What size feed mill do I need? From animal numbers to tonnes per hour.
Size a feed mill in six steps: convert your animal population to annual tonnes, add third-party and growth volume, divide by the production hours you will actually run, divide by 0.85 for effective utilisation, add 20% headroom, then round up to a standard press size. As reference points on a single 8-hour shift over 300 days: 1,000,000 broilers a year needs about 2 t/h, 250,000 layers about 5 t/h, 2,000 dairy cows on concentrate about 3 t/h, 5,000 head of feedlot cattle about 8 t/h, 2,500 sows farrow-to-finish about 7 t/h, and 8,000 tonnes of tilapia about 6 t/h. Check the result against your smallest recipe run before you buy — mixer batch size, not tonnage, is what constrains mills with many small or medicated batches.
- Feed-per-head figures for 7 production systems
- Six-step sizing method with a stated basis
- Worked examples from population to t/h
- Why batch size can matter more than tonnage
Visual scope of feed mill planning
Short answer
Size a feed mill in six steps: convert your animal population to annual tonnes, add third-party and growth volume, divide by the production hours you will actually run, divide by 0.85 for effective utilisation, add 20% headroom, then round up to a standard press size. As reference points on a single 8-hour shift over 300 days: 1,000,000 broilers a year needs about 2 t/h, 250,000 layers about 5 t/h, 2,000 dairy cows on concentrate about 3 t/h, 5,000 head of feedlot cattle about 8 t/h, 2,500 sows farrow-to-finish about 7 t/h, and 8,000 tonnes of tilapia about 6 t/h. Check the result against your smallest recipe run before you buy — mixer batch size, not tonnage, is what constrains mills with many small or medicated batches.
Start from tonnes, not from a target machine
The most reliable sizing starts with the animals you feed, not with a capacity someone quoted you. Multiply your population by the feed-per-unit figure for the species and cycle length, then count every phase you will actually mill. Pullet rearing feed, sow gestation feed and starter phases for a second species are forgotten routinely, and together they can add 15–25% to the annual total. If you intend to sell feed to third parties or expand within the mill's fifteen-year life, add that tonnage now: extending a tower later costs far more per tonne of added capacity than specifying one size up at the start.
The two divisions everyone gets wrong
Annual tonnes become tonnes per hour through two divisions, and both are commonly skipped. The first is the shift pattern: one 8-hour shift over 300 days gives 2,400 production hours a year, two shifts give 4,800 and roughly halve the capacity you must buy. The second is effective utilisation — divide by 0.85, because recipe changeovers, flushing between medicated and non-medicated batches, cleaning and short stops mean no mill produces at nameplate for a whole shift. A mill sized without these two steps looks affordable in the feasibility study and runs at 100% of nameplate from day one, which leaves no room for a breakdown, a demand peak or a bad month.
One shift or two
Buy for one shift and keep the second in reserve. Two-shift operation is the cheapest way to add output on paper, but it removes the maintenance window and turns every breakdown into a delivery failure. Mills that genuinely plan two shifts from the start should still size intake, silos and finished-feed bins for peak flow rather than average, because those buffers are what make a later shift change possible. Intake in particular should be sized at 2.5–3× mill capacity so a 30-tonne truck discharges in under twenty minutes; a slow pit creates a queue that no amount of production capacity downstream can clear.
When batch size, not tonnage, sets the specification
A mill producing a handful of large-volume recipes is constrained by tonnes per hour. A mill producing many small, medicated or species-specific runs is constrained by something else: the mixer must produce your smallest run without splitting it, and the micro-dosing system must reliably weigh the smallest premix inclusion inside that batch. Add changeover and flushing time between runs and the effective utilisation of a many-recipe mill falls well below 85%. If your recipe book is long, size the mixer and the dosing accuracy first and let the tonnage follow — the reverse order produces a mill that is technically large enough and practically unable to make its own product range.
Feed requirement by production system, with worked examples
| Production system | Unit | Feed per unit | Worked example | Required capacity |
|---|---|---|---|---|
| Broilers (grow-out, 6 cycles/yr) | per 1,000 birds placed | ~4.2 t per cycle to 2.4 kg live weight at FCR 1.55 | 1,000,000 birds/yr → ~4,200 t/yr | ~2.0 t/h |
| Layers (in production) | per 1,000 hens | ~41 t per year at 112 g/hen/day | 250,000 hens → ~10,300 t/yr | ~5.0 t/h |
| Pullets (rearing to 17 weeks) | per 1,000 pullets | ~6.2 t per rearing cycle | 300,000 pullets/yr → ~1,900 t/yr | ~0.9 t/h |
| Dairy cows (concentrate only) | per 100 cows | ~290 t per year at 8 kg concentrate/cow/day | 2,000 cows → ~5,800 t/yr | ~2.8 t/h |
| Beef feedlot | per 1,000 head on feed | ~3,300 t per year at 9 kg/head/day | 5,000 head → ~16,500 t/yr | ~8.0 t/h |
| Swine (farrow-to-finish) | per 100 sows | ~600 t per year including progeny to 115 kg | 2,500 sows → ~15,000 t/yr | ~7.3 t/h |
| Tilapia / catfish (pond) | per 100 t of fish harvested | ~150 t of feed at FCR 1.5 | 8,000 t fish/yr → ~12,000 t/yr | ~5.8 t/h |
Six-step sizing method
| Step | What to do |
|---|---|
| 1. Convert population to annual tonnes | Multiply your animal numbers by the feed-per-unit figure for that species and cycle length. Count every phase you will actually mill — pullet rearing and sow gestation feed are routinely forgotten and can add 15–25% to the total. |
| 2. Add third-party or growth volume | If you plan to sell feed externally, or expand within the mill's 15-year life, add that tonnage now. Extending a tower later costs far more than specifying one size up at the start. |
| 3. Choose the shift pattern | Divide annual tonnes by production days and shift hours. One 8-hour shift over 300 days gives 2,400 hours a year; two shifts give 4,800 and roughly halve the capacity you need to buy. |
| 4. Apply an effective-utilisation factor | Divide by 0.85. Recipe changeovers, flushing between medicated and non-medicated batches, cleaning and short stops mean no mill produces at nameplate for a full shift. |
| 5. Add headroom and round to a standard size | Add 20% and round up to the nearest press size the manufacturers actually build. A mill that runs at 70–85% of nameplate is healthy; one that must run at 100% has no room for a breakdown or a bad month. |
| 6. Check the batch size against the smallest recipe | Mixer batch size must let you produce your smallest run without splitting it, and your micro-dosing must reliably weigh the smallest premix inclusion in that batch. This constraint often decides the mixer, not the tonnage. |
Required-capacity figures assume one 8-hour shift over 300 production days at 85% effective utilisation, with a 20% headroom allowance. Feed-per-unit figures are commercial planning averages; substitute your own breed-guide intake and FCR data where you have it, since climate, genetics and management move them materially.