and if you're not careful, the shipping line will try and charge you detention fees. If you deliver it "early" to the terminal (e.g. in time for the old ETA, but early according to the new ETA) you've created a problem for the terminal (high yard utilization), you're going to get charged a demurrage fee, or the terminal will not accept the container and send the trucker on their way again, causing you to have to pay for the transport. Notice how in none of these cases the shipping line is impacted, and sometimes they even profit off of it.
One of the ways to avoid this might be to book more door-to-door transports (or 'carrier haulage') as opposed to arranging the hinterland transports yourself (or 'merchant haulage'). This is often not ideal because it requires shipping lines to have specific knowledge of the hinterlands they serve, but also puts the onus on them and them only to fix this. It also does put even more power into the hands of shipping lines, which is something the sector should probably avoid.
The removal of the 2-high stacking limit only helps to relieve pressure on the 'hinterland' storage equation of it all, it does nothing for the sea side terminals which are already running at capacity.