Show HN: A Modern Palletization App

by vld5 • 2 years ago

153 42 2 years ago

When searching on the internet for these type of apps, I didn't find many that were open-source AND easy to use. A lot of them had complicated interfaces, although they had loads of features as well.

So what I had in mind when making Stack Solver was an app with a modern interface that has the most essential features. It is also well integrated with Microsoft Excel and renders a 3D customizable drawing.

Stack Solver is programmed in C# using the WPF framework to ensure it is fast and light. The interface is built using WPF UI, a library that allows it to keep up with modern trends (specifically the Fluent design).

It is a work in progress with tons of new features planned and it is my first "serious" project so I would appreciate any feedback :)

aiiotnoodle• 2 years ago

Stuff like this is great.

Fast moving consumer goods is a vastly underrated represented technology sector, they do so much cool stuff that just goes completely unnoticed.

I started my first job in a warehouse and it really gave me a good look at the proprietary technology powering day to day life. It should be more open but few work on stuff like this.

dehugger• 2 years ago

I work in this field, as a developer for a proprietary warehouse management system.

The entire industry runs off closed source systems with little to no accessibility for external developers. The companies developing these systems like it this way, as it means they get a large number of billable hours through their service departments, since there are very few developers available with the expertise to work on these systems.

KomoD• 2 years ago

Not something I have a use for, but that's pretty cool.

Animats• 2 years ago

Robot palletizers use such algorithms, but do humans need a program for this?

Another case of "Machines should think. People should work."

chime• 2 years ago

Absolutely. Pallet arrangements are usually a required spec from the customer to the vendor. The arrangement order matters considerably for shipping and needs to be agreed upon beforehand.

Closi3• 2 years ago

Looks great! If you did do something related to truck fill, would be great to see if it could be generalised to 3D bin packing.

I’m currently working a free / open source warehouse management system so great to see other people working on supply chain problems!

tomtom13372• 2 years ago

This sounds interesting! Have you published it yet?

Closi3• 2 years ago

Not yet but will shoot you a message when I do!

rabotnik2• 2 years ago

Would love to hear more about a FOSS WMS - which aspects are you focusing on initially?

Closi3• 2 years ago

Amazing! I’m trying to make a WMS that’s very easy to customise (rather than configure) - will come with a highly opinionated set of base processes but then will be built in a way that it’s incredibly easy to edit them (rather than building a million parameters/toggles).

Handset processes are installed like apps.

avgDev• 2 years ago

I would maybe like to contribute to WMS. Wrote a WMS for a company I work for.

I use .net, C#, SQL, blazor.

bobim4• 2 years ago

Ah! That's a pretty much orphan subject yet the world is constantly moving pallets.

For pallet stability you would like to have a convex perimeter so the stretch wrapping maintains the boxes effectively. And also criss-crossing boxes for shear resistance.

But you need to align as much boxes corners as possible to get vertical stiffness.

A solver proposing stacking patterns with these constraints would be outstanding.

dist-epoch• 2 years ago

Does center of mass matter? Or not so much.

Should lighter boxes be higher, so they don't get crushed? Or again, doesn't matter.

bobim4• 2 years ago

Yes right, I'm biased toward single product pallet. Then high mass items should be placed on the first lower pallet, low mass on the pallet put on top of the first one in the truck. The solver starts to be complex.

bobim4• 2 years ago

And if it can spit out an OpenRadioss input file, one could launch pallets (almost) directly.

6510• 2 years ago

But you need to align as much boxes corners as possible to get vertical stiffness.

The magic might not be obvious?

I've made an animation[0] using the demo image[1] from VladM7's Stack-Solver.

[0] https://img.go-here.nl/palet-stack.gif

[1] https://github.com/VladM7/Stack-Solver/blob/master/img/scree...

pallet is 120 long. 21+21+21+21+35=119 pallet is 80 width. 21+21+35=77

And also criss-crossing boxes for shear resistance.

Mirror the next layer.

Besides useful it also looks impressive to have a full looking pallet with criss-crossing.

FredPret3• 2 years ago

The humble pallet is the red blood cell of civilization.

There's a ton of really cool optimization ideas like this in Industrial Engineering - factory production schedule optimizations, path optimizations, lots of knapsack problems. What a treat that there are people willing to pay for this kind of work.

bobim4• 2 years ago

Nice analogy!

a3n2• 2 years ago

I agree, very nice.

I'm a truck driver. My analogy, zooming out a bit, is that the entire road system, at least the National Network, is a system of interconnecting and overlapping conveyor belts.

The trailers are trays on the belt.

https://en.m.wikipedia.org/wiki/National_Network

rbaudibert• 2 years ago

I've had some friends in the logistics industry ask about a mobile app for this. How hard would it be to convert that to mobile? Are you depending on some complex libraries for the solver, or did you implement the algorithm yourself? I have 0 to no knowledge of C#.

vld53• 2 years ago

I implemented the algorithm myself and although I don't have much experience with mobile apps programming I think that converting it to other programming languages or adapting it for other frameworks would be fairly easy. The real problem is the 3D rendering which is done using WPF 3D (Windows only), so I would need to completely rework it.

jtriangle• 2 years ago

Even if mobile v1 was simple level by level text output, it'd probably still be useful.

neonsunset• 2 years ago

Moving to Avalonia from WPF should be relatively straightforward (except, possibly, the WPF 3D part), and Avalonia is capable of targeting mobile platforms (even though it really is desktop-first, when compared to Uno or MAUI). As a bonus it will run on macOS and Linux too.

crote• 2 years ago

If you're going to rework it, perhaps consider making it a web app? Wrap it in Electron for desktop use, I bet there's something similar for iOS / Android.

That'd make it available as both a dedicated app for power users, and an (ad-supported) variant for occasional use.

sirjaz• 2 years ago

Amazing job, love seeing native apps like this rather than webapps. Keep it up!

maw• 2 years ago

I tried to make some sort of joke or pun based off of shipping containerized software, I couldn't make it work. Oh well.

datboi2hawt• 2 years ago

Thanks for this! We ditched our Tops Pallet software a year ago and have been looking for something similar. Where can we donate to the development?

jayfiro• 2 years ago

We considered using Tops Pallet software at one point. What made you ditch it?

vld53• 2 years ago

I just set up a donation page, you can find it on the github repo page, on the right. But please keep in mind that for now I have to finish my exams so you may have to wait a bit for new features. Also, any feedback and suggestions for improvement are appreciated especially since you work in the field.

MisterBastahrd• 2 years ago

I suppose this would be useful for large supply chains. Once you get away from that and move to shipping from warehouse to store, though, this would be a largely be a waste of time. Stacking a pallet with various different sized boxes and shapes is a bit of an art form that is largely learned through trial and error until you "get" it (I was on a full time night stock crew while I was in college). It would literally take more time for workers to refer to the layout than it would for them to simply stack the goods onto the pallet.

rabotnik2• 2 years ago

Pallet stacking can also be done automatically (with robots), and there are solutions (only privately developed/owned to my knowledge) that calculate optimal stacking for heterogeneous loads - interlock boxes for stability, weight (distribution & total), height, etc. Each box is characterized on induct (weighed, scanned, sometimes manually updated with metadata on crushability/fragility). Then an algo figures out best way to put the boxes on a pallet. I've seen this for grocery, but applicable in other use cases.

ssl-3• 2 years ago

It's only time-consuming for workers to follow the planned layout when those workers are human.

thy77• 2 years ago

Just lasted 2 days moving pallets barrels and cylinders. I needed the customers to accept non-full 300 lb barrels instead of the full 500 lb. Couldn't do it. Pallets and wrapping were to load a bit easier but often we cut the shrink wrap at the first stop and the pallet had stuff for like 3 stops. Seems like the shrink wrap was practical to keep stuff from flying around but also regulatory, cops could write a ticket.

SoftTalker• 2 years ago

And yet people are allowed to load and drive their own rented U-Haul vans.

reverseblade2• 2 years ago

here's my alternative 3d bin packer https://3dpack.ing

dewfaced• 2 years ago

Only thing I think it needs is the option to change the UOM from metric to imperial.

Show HN: A Modern Palletization App

Related Stories