Current Assets in Factorio Controlling: When Your Own Accounting Meets the Reality of Assembly Lines

When optimizing factories from a business management perspective, it’s important not to overlook current assets: Materials on assembly lines and in buffers represent tied-up capital that affects return on investment.To accurately track these dynamic inventories, I’ve updated my Factory Ledger. The virtual accounts (Transit and Work in Progress) were refined, as static snapshots provide inaccurate data in the face of constant fluctuations. 

baseline was break even
this was my first optimisation
net profit margin now 17.26% 

In the previous post, I had already dealt extensively with various valuation perspectives for our factories. We accounted for fixed assets, analyzed environmental pollution, assessed land use, and compared various cost and efficiency aspects. Yet, despite all these metrics, one crucial factor was still missing—one that determines success or failure in industrial practice: current assets.

In Factorio in particular, current assets can reach considerable proportions depending on the chosen factory structure. This is especially evident in classic main-bus factories, but the principle applies to every layout. There is constantly a massive amount of material on the conveyor belts: iron plates, copper plates, and advanced intermediate products, some in different qualities or stages of processing. These inventories aren’t simply “there”; they tie up capital. From an accounting perspective, they are current assets that are deeply embedded in the factory but have not yet been realized as finished, saleable output. To reflect this economic reality, I have now firmly incorporated this point into my calculation sheet.

The introduction of this expanded evaluation logic had an immediate and noticeable impact on my ongoing optimization scenarios, which I have gradually refined. This clearly demonstrates that, at a certain point, a typical Pareto effect sets in.  (barchr below)

The numbers on the x-axis indicate which scenario we have there. Number 1 is the baseline, which means we have a scenario where the return on investment is 0, in other words, we earn nothing. It’s simply neutral. Excel sheet
Number 2 is my first optimization, and from there on, the returns increase successively, and we earn more and more money. And the last scenario is Scenario Number 8, where, in principle, all optimizations have been implemented under the given conditions.

At the beginning, the improvements were massive. Through relatively clear, structural changes to the production lines, we were able to significantly increase throughput and make the factory much more efficient. In the meantime, maximum throughput has essentially been reached. Further optimizations therefore no longer stem primarily from even more output, but from more cost-effective, leaner operations. The central question now is: How can the same throughput be achieved with less tied-up capital, fewer fixed assets, lower working capital, and an overall leaner structure?

So now we're looking at Scenario 8. As you can see,
I've managed to cut costs a bit more.
net profit margin is now 29,36%
Revenue can't be increased any further and
it's very clear that, in my opinion,
no further optimizations are possible.

This new perspective also forced me to thoroughly re-examine my existing accounting logic in the Factory Ledger. This particularly affected the virtual accounts, especially the posting of transit and work in progress (WIP). As is often the case when introducing a new dimension, weaknesses in the old logic immediately became apparent. The previous structure simply wasn’t precise enough to clearly delineate the dynamic inventories. I fixed these issues directly in the Factory Ledger’s code, so that the calculation now runs absolutely consistently and the data can be transferred directly from the simulation into Controlling.

In doing so, we faced a methodological problem head-on: How do you sensibly calculate current assets in a running factory? A simple snapshot at a random point in time falls short, since inventory on the production lines fluctuates constantly. To smooth out this dynamic, I now use two parallel exponential moving averages—one that reacts quickly and one that is slower. The major advantage of this approach lies in the direct comparison: If the current actual inventory, the fast moving average, and the slow moving average are approximately identical, I know that the factory is in a stable steady state. Only then is the data reliable.

I’m now satisfied with this setup. It proves once again that Factorio plays out perfectly like a real-life consulting project. You analyze the initial state, identify the weak points, optimize the system step by step under realistic business conditions, and ultimately extract the maximum return from the structure.

More: 

• The excel sheet containing the data of the optimizations
• Alle calculations and the save games in one-zip archive
• the excel for the calculations 

Blog: , Seite:
Version: 1.4 April 2025, Kontakt: E-Mail Martin Wölker
Pirmasens, Germany, 2018-, ausgelesen am: , Licence CC BY