(by Acido)

Let’s say you already have a plan: what production at what time to build. Now you need a base design, which you will reproduce in the run.

Step 1: Determinate how many assemblers and raw materials you need.

For example we will use early game mall + Red and Green science that you need to produce at the rate of 1 science / second (good for 2.5h rocket).

Fig 1

The science itself takes only 7.5 iron /second. We add some intermediates overproduction to round it up to full iron belt (15 iron / second)

Fig 2

These assemblers also take 5.5 copper / second that are a bit less than half of a line. Keep in mind that you have extra 2 copper / second there. You may buffer it and use in the other part of the later. Or if you don’t find use for it then adjust the number of miners and furnaces.

Step 2: Connect all production chain in a simple way.

Just make it work. It doesn’t have to be perfect for now. “Simple” is important here – It should be easy to remember and fast to build. You will make them more advance if you need to later, after you get the filing of what part is not working out.

I will use the standard smelting line, than put GC assemblers between iron and copper.

Fig 3

Only GC and Red science use the copper, so I inclined to put Red science nearby. It also needs Gears.

Fig 4

For Green science we need Inserters and Belt. We can easy access iron, but there are limited options to get GC for Inserters.

Fig 5

Test that everything working.

Step 3: Think how you will build it step by step. Adjust you design based on that.

This is mostly important for early parts of the game, when you don’t have resources to build it at once. In late game you will usually build along the production chain (miners->smelting->intermediates->final product), to reduce the belt delay.

It will be our first automate iron line, so we start small with half of a belt. Without copper we can only automate Gears and Belts, but it is totally fine.

Fig 6

Belt production speed is limited by inserter, so it consumes ~2.5 iron / second, not 3. Gear assembler consumes 2 iron / second and 3 iron /second remain unused (not needs to buffer iron in a chest – it will buffer up to 100 in the last furnaces, just collect from them time to time). My inclination here is to add the second Gear assembler. Yes, we can change the plan on the fly on that step.

Fig 7

Next we add copper. 3 level 1 assemblers consume only 3 copper / second. Early Red science will consume 0.1 copper /second / assembler. We will run away with ¼ of a copper lane to start.

Fig 8

Belts, Gears, GC, automate research is a solid start. After that I will take care of steam power while Fast inserters are researching. Than second half of iron, Inserters, Green science.

Fig 9

Than expend copper to half line. Finally, we will upgrade some assemblers to level 2, and some Inserters to Fast inserters.

Fig 10

Step 4: Do the test run.

It doesn’t have to be a full run. Stick to the plan this time – see how it goes. You will get a better idea of what part of the base needs to be improved. Back to step 3, try other ideas. Soon you will find the base design that you are happy with!