Main Bus Design: How Many Lanes Do You Need?

The main bus is the backbone of most Factorio factories — a set of parallel belts carrying raw and intermediate materials from production to consumption. But "how many lanes?" has no single answer. It depends on your target SPM (science per minute), which belt tier you use, how far you plan to expand, and how much space you reserve for materials you have not started making yet.

This page does what most guides cannot: it computes the actual material demand at 60, 90, and 1000 SPM using the ratio solver built from data/2.1/recipes.json, then divides by belt throughput (also computed from game data) to give you belt counts. No forum posts, no memorised ratios — just the numbers the game data produces.

Figures computed from game version 2.1.12, updated 2026-08-21

Belt Throughput (the Baseline)

Before asking how many lanes you need, you need to know how much one lane carries. Belt throughput is computed from data/2.1/logistics.json using the formula speed × 8 × 60:

A yellow belt carries 15 items/s; a turbo belt carries 60 items/s — four times as much. This means the same factory needs four times fewer lanes with turbo belts than with yellow belts. The belt tier you choose is the single biggest factor in how wide your bus will be.

Lane Counts by SPM Target

The table below is computed by running the ratio solver for each science pack at the target SPM, summing the demand for each bus material across all packs, and dividing by belt throughput. "Yellow belts" uses the yellow transport belt (15 items/s); "Turbo belts" uses the turbo transport belt (60 items/s).

If a cell shows NOT FOUND, the material was not found in the supply chain stages for any science pack at that SPM. This means no science pack recipe chain produces or consumes that material as an intermediate step.

Reading the 1000 SPM Numbers

The 1000 SPM column deserves special attention because it represents the scale where a main bus starts to strain against practical limits. At 1000 SPM, iron plate demand reaches 1845.83 items per second — that is 124 yellow belts or 31 turbo belts dedicated solely to iron. Copper plate is close behind at 1325.00 items/s (89 yellow / 23 turbo).

To put this in perspective: a standard 4-lane yellow belt group (the most common bus design for new players) carries 60 items/s total. At 1000 SPM, iron plate alone needs more than double that. A yellow-belt bus for 1000 SPM iron and copper combined would require 213 lanes — just for two materials. Add steel, gears, circuits, and plastic, and the bus exceeds 295 yellow lanes in total.

This is why megabase designs typically abandon the main bus in favour of train-based logistics or city-block layouts. A bus of this width is technically possible but impractical: it is difficult to route, difficult to tap without throughput loss, and takes up enormous space. The turbo belt reduces the total to roughly 78 lanes — still substantial, but manageable.

The crossover point — where a bus becomes too wide to manage — depends on player preference and factory layout, but the data suggests it falls between 300 and 500 SPM for yellow belts and between 800 and 1200 SPM for turbo belts. Above those ranges, the lane count grows faster than most players can practically manage.

What Determines Bus Width

The table above gives you the raw numbers, but the actual width of your bus depends on several decisions that the data alone cannot make for you:

1. Target SPM

The most obvious factor. A 60 SPM factory needs far fewer lanes than a 1000 SPM factory. The demand scales roughly linearly — doubling your SPM target roughly doubles the belt count for each material. But "roughly" is important: some materials are used by multiple science packs, so the demand curve is not perfectly linear.

2. Belt Tier

Using turbo belts (60 items/s) instead of yellow belts (15 items/s) reduces the lane count by a factor of four. This is the most effective way to keep your bus narrow. However, turbo belts require turbo-transport-belt recipes, which are locked behind Space Age research — they are not available in a base-game-only factory.

3. Which Materials to Bus

Not every material belongs on a bus. The general criteria:

• High consumption rate — if a material is used by multiple production lines at high volume, it deserves a bus lane. Iron plate and copper plate are the classic examples.
• Many consumers — if a material is used by many different recipes, busing it avoids routing individual belt lines to each consumer. Electronic circuits are used in almost everything.
• Not worth busing — low-volume materials or materials with a single consumer are better served by direct belt lines or logistic robots. Stone bricks for walls, or sulfuric acid for battery production, are often not bused.

The decision is a judgment call, not a formula. The table above shows which materials have the highest demand — those are the strongest candidates for bus lanes. But a material with moderate demand and many consumers (like iron gear wheels) may still be worth busing for convenience, even if its raw item/s count is lower than iron plate.

What the Numbers Say About Material Selection

The bus table provides quantitative support for the qualitative criteria above. At 60 SPM (a typical early-game target), the demand distribution is highly skewed:

• Iron plate and copper plate dominate, consuming 8 and 6 yellow belts respectively at 60 SPM. Together they account for over half of the total bus width.
• Electronic circuits (2 belts) and plastic bars (2 belts) form the second tier — high enough to justify bus lanes, but far below the plate demand.
• Processing units (1 belt) and sulfur (1 belt) are borderline. A single belt each is minimal, but their consumption by multiple science packs justifies a bus lane for convenience.
• Lubricant shows NOT FOUND — it does not appear in any science pack supply chain. This is a strong data-driven signal that lubricant should not be on a science-focused bus. It may still be needed for engine unit production or electric motor assembly, but those are separate production lines that do not require bus infrastructure.

The pattern holds at higher SPM: iron and copper scale together (approximately 1.4 to 1 ratio by items/s), while secondary materials grow proportionally. The implication for bus design is that you should allocate bus width roughly in proportion to demand: more lanes for iron and copper, fewer for sulfur and processing units.

4. Expansion Room

A bus that is exactly wide enough for your current SPM will need to be rebuilt when you expand. Most experienced players add 2-4 empty lanes (or a gap for future lanes) on one side of the bus. This costs nothing now and saves a painful rebuild later.

When a Main Bus Becomes a Bottleneck

A bus can bottleneck in two different ways, and the symptoms are different:

Bandwidth bottleneck — the belt is saturated and cannot carry more items. The symptom: items back up at production inputs, and consumers at the end of the bus starve. The fix is either upgrading belt tier (yellow to red to blue to turbo) or adding parallel lanes for the saturated material.

Tap bottleneck — the belt has capacity, but the inserters pulling from it cannot keep up. The symptom: the belt remains full past a tap point, but the consumer still does not get enough. This is harder to diagnose because the belt looks fine. The fix is upgrading inserters (to fast, stack, or bulk inserters) or adding more tap points in parallel.

Note: inserter throughput cannot be computed from the available game data (the speed field is null for inserters in data/2.1/logistics.json). We know the rotation speed and extension speed, but not the items-per-second. This means we cannot tell you exactly when a tap bottleneck will occur — only that it is a different failure mode from a bandwidth bottleneck.

Diagnosing Bottlenecks in Practice

The two bottleneck types require different diagnostic approaches:

• Bandwidth bottleneck diagnosis: Look at the belt between the last tap and the source. If items are backing up at the source end but the belt is empty near the last consumer, the belt is saturated. The fix is adding parallel lanes or upgrading the belt tier for that material. The bus table above tells you exactly how many lanes you need at each SPM — if you have fewer lanes than the table recommends, you have a bandwidth bottleneck by definition.
• Tap bottleneck diagnosis: The belt remains full past the tap point, meaning items are available but cannot be pulled off fast enough. This is harder to diagnose because the belt looks healthy. The fix involves inserter upgrades (to fast, stack, or bulk inserters) or parallel tap points. Since inserter throughput is NOT FOUND in the game data, we cannot provide exact numbers for when this will occur — but the symptom (full belt, starving consumer) is distinctive enough to identify.

A third, less obvious failure mode is the splitter bottleneck: if a splitter on the bus cannot handle the full throughput of the incoming belt, it becomes a bottleneck even when the belt itself has capacity. This is why all splitters on a bus should match the belt tier — a yellow splitter on a turbo belt line limits the effective throughput to yellow belt levels.

How Turbo Belts Change the Bus

The turbo transport belt (0.125 tiles/tick, 60 items/s) was introduced in Factorio 2.0 with the Space Age expansion. Its impact on main bus design is significant:

• Four-fold density increase — a single turbo belt carries as much as four yellow belts. A bus that needed 16 yellow lanes for iron plate can be served by 4 turbo lanes.
• Existing buses may become over-built — if you designed your bus with yellow or red belts and later unlock turbo belts, your bus will be wider than necessary. This is not a problem (extra width does not hurt), but it means you could have built a narrower bus from the start if you had planned for turbo.
• Splitter and underground belt upgrades — every component on the bus (splitters, underground belts) must also be upgraded to turbo tier to maintain throughput. A single yellow splitter in a turbo line becomes a bottleneck.

Belt Tier Upgrades: When to Switch

Most players start with yellow belts and upgrade as they unlock better tiers. The data shows exactly how much each upgrade buys you, using iron plate demand at 1000 SPM as the benchmark:

The biggest single jump is yellow to red (doubling throughput), followed by red to blue (1.5x) and blue to turbo (1.33x). The yellow-to-turbo upgrade is a 4x improvement — the equivalent of compressing four lanes into one. This is why turbo belts are transformative for bus design: they do not just reduce width, they change the fundamental economics of how much space a bus occupies.

The practical upgrade order is usually yellow, then red, then blue, with turbo reserved for late-game or Space Age playthroughs. Each upgrade lets you either narrow the existing bus or absorb more SPM demand without widening it. The key insight from the data: you do not need to upgrade all belts simultaneously. Upgrading only the saturated lanes (typically iron and copper) while leaving low-demand lanes on cheaper belts is a valid cost-saving strategy.

When You Should NOT Use a Main Bus

A main bus is not the only layout, and it is not always the best one. Consider alternatives when:

• Your factory is small — for a 10-30 SPM base, a bus is more infrastructure than you need. Direct belt lines from production to consumption are simpler and faster to build.
• You are making a megabase — at very high SPM (500+), a single bus becomes impractically wide. Megabases typically use train-based logistics or city-block designs instead.
• Your production is decentralised — if you have separate production blocks for different science packs (a " spaghetti" or modular layout), a central bus may not fit the architecture.
• You rely on bots — logistic robot networks can replace belts for low-volume, high-variety materials. A hybrid approach (bus for bulk, bots for diversity) is common.

The bus is a tool, not a requirement. The best layout is the one that matches your factory's scale, your expansion plans, and your personal preference for visual organisation.

Applies When

Related

• Smelting Ratios: Furnaces to Belts — How many furnaces fill a belt of plates? Directly relevant to bus input sizing.
• Train Throughput: Wagons vs. Belts — When does a train outperform a belt? Relevant for bus-to-train transitions at high SPM.
• Belt Balancer Designs — Splitter counts and throughput for balanced belt networks. Useful for designing bus splits and merges.