Factorio Forums

TL;DR

Introduce a form of compressing items on trains (not transportable by other means) as a way to keep train transportation relevant in the late game

Why?

In Space Age, with the introduction of the Turbo Belts and belt stacking, belt throughput is increased through the roof. So much so that it easily rivals train transportation over long distances, displacing trains as the undisputed long distance, high throughput transportation option of choice, leaving trains mostly irrelevant.

A way to compress items on trains would help bring the balance between trains and belts back into alignment, by vastly increasing item throughput on trains.

The idea:

Instead of loading trains with loose items like we do currently, introduce palletization, a way to package multiple items into a single one, and then load those onto the trains. These palletized items (pallets from now on) would not be transportable via bots or belts.

By manipulating how many items it takes to make each pallet, and the stack size of the pallets, the number of items a regular cargo wagon can transport could be easily controlled.

There could either be just a single pallet research, so the change only has to be made once, or several tiers of pallet research, to help keep the trains in lockstep with the belt stacking.

Each new tier of pallet would be increasing the amount of items that can be transported through pallets in each cargo wagon slot, achieved by the new pallets packing more regular items and / or having a higher stack size.

At the loading stations, the items would be packed onto pallets before sent to the loading buffer. Then those pallets would be loaded into trains, transported to the unloading station, and unloaded into the unloading buffer. From there, the pallets would need to be unpacked before the items in it could be used in crafting, or further transported by bots or belts.

This would necessitate a different train loading and unloading strategy than what is used currently. While not strictly necessary, and regular inserters, chests, and assemblers could be used to handle the pallet logistics, thematically, I think it would be best to have a small set of dedicated entities to manage the loading / unloading process, for example a "pallet crane" (essentially a supersized inserter) to move the pallets around, and a depot (like a special chest that only accepts pallets) where those pallets could be stored.


In case there would be multiple tiers of pallet research, there are some QoL features that would be really nice to have:

1) for the unpacking this would be almost essential, because one would need to drain the system of old pallets before changing over to the new ones, otherwise the old pallets would clog the system.

Fortunately this can easily be solved - the unpacking plant simply needs to work as a recycler that only accepts pallets, and returns 100% of the inputs.

2) for packing, this would not be anywhere near as essential, but having to change the packing recipe for every single loading station for every new tier of pallets would be really annoying and tedious. Therefore, a way to have the packing plants always make the highest tier of pallets would be nice.

3) for train schedules, in order to keep generic interrupts relevant, it would be really good to have the different tiers of palletized cargo detected as a single signal - be that either the raw item, or, preferrably, a general "pallet of [item]" signal (eg "pallet of iron ore" instead of "small / medium / large pallet of iron ore", which could then be used to send the train to the appropriate unloading station. Without this, the unloading train stations would have to be renamed after adopting each new tier of pallets.


Of course, having only a single tier of pallets would neatly sidestep the need for these QoL features altogether.

This gets bonus points if a pallet was using the weight calculation of the rocket as well, giving the opportunity to work out balanced pallets of varieties of material, just for fun.

The crane could take the graphics off the agtower

The selector combinator could be used to ascertain the current contents of the pallet and even more useful if it could work on a per wagon and per wagon slot calculation, such as wagon A slot 20. In fact, giving the pallet that encoding on loading, with the crane able to obey the ordering of items, would go a long way to fixing the inability of reading mixed wagon loads across a long train

+1

there is already a mod if you want to play around with this idea:
https://mods.factorio.com/mod/Intermoda ... rom=search

I like this idea conceptually - to speed up train throughput one could load items into a specialized "train dock" building, placed next to the cars.

In my head the best approach would be that this "dock" building would be something like a 2x2 (so three can fit on one side of a train) and would accept items into a storage buffer; when a train shows up it could then load the "pallets" onto a train, consuming items from its buffer to do so.

Ideally then at the other end the same building rotated the other way would remove pallets from a train and fill its buffer with items as it does so.

One could also implement quality on these buildings to give trains some degree of quality scaling without messing with the trains themselves.

I do see a few issues;

1) pallets would be "unique" items, i.e. they would only ever be useful for trains and if they somehow ended up outside of a train, you'd have useless items stuck somewhere in logistics or in your base

1a) you could have separate "packing" and "loading" buildings, but that would still imply that inserters can move pallets and then the meta for long-distance item transfer might become train wagon inserter chains

2) AFAIK all items so far can go on belts, and if the idea is to make this a buff to trains specifically (which would be nice) then there would have to be some other type of unique behaviour again for these items.

the issues I have are something that, while I could 100% overlook for a mod, would be a big change to base game / space age factorio feeling-wise as they'd effectively be a brand new logistics feature.

That's not to say I don't think it could work; it just might need some work to feel like it "belongs" with the rest of the game to me

edit / addendum:

I think having a "pallet size" research to go along with the belt stacking research makes sense, though ideally they would somehow always just show up as "[item] Pallet" and maybe have a number attached to them which is the number of items in the pallet. The different numbered pallets would not stack with eachother, but could coexist in a wagon when you research the next level of tech. Essentially like partial stacks on belts.

in other words, I see the ideal version of this basically being belt stacking for trains, and working similar to the already implemented belt stacking system.

Maybe instead of palletization, just have a car / tank crane to load / unload vehicles with inventory to / from trains? I mean in real world trains are used to carry cars. Currently cars / tanks can be transported with belts but are really clumsy at it, so it provides a meaningful way to load stuff onto trains without being overpowered to replace trains.

h.q.droid wrote: Thu Oct 16, 2025 3:03 am Maybe instead of palletization, just have a car / tank crane to load / unload vehicles with inventory to / from trains? I mean in real world trains are used to carry cars. Currently cars / tanks can be transported with belts but are really clumsy at it, so it provides a meaningful way to load stuff onto trains without being overpowered to replace trains.

Loading / unloading actual vehicles to transport other cargo just feels wrong to me tbh. I could accept loading / unloading whole chests, but I don't think that's the way to go.

The reason I suggested item compression is because that way existing trains can be upgraded without needing to change the trains, which was a pain point in suggestions about quality cargo wagons having expanded cargo space. Loading vehicles or chests onto trains would thematically require a different sort of cargo wagon.

And if you want to bring IRL examples into the mix, when cars are transported on trains IRL, that's because they want the cars themselves moved from one place to another, and not because they want to use the trunks of those cars to transport iron ore, or something similar

juliawtapp wrote: Wed Oct 15, 2025 8:20 pm I like this idea conceptually - to speed up train throughput one could load items into a specialized "train dock" building, placed next to the cars.

In my head the best approach would be that this "dock" building would be something like a 2x2 (so three can fit on one side of a train) and would accept items into a storage buffer; when a train shows up it could then load the "pallets" onto a train, consuming items from its buffer to do so.

Ideally then at the other end the same building rotated the other way would remove pallets from a train and fill its buffer with items as it does so.

One could also implement quality on these buildings to give trains some degree of quality scaling without messing with the trains themselves.

This is pretty much how I imagined this, although with a few different buildings working in concert.

juliawtapp wrote: Wed Oct 15, 2025 8:20 pm I do see a few issues;

1) pallets would be "unique" items, i.e. they would only ever be useful for trains and if they somehow ended up outside of a train, you'd have useless items stuck somewhere in logistics or in your base

The way I imagine the pallets outside of their intended buildings is similar to the parts of the crashed ship: a pallet "building" that has the constituent items in it's inventory, and then they could be deconstructed the regular way with bots, or by hand.

These pallet "buildings" could be spawned item shower style whenever something happens that would have them appear outside their intended buildings. They could automatically be marked for deconstruction as well.

juliawtapp wrote: Wed Oct 15, 2025 8:20 pm 1a) you could have separate "packing" and "loading" buildings, but that would still imply that inserters can move pallets and then the meta for long-distance item transfer might become train wagon inserter chains

I don't think this is a big issue. Technically, the current meta for belt throughput is cars on belts, but noone really uses them for the simple reason that their use is way too inconvenient. The same would apply to cargo wagon + pallet crane chains.

juliawtapp wrote: Wed Oct 15, 2025 8:20 pm 2) AFAIK all items so far can go on belts, and if the idea is to make this a buff to trains specifically (which would be nice) then there would have to be some other type of unique behaviour again for these items.

That is true. Either the devs will need to implement items that can't be put on belts, or they will need to restrict inserters from grabbing pallets when their drop location is something that is not supposed to handle pallets.

The latter I could imagine implemented even in just a regular mod.

juliawtapp wrote: Wed Oct 15, 2025 8:20 pm the issues I have are something that, while I could 100% overlook for a mod, would be a big change to base game / space age factorio feeling-wise as they'd effectively be a brand new logistics feature.

That's not to say I don't think it could work; it just might need some work to feel like it "belongs" with the rest of the game to me

I don't think it would be too dissimilar to belt stacking. Sure, you'd need considerably more supporting infrastructure than belt stacking, but I don't think it would be new infrastructure in a bad way.

juliawtapp wrote: Wed Oct 15, 2025 8:20 pm I think having a "pallet size" research to go along with the belt stacking research makes sense, though ideally they would somehow always just show up as "[item] Pallet" and maybe have a number attached to them which is the number of items in the pallet. The different numbered pallets would not stack with eachother, but could coexist in a wagon when you research the next level of tech. Essentially like partial stacks on belts.

in other words, I see the ideal version of this basically being belt stacking for trains, and working similar to the already implemented belt stacking system.

I'm thinking the same in this regard.

An idea I had is that "pallet size" could be implemented as a separate value, like how the spoilage timer is kept track of using a separate value unique to each item. Each pallet would have a value attached to it that would represent how many items are contained in it.

That way, the pallet upgrade research would only change how high that value can go. And also, each size of pallet would not be it's own unique item. Even more so if the pallet itself was just a generic item, and would also keep track of which type of item it contains, alongside the number of them.


The very lazy alternative to all this is to just have research that increases the number of stacks a cargo wagon can carry.

The downside if that would be that such a research would break vanilla LTN train designs unless there was a way to automatically read cargo wagon size.

Pallets (or other item compression methods) would not break LTN designs because the number of stacks would stay the same, therefore keeping the loading / unloading buffer size and cargo wagon size the same.

I feel like enabling quality for storage size of wagons may be more effective and fun. (Is there a mod for that?) And maybe a tier 2 wagon. (Some planet mod has one, but it's only 25% better)
The advantage of trains, to me, are that they carry goods dynamically. You do not need to hardwire your factory if you use them.

Edit: Long Handed Gleba-Chonk Inserters also double the unloading speed. Also available in some planet mod)

As a Logistics comparison, remember that Barrels exist in Vanilla, and they SUCK! They existed before Fluid Wagons to enable Train movement of Fluids, and were simply never removed... they are still necessary to move Fluids through Space (eg, Fluoroketone). I do not see how item-based Pallets could improve on this experience. Beltboxes are really not that helpful since the Stack Loader introduced built-in "belt heights".

Gladaed wrote: Sat Oct 18, 2025 2:36 pmI feel like enabling quality for storage size of wagons may be more effective and fun. (Is there a mod for that?) And maybe a tier 2 wagon. (Some planet mod has one, but it's only 25% better)

Several! as well as Bob's Mods.

eugenekay wrote: Sat Oct 18, 2025 5:14 pm As a Logistics comparison, remember that Barrels exist in Vanilla, and they SUCK! They existed before Fluid Wagons to enable Train movement of Fluids, and were simply never removed... they are still necessary to move Fluids through Space (eg, Fluoroketone). I do not see how item-based Pallets could improve on this experience.

I do not understand how barrels are relevant? Could you elaborate on that? My suggestion was never intended to influence how fluids would be moved around, any such influence on fluid logistics would only be an ancillary benefit of the particular, hypothetic implementation.

Zijkhal wrote: Sat Oct 18, 2025 5:50 pm

eugenekay wrote: Sat Oct 18, 2025 5:14 pm As a Logistics comparison, remember that Barrels exist in Vanilla, and they SUCK! They existed before Fluid Wagons to enable Train movement of Fluids, and were simply never removed... they are still necessary to move Fluids through Space (eg, Fluoroketone). I do not see how item-based Pallets could improve on this experience.

I do not understand how barrels are relevant? Could you elaborate on that? My suggestion was never intended to influence how fluids would be moved around, any such influence on fluid logistics would only be an ancillary benefit of the particular, hypothetic implementation.

It is only intended as a point of comparison - I am assuming that Pallets would use a similar set of mechanics.

The “handling” of the Barrel item adds extra Logistics challenges:

• The empty barrel must be crafted before any movement can occur
• Constant supply of barrels is required or the chain is broken
• Empties must be handled at each un-loading point by returning them to the origin or destroying them.

Before the introduction of reliable Voiding with Recycler mechanics this would often cause an inventory-full state if too many barrels were crafted, requiring manual intervention to resolve.

eugenekay wrote: Sat Oct 18, 2025 6:31 pm It is only intended as a point of comparison - I am assuming that Pallets would use a similar set of mechanics.

The “handling” of the Barrel item adds extra Logistics challenges:

• The empty barrel must be crafted before any movement can occur
• Constant supply of barrels is required or the chain is broken
• Empties must be handled at each un-loading point by returning them to the origin or destroying them.

Before the introduction of reliable Voiding with Recycler mechanics this would often cause an inventory-full state if too many barrels were crafted, requiring manual intervention to resolve.

Those are all implementation details, not an inherent part of my suggestion.

Heck, one could relatively easily make a barreling mod that would not require any empty barrel logistics.

My suggestion was really about using item compression to store more of each item in each slot in the cargo wagon, in order to solve the train throughput balance issues with belts. "Pallets" is really just a term used to help people who read it intuitively understand what I'm getting at, and to help sell the idea that item compression would fit the game thematically.

eugenekay wrote: Sat Oct 18, 2025 5:14 pm As a Logistics comparison, remember that Barrels exist in Vanilla, and they SUCK! They existed before Fluid Wagons to enable Train movement of Fluids, and were simply never removed... they are still necessary to move Fluids through Space (eg, Fluoroketone). I do not see how item-based Pallets could improve on this experience. Beltboxes are really not that helpful since the Stack Loader introduced built-in "belt heights".

Gladaed wrote: Sat Oct 18, 2025 2:36 pmI feel like enabling quality for storage size of wagons may be more effective and fun. (Is there a mod for that?) And maybe a tier 2 wagon. (Some planet mod has one, but it's only 25% better)

Several! as well as Bob's Mods.

Yeah, but the one I tried emptied all my locomotives. They don't work quite right.

Zijkhal wrote: Sat Oct 18, 2025 8:52 pmMy suggestion was really about using item compression to store more of each item in each slot in the cargo wagon, in order to solve the train throughput balance issues with belts. "Pallets" is really just a term used to help people who read it intuitively understand what I'm getting at, and to help sell the idea that item compression would fit the game thematically.

If there is no “Pallet item” involved then it is not really palletization per se; just increasing the effective Slot/Stack size by converting to/from an intermediate item. You can already sort-of do this by shipping products, such as Gears instead of Plates. I am not against the idea of adding a Pallet Item; but I think it is unlikely to add much value or interesting gameplay compared to existing solutions. It might give a real reason to farm Wood, if worth the hassle of handling the empty pallets.

There is a mod which lets Cargo wagons benefit from belt stack size bonuses, without having a processing step at either end.

Gladaed wrote: Sat Oct 18, 2025 9:43 pmYeah, but the one I tried emptied all my locomotives. They don't work quite right.

Sounds like it replaced all the existing entities in a Migration? A new game is usually a good idea when changing mod settings drastically.

eugenekay wrote: Sat Oct 18, 2025 10:31 pm If there is no “Pallet item” involved then it is not really palletization per se;

Those are implementation details that we are specifically told in the how to use section of this forum to keep out of our suggestions. Worrying about implementation details of a non-existent feature is not really what this forum is about.

eugenekay wrote: Sat Oct 18, 2025 10:31 pm There is a mod which lets Cargo wagons benefit from belt stack size bonuses, without having a processing step at either end.

That is a neat solution. My only gripe with it is that increasing cargo wagon slot count breaks LTN setups where train stations set train limits based on the available supply / available empty space to unload them.

I would think that simply making locomotives and cargo wagons be boosted by quality might be enough.