Factorio All-Colour Science Production Line Ratios and Design

In Factorio, most players sail smoothly through red and green science, only to hit a wall when blue science arrives—suddenly oil refining, steel, and advanced circuits all struggle at once. This guide covers vanilla 2.0's full spectrum (red, green, black, blue, purple, yellow, white) with practical target ratios centred on 45 items/minute, plus scaling strategies from low SPM start-up to full expansion.

The audience ranges from beginners wanting to build all-colour packs in one go, to intermediate players unsure how much to feed the main bus versus set up on-site. I've hit the same acceleration wall—red and green were fine until blue hit, when I had to split advanced circuits into their own factory to stabilize. This article avoids claiming "one true way"; instead, it lays out the design philosophy behind factories that rarely jam, grounded in concrete numbers.

Preconditions for an All-Colour Science Line in Factorio

What This Article Covers

This piece addresses how to make vanilla 2.0's all-colour science (red, green, black, blue, purple, yellow, white) work as a single research feeding system. The official Factorio each pack's role and placement, but here we focus on the design bottlenecks rather than a bare overview.

Specifically: each colour's role, primary materials, common jam points, assembler ratios, and layout philosophy. Red and green slot in as early-game continuations. Black introduces military-side materials and shared resources. Blue marks the jump: oil, steel, advanced circuits, and engines all converge at once. Purple and yellow then squeeze development circuits and low-density structures harder, plus lubricant strain rises sharply. White isn't continuous production at all—it's rocket launches. When you see the ratios side by side, it's clear: all-colour lines aren't about lining up the packs; they're about whether you've built enough intermediate supply chains ahead of time.

Baseline infrastructure worth stabilizing: iron plate, copper plate, steel, electronic circuits (green circuits), and oil products (including sulphur, plastic, solid fuel, and lubricant from yellow onwards). I've entered blue science with weak base smelting, watched iron plates vanish alongside oil, and had to redirect iron belts to red circuits just to recover. In hindsight, the bottleneck wasn't blue—it was overestimating my capacity at red/green. Most failing blue-stage factories have a thin spot somewhere upstream.

Transport planning anchors well on yellow conveyor belts at 15 items/second. If you're building a main bus, use that ceiling as your benchmark: "How many iron lines do I need?" "Should I split green circuits into a dedicated line?" becomes answerable. Red and green flow by feel, but once you eye purple and beyond, splitting electronic circuits into an independent upstream line pays dividends. Main-bus designs do this everywhere because green circuits feed nearly every downstream process.

Assembler 2 (crafting speed = 0.75) provides a workable default. A solid rule of thumb: red 5 : green 6 : black 5 : blue 12 : yellow 7 : purple 7 at roughly 45 items/minute each. This is an operational guideline, not gospel. Assembler 3 users should scale that down to about 0.6× (= 0.75 / 1.25). Always recalculate against the actual recipe times before final construction.

White is its own beast. Red through purple are assembly-line design; white is launch logistics. One satellite launch yields 1000 space science packs at once, so think in terms of launch frequency, not per-second throughput. Low-density structures take 20 seconds each and need 1000 per rocket, meaning solo production is a 5+ hour grind. By that stage, white science isn't a parallel line—it's the endgame factory's resource footprint.

Space Age bears mention: the same "all-colour" framing breaks down entirely. New pack types, per-planet equipment gaps, space logistics, and quality mechanics mean hauling vanilla 2.0 main-bus logic directly won't work. This article stays focused on vanilla 2.0 all-colour ground-based design. Space Age deserves separate treatment.

To anchor terminology and roles, the official Factorio Science pack wiki|https://wiki.factorio.com/Science_pack is the reference. This article presumes you've synced names and high-level layout there, then dives into the assembly and bottleneck design.

Science pack - Factorio Wiki

wiki.factorio.com

Setting Your Target Throughput First: 45 Items/Minute Is Plenty for Beginners

Low SPM (≈ 0.75 items/second): Advantages and Pitfalls

Before optimizing for speed, ask yourself: at what rate can I keep everything moving without constant jams? I once built for 1+ items/second from the start, then doubled labs before upstream caught up. Blue immediately strangled oil, yellow/purple squeezed circuits, power hung, and the factory seized. Once I reset to about 0.75 items/second and accepted slower research, the whole thing stabilized.

The upside of this band: upstream load collapses. Red and green look manageable until black, blue, yellow, and purple each add their own layer. Research-only acceleration doesn't help if supply chains are starved. Laboratories accept packs from a half-fed belt; they just run slower. Padding labs without feeding the lines only creates dead time.

Low SPM lets you prioritize reliability and expansion room over raw speed. Yellow belts cap at 15 items/second, so rather than filling them tight from day one, you can ask "Can I run a second iron belt here later?" "Is there space to tap oil differently?" Planning for duplication beats designing for peak efficiency on the first try. Electronic circuits, steel, and oil benefit most: initial overload often causes the crash that better-thought-ahead design would prevent.

Standard 45 Items/Minute: the Sweet Ratio Spot

Once comfortable with low-SPM runs, scale up to roughly 45 items/minute per colour. That lands comfortably between "serious growth" and "still manageable without advanced optimizations." The ratio red 5 : green 6 : black 5 : blue 12 : yellow 7 : purple 7 slots neatly into that ballpark and stays friendly to Assembler 2 benchmarks.

Why this proportion works: blue suddenly doubles, making the shift in factory demand crystal clear. Yellow and purple appear balanced with black, yet actually squeeze circuits and steel much harder. At 45 items/minute, that strain is visible but not crushing—you experience the "which bottleneck to uncork first" problem at a survivable scale. My own experience bears this out: starting at 0.75/s, watching where things first jam, then scaling to 45/min to see the full colour spectrum without panic.

A useful progression: start at ≈0.75/s, let it run stably, then proportionally scale all colours up to 45/min. This avoids the trap of blue outpacing everything else; you grow the whole thing together and spot imbalances early.

High SPM and Balanced All-Colour: Prerequisites

Don't rush to high SPM or per-colour balancing until all your supplies flow at roughly the same tempo. Oversizing labs before the rest catches up creates research "starvation" illusions: you see red and green surplus while blue, yellow, and purple dry up. That happens because high-end production is a system; no single bottleneck stands alone.

The 15 items/second yellow-belt ceiling becomes brutal at high SPM. Carrying iron or copper from one end to the other works at 45/min, but multi-pass expansion quickly stuffs the belt. Long-distance feedlines that work at 45/min will choke before doubling. You'll need either:

• Extra belt runs per material (red or blue belts), or
• Local production and train logistics.

High SPM designs work best when you stop thinking "one deep bus" and start thinking "clustered factories linked by trains". Medium-density output often doesn't travel well; making it on-site and shipping only high-level products scales further.

My own pattern: 0.75/s to stabilize, scale to 45/min, then multi-module factories at 1.5+/s. Each jump reveals where the next pressure point lives.

Overview of Red, Green, Black, Blue, Purple, Yellow, White

All-colour lines don't jam equally. Red and green are simple; blue is where stone and steel and oil converge, breaking many starter factories. Yellow and purple then compress circuits and steel further. White offloads to rockets, doubling the endgame load. The order of pain for most players: blue → yellow/purple → white. Red, green, and black stumble mainly on intermediate distribution, not core recipes.

Here's the landscape:

Red (Automation Science): Iron Gear Wheels Are the Hidden Gremlin

Automation science packs need copper plates and iron gears. They look light—and they are, in isolation—but gears disperse across the whole factory. Red doesn't starve for gears; gears starve for placement.

Gear locations matter enormously. If you co-locate red science and gear production, that spot runs dry while downstream green or engine builds are blocked. Lesson: don't defer gear design. Decide early whether you're centralizing them or spreading them; a bad choice leaves red suspended.

Green (Logistic Science): Belts and Inserters Hide Circuit Demand

Logistic science packs use inserters and transport belts. Internally, gears and electronic circuits are consumed too. Where red bled gears, green bleeds circuits and gears.

Jam signs: inserter lines starve, belt lines starve, or both oscillate. This usually points to competition for gears and circuits, not a shortage of inserters themselves. Early on, "why is green so slow?" rarely means green's not producing fast enough; it means the feeder chains are misconfigured.

Black (Military Science): Ammo and Explosives Split Your Stone Pipeline

Military science packs demand piercing rounds, grenades, and concrete walls—all offshoots of warfare. Stone and iron pressure spike here.

Black doesn't crater factories like blue does, but it forces a choice: feed ammo lines separately, or watch them steal stone from the main bus. Initial struggles usually mean ammo isn't a dedicated sub-factory; it's cannibalizing your general pipeline.

Blue (Chemical Science): The First Hard Wall

Chemical science packs require engine units, sulphur, and advanced circuits. This is the first major pivot. Solid-phase reasoning stops working; oil refining and fluid logistics enter the picture.

Blue's danger: every single input—engines, sulphur, circuits—pulls from a heavier upstream. Engine units need steel; sulphur needs oil and refining; circuits need plastic and copper. All three converge on steel, oil processing, or advanced-circuit output at once. When blue launches, a factory's research speed crashes unless oil, steel, and circuit supply are already fat.

I felt this acutely: my blue smelter hummed, but the moment I throttled research labs onto blue packs, oil runs began stuttering, steel vanished, and advanced circuits flickered. The bug wasn't blue; it was the foundation. This is the colour where "build the supply first, let blue flow as a side effect" wisdom hits hardest.

Chemical science pack - Factorio Wiki

wiki.factorio.com

Purple (Production Science): Module Hunger Beats Line Weight

Production science packs demand electric furnaces, productivity modules, and rails. Superficially, rails dominate (lots of items); actually, productivity modules and furnaces pin down advanced circuits hard.

Purple dangles a pair of loads: rail is straightforward, modules are not. Modules suck circuits and steel relentlessly. Pairing purple with blue (which also hungers for circuits) means circuit supply takes a second jump. If your blue line barely holds, purple will collapse it.

Yellow (Utility Science): Distribution Frames and Low-Density Structures

Utility science packs use advanced circuits, flying robot frames, and low-density structures. Yellow mirrors blue's pain: circuits are hungry, but now you add lubricant (from heavy oil) and copper-steel-plastic composite demand.

Flying robot frames drag lubricant requirements skyward. Lubricant means upstream oil reshuffling. Low-density structures pull copper, steel, and plastic all at once—and that competition is fierce. Yellow is blue expanded across more resource vectors.

White (Space Science): Rockets, Not Continuous Lines

Space science packs don't come from assemblers. They come from rocket launches. One satellite flight yields 1000 at once.

The crunch: low-density structures. Each rocket needs 1000; each takes 20 seconds solo. Pair that with the fuel, processing units, and rocket engines, and white stops being a "fifth parallel line" and becomes an entirely separate production ecosystem. By white, your factory isn't "five colours running in parallel"; it's "endgame resource management."

Assembler Counts and Practical Ratios

Standard 45 Items/Minute (Assembler 2 Baseline)

Using Assembler 2 (speed 0.75), the red 5 : green 6 : black 5 : blue 12 : yellow 7 : purple 7 ratio targets roughly 45 items/minute per colour. This assumes no modules, bare assemblers.

More important than raw count: local supply beats global count. Even with the right ratio, gears only fed via a long bus often starve local demand. I discovered yellow belts bottleneck far sooner than my math predicted; actual throughput loss during transport crashed designs that looked balanced on paper.

Low SPM Scaled-Down Version (≈ 0.75 Items/Second)

Shrinking the whole build uniformly—keeping ratios, cutting all numbers proportionally—often runs smoother than trying a "lite" subset. Standard ratios become your scaling anchor, not your starting point. I typically build red/green/black first at low SPM, reserve space for blue and beyond, then scale proportionally.

Where low SPM shines: weak points emerge fast. If circuits or steel waver even at 0.75/s, that's a design flaw, not a volume problem. Fixing it at low SPM costs far less than reworking a high-SPM blob.

Black and Blue Intermediate Breakdowns

Neither colour can be balanced by raw assembler count alone. Intermediate supply chains dominate.

Black internally needs:

• Piercing rounds (~12 assemblers for ~45/min)
• Regular rounds (~4, feeding into piercing)
• Grenades (~8)
• Concrete walls (~1–2)

The piercing-round line is the spine; everything else decorates it.

Blue pairs assemblers with:

• Engine units (~10 assemblers, roughly 2:1 to blue pack assemblers)
• Chemistry (refinery and cracking)
• Advanced-circuit line

Blue 12 + Engine units 10 is closer to reality than blue 12 alone. And engine units should live next to blue, not on a distant bus; they're heavy on gears and oil, and long pipes bleed pressure. I've learned: put blue and engines side by side, not end-to-end of a main bus.

Design Principles: Main Bus vs On-Site Production

What Goes on the Main Bus

Decide early: which materials survive long transport? My anchor: iron plates, copper plates, and electronic circuits (especially circuits). These are your bus backbones.

Electronic circuits deserve special attention. They feed red, green, blue, and nearly everything downstream. Make circuits in a dedicated upstream line, ship the finished product widely. Distributed small-batch production spreads consumption across sites and breaks when demand spikes. One master circuit factory feeding everywhere is far more stable.

Oil products (sulphur, plastic, lubricant) belong on their own pipe network, not mixed into belts. Especially lubricant—yellow's arrival suddenly quadruples its need. Plan ahead.

What Stays Local

Middle-tier items jam on long buses. Gears, pipes, copper cables, ammunition: these cluster at their use sites.

Gears are my textbook example. They're everywhere (red, green, engines, machinery), but demand oscillates by location. A long bus stuffed with gears works until you're three screen-lengths away and a gear source runs dry. Short local lines: problem solved.

Intermediate outputs (furnaced copper wire, assembled batteries, munitions) follow the same rule. Bring plates and ores to the job site; don't pre-fab and haul.

Long-Line Bottlenecks and Sectioning

A single linear main bus seems elegant until blue forces you to branch everywhere. The longer the bus, the harder it is to diagnose: "Is iron too thin, or is my green-circuit tap strangling the downstream?" Blame disperses.

Answer: section your bus. Solid-phase (red/green/black), then chemical (blue onwards), then synthesis (purple/yellow heavy materials). Insert inter-section buffers—storage chests or capped flows—so one crisis doesn't cascade backward.

I often feed major production zones through a buffer, then redistribute onward. This isolates shocks: if blue production spikes, it pulls from the blue buffer, not directly starving red.

Oil's worse; treat it as a separate infrastructure. Pipes, pump jacks, refineries, chemical plants—don't daisy-chain them into the solid belt network. Dedicated pipes and tank farm = stable, compartmentalized supply.

Common Pitfalls and Fixes

Blue Science: Oil and Steel Both Failing at Once

Blue enters and suddenly oil and steel run dry. But the root cause is earlier: red or green smelting was already marginal. Blue didn't cause collapse; it exposed undersizing.

Fix: don't add blue assemblers; add ore and refinery throughput. Increase mine output, smelter count, and refinery input first. Blue output then naturally follows.

Oil specifically: cracking and output balance often break. If heavy oil backs up, light oil underproduces, and advanced circuits starve for plastic. Spend time on refinery layout and recycle loops before scaling blue.

Yellow / Purple: Advanced-Circuit Drought

Both colours hammer circuits. Yellow adds lubricant pressure; purple adds module hunger. Circuits are the shared battleground.

Mistake: growing yellow's throughput without checking purple's hunger. You solve one, the other implodes.

Fix: feed circuits from a dedicated line independent of blue. Carve out a second circuit factory, or reroute a circuit line to supply yellow/purple separately from blue. This sounds redundant; it's actually the only stable pattern. My factories that faltered here all tried to split a single circuit bus four ways.

Lubricant strain: anticipate it during blue design. Yellow's launch will double or triple demand. Pre-install pipe capacity that you don't fill yet.

Lab Growth: The Research Speed Trap

Temptation: drop more labs into empty space. Never do this without checking pack supply.

If red and green run surplus, blue lags, and purple starves, adding labs doesn't help. You've just put more pull on an already-broken input. Labs are a downstream load; they're only useful once upstream flows steadily.

Reliable pattern: grow all packs proportionally until they all fill labs simultaneously. Then expand labs a little. Repeat until the next bottleneck surfaces.

Construction Order and Next Steps

Phase 1: Red, Green, Black Stability

Don't chase all colours at once. Start with red, green, black as a unit. Shrunk-down versions of the standard ratios work fine. The goal: continuous supply of iron, copper, and stone, with gears and stone-derived goods flowing steadily.

Black is where this trio reveals weak spots. Watch stone and iron consumption patterns shift. If black causes visible jams, your red/green base was already stretched; black just exposed it.

This phase teaches you main-bus thinking, intermediate sourcing, and the boundary between long-haul and local production. Lock these in before blue.

Phase 2: Pre-Blue Audit

Before blue arrives, verify oil refining, steel, and advanced circuits are robust. Not "adequate for now"—robust for the load blue brings.

Triple-check:

• Oil input: pump jacks, railheads, or however you feed crude
• Cracking balance: heavy → light, as needed
• Advanced circuits: dedicated line or shared? (Recommend dedicated)
• Steel: electric furnaces running smoothly?

If any wavered during red/green/black, blue will bury them. Fix upstream before blue launches.

Phase 3: Yellow and Purple at Parity, Then Lab Scaling

Bring yellow and purple online together, at the same throughput target. Don't favour one; the asymmetry will haunt you.

Prepare a separate advanced-circuit input for yellow/purple, or run a second circuit factory. Prepare lubricant provisioning in advance. These aren't "nice to have"; they're load-bearing.

Once both run stably, expand labs proportionally. Labs should grow as packs flow equally; never before.

What Space Age Changes

Space Age redesigns the entire premise:

• New science types, per-planet equipment gaps, space logistics, and quality mechanics all arrive.
• Ground-based simultaneous production (main bus logic) no longer fits.
• Balanced-speed modular designs (e.g., 1.5/s chunks, replicated across planets) become far more practical.
• Quality benefits scale best at final assembly, not sprinkled throughout the supply chain.

Space Age deserves separate coverage. Vanilla 2.0 all-colour design is foundational but not portable directly. Learn vanilla stability first; treat Space Age as a new game.

Conclusion

Start with a throughput you can sustain without constant fires. Red 5 : green 6 : black 5 : blue 12 : yellow 7 : purple 7 at 45 items/minute is a solid checkpoint. Centralise circuits and oil; spread gears and construction materials locally. Anticipate blue by thickening oil and steel before it launches. Grow all colours proportionally; labs follow supply, never lead it.

There is no "correct" factory. The strongest approach combines low initial SPM → stable multi-colour flow → deliberate scaling → space-efficient per-module designs. Watch where production falters, widen that section, and repeat. Factories that never jam are rarely the most elegant; they're the ones where you planned the pressure points right.