foratvise
Get Started
Forming, baking and cooking

Extrusion

Cooking extrusion of cereal-based expanded products (snacks, breakfast cereal, pet food)

Validated modelOEE bound 0.979 [0.947–1.000]20,000 Monte Carlo samples · seed 20261115 · computed November 16, 2026

In short

Extrusion can use at most 98 of every 100 planned hours. The rest goes to limits no plant can avoid.

  1. 98usable at most
  2. 2are a physical limit

The three limits that set it

Running time: up to 100 %

The rest are stops imposed by physics or by regulation: cleaning, changeovers, defrosting.

Speed: up to 100 %

How close to its maximum speed it can run without damaging the product or the equipment.

Good product: up to 98 %

Some product falls outside specification even with the process well under control.

See the technical detail

Technical detail

Uncertainty intervals, physical models per station and sources, for anyone who wants to check each figure.

1Scenario

Moistened corn grits are fed to a single-screw cooking extruder. Viscous dissipation and barrel heating raise the mass to the melt temperature, the starch is gelatinised and melted, and the melt is forced through a die where the superheated water flashes to steam and expands the product. The reference geometry and operating point are those of a documented laboratory extruder (19 mm barrel, length/diameter 25, 150 rpm, barrel sections at 25/70/170/175 °C); the same equations apply to industrial single-screw machines.

Provenance of inputsComputed1Declared assumption17Verified quote18

2Results

Point value from the nominal parameters and 95% interval from 20,000 Monte Carlo draws of the uncertain parameters.
  • Gelatinisation enthalpy of the starch per kg of feed7.02 [6.21–8.05] kJ/kg
  • Sensible heat of dry matter and water from feed to melt temperature305 [271–332] kJ/kg
  • TheoreticalThermodynamic minimum: net enthalpy change of the starch (gelatinisation)7.02 [6.21–8.05] kJ/kg
  • Adiabatic energy input without heat recovery from the extrudate (sensible heat plus gelatinisation; computed, not a documented BAT)312 [279–339] kJ/kg
  • Best documentedDocumented specific mechanical energy of corn extrusion (lower bound of the total input)900 [905–1,095] kJ/kg
  • TypicalBAT indicative specific energy of dry pet food plants (whole plant)1,404 [1,414–1,791] kJ/kg
  • Water flashed at the die per kg of feed0.0669 [0.0578–0.0728]
  • Moisture of the extrudate after the die flash (wet basis)0.110 [0.092–0.143]
  • Drag-flow capacity of the screw net of die back-flow6.78 [4.25–10.13] kg/h
  • Throughput at which the melt stays the minimum cooking time in the hot zone4.98 [1.67–8.91] kg/h
  • Throughput limit (lower of drag and cooking capacities; declared cooking time and back-flow, no ceiling)4.98 [1.67–8.03] kg/h
  • Minimum residence time at cooking temperature (declared, no ceiling)20.0 [10.8–39.3] s
  • Mean residence time in the hot zone at the maximum throughput20.0 [12.0–39.3] s
  • Start-up and shut-down scrap per run (share of run production)0.00861 [0.00273–0.03237]
  • TheoreticalMaximum dry-matter yield (no forced stop, hence no transient losses)1.00
  • Dry-matter yield with the start-up and wash-out scrap of the declared runs0.991 [0.968–0.997]
  • Share of product outside the bulk-density specification0.0214 [0.0000–0.0526]
  • TheoreticalMaximum OEE quality factor (bulk-density capability)0.979 [0.947–1.000]
  • Quality factor with the start-up and wash-out scrap of the declared runs0.970 [0.936–0.996]
  • TheoreticalPerformance factor at the physical maximum throughput1.00
  • TheoreticalMaximum availability (no stop forced by physics or regulation)1.00
  • Availability with the declared heat-up and changeover per run0.941 [0.827–0.982]
  • TheoreticalPhysics-bounded OEE0.979 [0.947–1.000]
  • OEE with the declared heat-up, changeovers and start-up scrap0.913 [0.794–0.975]

3Ceilings per dimension

Each dimension shows the ceilings the model computes and, where units are comparable, documented benchmarks. Colour and marker shape identify the ceiling type throughout the Oracle.
  • Theoretical
  • Best available technique
  • Best documented
  • Typical
Energy per unit of productkJ/kg · lower is better
  • TheoreticalDOE bandwidth study: zero thermodynamic minimum for most food products; this model keeps the small gelatinisation enthalpy of starch as the irreducible net change · Benchmark0 kJ/kg
  • TheoreticalThermodynamic minimum: net enthalpy change of the starch (gelatinisation)7.02 [6.21–8.05] kJ/kg
  • Best documentedLowest documented SME of expanded maize starch on a pre-pilot twin-screw extruder (20-36 % moisture; mechanical energy only, barrel heat not included, so not a total) · Benchmark250 kJ/kg
  • Best documentedDocumented specific mechanical energy of corn extrusion (lower bound of the total input)900 [905–1,095] kJ/kg
  • TypicalBAT indicative specific energy of dry pet food plants (whole plant)1,404 [1,414–1,791] kJ/kg
  • TypicalHighest documented SME of expanded maize starch on the same extruder · Benchmark2,000 kJ/kg
Yieldhigher is better
  • TheoreticalMaximum dry-matter yield (no forced stop, hence no transient losses)1.00
Qualityhigher is better
  • TheoreticalMaximum OEE quality factor (bulk-density capability)0.979 [0.947–1.000]
Performancehigher is better
  • TheoreticalPerformance factor at the physical maximum throughput1.00
Availabilityhigher is better
  • TheoreticalMaximum availability (no stop forced by physics or regulation)1.00
OEEhigher is better
  • TheoreticalPhysics-bounded OEE0.979 [0.947–1.000]

4Loss cascade

What separates each ceiling from the next one, from the physical limit to typical operation.
  1. TheoreticalBest documentedEnergy per unit of product
    7.02 [6.21–8.05] kJ/kg → 900 [905–1,095] kJ/kg (gap 893 kJ/kg)

    The only irreducible net enthalpy change between the raw grits and the cooled product is the gelatinisation and melting of the starch (about 7 kJ per kg of feed). Everything else is sensible heat: the mass is heated to about 170 °C and all of that heat leaves the process again when the steam flashes at the die and the product cools. No extruder recovers the heat of the extrudate, so the best achievable input is the adiabatic one: sensible heat of dry matter (Choi–Okos polynomial) and water (IAPWS, liquid under the die pressure) from feed to melt temperature, plus the gelatinisation enthalpy — about forty times the thermodynamic minimum. The documented specific mechanical energy of the reference corn extrusion (900–1100 kJ/kg) is about three times the adiabatic enthalpy rise, before counting barrel heating. The excess is mechanical energy converted into heat that leaves through the barrel cooling and the barrel wall, friction of the solid bed in the feed section, and the deliberate over-shearing that low moisture requires for expansion; SME also includes the mechanical degradation of starch molecules that sets texture. Twin-screw machines at higher moisture document lower SME (from 250 kJ/kg) because barrel heat supplies part of the enthalpy.

  2. Best documentedTypicalEnergy per unit of product
    900 [905–1,095] kJ/kg → 1,404 [1,414–1,791] kJ/kg (gap 504 kJ/kg)

    Whole dry pet food plants (0.39–0.50 MWh/t) add steam preconditioning, the dryer that removes the water remaining after the die flash (latent heat plus air losses), coating, cooling and conveying: the dryer alone typically exceeds the extruder.

5Physics-bounded OEE

The highest OEE the operation can reach: availability, performance and quality limited only by unavoidable stops and physical capability.
Fraction of loading time that remains after each unavoidable loss. Axis starts at 0.90.
  • Loading time (ideal)
    1.000
  • Availability
    factor 1.00
    1.000 (−0.000)
  • Performance
    factor 1.00
    1.000 (−0.000)
  • Quality
    factor 0.979 [0.947–1.000]
    0.979 (−0.021)
  • Physics-bounded OEE
    0.979 [0.947–1.000]

6Full derivation

Mass and energy balances, rate-limiting step, process capability and unavoidable stops, with every term explained.

Scope. Moistened corn grits (feed moisture xwx_w about 17 %, wet basis) are fed to a single-screw cooking extruder. Viscous dissipation and barrel heating raise the mass to the melt temperature TmT_m, the starch is gelatinised and melted, and the melt is forced through a die where the superheated water flashes to steam and expands the product. The reference geometry and operating point are those of a documented laboratory extruder: barrel diameter 19 mm, length/diameter 25, screw speed 150 rpm, barrel sections at 25, 70, 170 and 175 °C, head pressure 112–222 bar. The same equations apply to industrial single-screw snack, breakfast-cereal and pet-food extruders. Symbols and their values, units, uncertainty distributions and sources are listed in the parameter table generated from config/physics/extrusion.yaml.

1. Mass balance — maximum yield

Dry matter is conserved through the extruder; only water changes phase. Per kilogram of feed, 1−xw1 - x_w kg of dry matter and xwx_w kg of water enter. Inside the screw the water stays liquid because the melt pressure exceeds the saturation pressure at TmT_m (about 8 bar at 170 °C against more than 100 bar). At the die the pressure drops to atmospheric and the melt, superheated above the boiling point TbT_b = 100 °C, flashes water adiabatically until it reaches TbT_b:

wev=min⁡ ⁣(xw, (1−xw) Δhdry(Tb→Tm)+xw [hw(Tm,Pd)−hw,sat(Tb)]hfg(Tb)).w_{ev} = \min\!\left(x_w,\ \frac{(1-x_w)\,\Delta h_{dry}(T_b \to T_m) + x_w\,[h_w(T_m, P_d) - h_{w,sat}(T_b)]}{h_{fg}(T_b)}\right).
  • wevw_{ev} — water flashed per kg of feed (about 0.067 kg/kg with the point values).
  • Δhdry\Delta h_{dry} — enthalpy rise of the dry matter between the two temperatures (Section 2).
  • hw(Tm,Pd)h_w(T_m, P_d) — enthalpy of liquid water at the melt temperature and die pressure, and hw,sat(Tb)h_{w,sat}(T_b) that of saturated liquid at 100 °C (IAPWS-95 through CoolProp).
  • hfg(Tb)h_{fg}(T_b) — latent heat of vaporisation at 100 °C (2256 kJ/kg).

The extrudate leaves with moisture (xw−wev)/(1−wev)(x_w - w_{ev})/(1 - w_{ev}), about 11 %, and is dried further downstream. Water flashed at the die is not a yield loss: yield is counted on dry matter.

Dry matter is lost only in transients, and transients happen only when the extruder is stopped and restarted. No stop is forced by regulation or by a documented physical mechanism (Section 5), so the physical ceiling is Ymax=1Y_{max} = 1; the transient losses of the declared runs are reported as yield_scenario. At start-up the product is out of specification until melt temperature and pressure stabilise (thermal equilibrium of screw and barrel with viscous dissipation, tstabt_{stab}), and each composition change takes a number nfln_{fl} of mean residence times trest_{res} to wash out, once at start-up and once at shut-down:

Yscen=1−tstab+2 nfl trestrun,tres=ρm πDHL ffillm˙max.Y_{scen} = 1 - \frac{t_{stab} + 2\,n_{fl}\,t_{res}}{t_{run}}, \qquad t_{res} = \frac{\rho_m\,\pi D H L\,f_{fill}}{\dot m_{max}} .
  • trunt_{run} — run length between changeovers; ffillf_{fill} — average degree of fill of the screw channel; ρm\rho_m — melt density; DD, HH, LL — screw diameter, channel depth and length; m˙max\dot m_{max} — the maximum throughput of Section 3.
  • For a well-mixed vessel three residence times leave 5 % (e−3e^{-3}) of the old material; a screw is closer to plug flow, so nfln_{fl} = 3–5 is conservative.

2. Energy balance — minimum energy per unit of product

Thermodynamic minimum. Compare the raw grits with the cooled, dried product at the same temperature. The water that flashes and is dried off leaves as vapour, but its enthalpy can in principle be recovered by condensation; the sensible heat given to the mass returns when the product cools. The only irreducible net enthalpy change is the conversion of the starch — gelatinisation and melting of the crystalline order of the granules, an endothermic transition:

emin=(1−xw) fs Δhgel.e_{min} = (1 - x_w)\, f_s\, \Delta h_{gel} .
  • fsf_s — starch fraction of the dry matter (declared 0.75–0.88 for degermed grits).
  • Δhgel\Delta h_{gel} — gelatinisation enthalpy of normal maize starch by differential scanning calorimetry, 10.57 J/g of starch.

With the point values emin≈7e_{min} \approx 7 kJ/kg. The US DOE bandwidth study assigns a zero thermodynamic minimum to most food products; the gelatinisation enthalpy is the small correction this model keeps.

Adiabatic bound (minimum without heat recovery). No extruder recovers the heat of the extrudate: the steam flashes to the atmosphere and the product cools in the dryer and cooler. The least energy an extruder can then put into the mass is the enthalpy rise from feed to die, with no heat lost through the barrel:

ead=(1−xw) Δhdry(T0→Tm)+xw [hw(Tm,Pd)−hw(T0,Pd)]+emin,e_{ad} = (1 - x_w)\,\Delta h_{dry}(T_0 \to T_m) + x_w\,[h_w(T_m, P_d) - h_w(T_0, P_d)] + e_{min}, Δhdry(T1→T2)=∫T1T2cp,dry dT=a0(t2−t1)+a12(t22−t12)+a23(t23−t13).\Delta h_{dry}(T_1 \to T_2) = \int_{T_1}^{T_2} c_{p,dry}\,dT = a_0 (t_2 - t_1) + \frac{a_1}{2}(t_2^2 - t_1^2) + \frac{a_2}{3}(t_2^3 - t_1^3).
  • T0T_0 — feed temperature; tt — temperature in °C.
  • a0a_0, a1a_1, a2a_2 — coefficients of the Choi–Okos specific-heat polynomial for carbohydrate (1.5488 kJ/(kg·K), 1.9625×10⁻³ and −5.9399×10⁻⁶ per °C and °C²), from the open CoolProp source file that reproduces the 2006 ASHRAE tables. Treating all dry matter as carbohydrate slightly underestimates the specific heat (protein, fat and fibre have higher values), which makes the bound favourable.
  • Water enthalpies come from IAPWS-95 at the die pressure (liquid).

With the point values the sensible heat is about 305 kJ/kg and ead≈312e_{ad} \approx 312 kJ/kg — about forty times the thermodynamic minimum. The whole difference is heat that the process gives to the mass and then throws away at the die and in the cooler.

Documented and typical. The documented specific mechanical energy (SME, motor work per kg) of the reference corn extrusion is 900–1100 kJ/kg, about three times the adiabatic bound, before counting barrel heating. The excess leaves as heat through the barrel wall and cooling, friction of the solid bed in the feed section, and the deliberate over-shearing that low moisture requires for expansion (SME also degrades starch molecules, which sets the texture). On twin-screw machines at higher moisture (20–36 %) documented SME of expanded maize starch spans 250–2000 kJ/kg; the low end is not a total because barrel heat supplies part of the enthalpy, and it is shown as a benchmark with that caveat. Whole dry pet-food plants, with steam preconditioning, drying, coating and cooling, have BAT indicative levels of 0.39–0.50 MWh/t (1400–1800 kJ/kg), the typical ceiling: the dryer alone usually exceeds the extruder.

3. Rate-limiting step — minimum residence time and maximum throughput

Drag flow. Unrolling the screw channel gives flow between two plates, one (the barrel) moving relative to the other at the peripheral speed πDN\pi D N. The component of that velocity along the channel, πDNcos⁡φ\pi D N \cos\varphi, drags a Couette flow of mean velocity half the wall velocity through a channel of width πDsin⁡φ\pi D \sin\varphi (zero flight width) and depth HH:

Qd=12 π2D2NHsin⁡φcos⁡φ.Q_d = \tfrac{1}{2}\,\pi^2 D^2 N H \sin\varphi \cos\varphi .
  • NN — screw speed in revolutions per second; φ\varphi — helix angle, 17.66° for a square-pitch screw (tan⁡φ=1/π\tan\varphi = 1/\pi); HH — channel depth of the metering section (declared 1.0–2.0 mm for a 19 mm screw with 3:1 compression).

The die resistance builds a pressure gradient that drives a back-flow (pressure flow) opposing the drag flow; the open extrusion-simulation literature writes the throughput as exactly these two terms, drag minus pressure flow. Without a viscosity model of the melt, the back-flow is a declared fraction aa of the drag flow:

m˙drag=ρm Qd (1−a).\dot m_{drag} = \rho_m\,Q_d\,(1 - a).

Cooking residence time. Starch must stay long enough at cooking temperature to convert completely. The melt-filled hot zone has the channel volume Vhot=πDHLfhotV_{hot} = \pi D H L f_{hot} (channel width × depth × helical length, πDsin⁡φ⋅H⋅L/sin⁡φ\pi D \sin\varphi \cdot H \cdot L/\sin\varphi), where fhotf_{hot} is the fraction of the screw at cooking temperature (two of four barrel sections in the reference set-up). A throughput m˙\dot m gives a mean hot-zone residence ρmVhot/m˙\rho_m V_{hot}/\dot m, which must be at least tcookt_{cook}:

m˙cook=ρmVhottcook,m˙max=min⁡(m˙drag,m˙cook).\dot m_{cook} = \frac{\rho_m V_{hot}}{t_{cook}}, \qquad \dot m_{max} = \min(\dot m_{drag}, \dot m_{cook}).

No open source with rate constants of starch conversion under extrusion shear was found, so tcookt_{cook} = 10–40 s is a declared scenario. With the point values the drag capacity is about 6.8 kg/h and the cooking capacity about 5.0 kg/h, so the residence time limits: m˙max≈5\dot m_{max} \approx 5 kg/h. The documented feed rate of the reference extruder, 3.4 kg/h, is below both limits — the machine is starve-fed and set for expansion quality. Because m˙max\dot m_{max} and the minimum residence time follow from the declared cooking time and back-flow ratio, they are scenario values passed through a formula and are reported without a ceiling label. An extruder running at m˙max\dot m_{max} has an OEE performance factor of 1.

4. Process capability — quality ceiling

The expansion specification is expressed through bulk density. The documented bulk density of corn extrudates is 0.092 g/cm³ with a standard deviation of 0.008 g/cm³ (15 measurements, laboratory; a better-controlled line is declared down to 0.004). With a centred target and a two-sided specification of relative half-width δ\delta,

k=δσρ/ρˉb,dρ=2 Φ(−k),Qmax=1−dρ,Qscen=(1−dρ) Yscen.k = \frac{\delta}{\sigma_\rho/\bar\rho_b}, \qquad d_\rho = 2\,\Phi(-k), \qquad Q_{max} = 1 - d_\rho, \qquad Q_{scen} = (1 - d_\rho)\,Y_{scen}.
  • σρ/ρˉb\sigma_\rho/\bar\rho_b — coefficient of variation of bulk density; Φ\Phi — standard normal distribution function; dρd_\rho — share of product out of specification.

Start-up and shut-down scrap is unsaleable material made during run time, so it belongs to the quality factor (as in pasteurization and evaporation); it exists only with the declared stops and therefore enters QscenQ_{scen}, not QmaxQ_{max}.

5. Unavoidable stops — maximum availability

No stop of the extruder is forced by regulation or by a documented physical mechanism: an extruder can in principle run without interruption, so Amax=1A_{max} = 1. The stops of the scenario follow the declared run length between product changeovers: each run needs the barrel to be heated before feeding (theatt_{heat}; feeding a cold barrel blocks the screw with unmelted material) and, at the end, a die and screw changeover with purge and cleaning (tchgt_{chg}; starch left in a stopped barrel carbonises). The extruder is purged and cleaned by hand, not in place, so the cleaning model does not apply and the durations are declared:

Ascen=truntrun+theat+tchg=0.941.A_{scen} = \frac{t_{run}}{t_{run} + t_{heat} + t_{chg}} = 0.941 .

Wear-driven maintenance of screw and die is outside the model.

6. Physics-bounded OEE

Scope of the bound. The theoretical label of OEEmaxOEE_{max} means physics-bounded under the declared scenario: the ceiling follows from the physics of the operation for the declared product, equipment and schedule, and it is a bound only conditional on those declarations. It follows the common definition of all models: AmaxA_{max} counts only stops required by regulation or forced by a physical mechanism, at their physical minimum duration, plus the start-up they force; PmaxP_{max} is the ratio of the physics-limited ideal cycle to the cycle of the capable process; QmaxQ_{max} is the share of conforming units, and mass losses of conforming product are reported only as yield.

OEEmax=Amax×Pmax×Qmax,Pmax=1,OEEscen=Ascen Pmax Qscen.OEE_{max} = A_{max} \times P_{max} \times Q_{max}, \qquad P_{max} = 1, \qquad OEE_{scen} = A_{scen}\,P_{max}\,Q_{scen} .

With the point values OEEmaxOEE_{max} = 0.979, set by the bulk-density capability, and oee_scenario = 0.913 with the declared stops and start-up scrap. The Monte Carlo intervals are reported in the output table.

7. Assumptions and validity

See the assumption and validity lists generated from the configuration file. The channel depth, the hot-zone fraction and the cooking time dominate the maximum throughput; melt temperature and feed moisture dominate the energy bound; the run length dominates availability.

References

  • Effect of Medicago sativa addition on physicochemical, nutritional and functional characteristics of corn extrudates (2021). Foods, 10(5), 928. doi:10.3390/foods10050928
  • Extrusion simulation for the design of cereal and legume foods (2022). Foods, 11(12), 1780. doi:10.3390/foods11121780
  • Modulating the physicochemical properties and internal structure of maize starches with differing amylose contents via non-covalent interaction with tea polyphenols (2026). Foods, 15(4), 766. doi:10.3390/foods15040766
  • CoolProp open-source project. Source file dev/incompressible_liquids/CPIncomp/CoefficientFluids.py (food component polynomials based on Choi and Okos, 1986, as reproduced in the 2006 ASHRAE Handbook).
  • European Commission, Joint Research Centre. Best Available Techniques (BAT) Reference Document for the Food, Drink and Milk Industries, 2019.
  • U.S. Department of Energy. Bandwidth Study on Energy Use and Potential Energy Savings Opportunities in U.S. Food and Beverage Manufacturing, 2017.
  • Bell, I. H., Wronski, J., Quoilin, S., Lemort, V. (2014). Pure and pseudo-pure fluid thermophysical property evaluation and the open-source thermophysical property library CoolProp. Industrial & Engineering Chemistry Research, 53, 2498–2508. doi:10.1021/ie4033999

7Parameters and provenance

Every input with its value, unit, role, uncertainty distribution and provenance. Open a quote to read the passage of the source that states the value.

8Benchmarks

Documented values from open sources used to check the ceilings; no documented value may beat the theoretical limit.
BenchmarkCeilingValueProvenance
Lowest documented SME of expanded maize starch on a pre-pilot twin-screw extruder (20-36 % moisture; mechanical energy only, barrel heat not included, so not a total)
Show verbatim quote
SME (kJ/kg) 250 < SME < 2000 350 < SME < 1250 850 < SME < 5500 confidentialExtrusion Simulation for the Design of Cereal and Legume Foods (Foods 11(12): 1780) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11121780; PMC9222340
Best documented250 kJ/kgVerified quote
Highest documented SME of expanded maize starch on the same extruder
Show verbatim quote
SME (kJ/kg) 250 < SME < 2000 350 < SME < 1250 850 < SME < 5500 confidentialExtrusion Simulation for the Design of Cereal and Legume Foods (Foods 11(12): 1780) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11121780; PMC9222340
Typical2,000 kJ/kgVerified quote
BAT indicative specific energy of dry pet food plants, upper end (whole plant including drying)
Show verbatim quote
Dry pet food 0.39–0.50Best Available Techniques (BAT) Reference Document for the Food, Drink and Milk Industries (opens in a new tab). European Commission, Joint Research Centre; Publications Office of the European Union, 2019. p. 714. EUR 29978 EN; JRC118627; doi:10.2760/243911
Typical0.5 MWh/tonneVerified quote
DOE bandwidth study: zero thermodynamic minimum for most food products; this model keeps the small gelatinisation enthalpy of starch as the irreducible net change
Show verbatim quote
Most food and beverage products have a zero TM energy intensity because there are no significant chemical reactions involved.Bandwidth Study on Energy Use and Potential Energy Savings Opportunities in U.S. Food and Beverage Manufacturing (opens in a new tab). U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Advanced Manufacturing Office, 2017. p. 41. DOE/EE-1571; OSTI ID 1512366
Theoretical0 Btu/lbVerified quote
Feed rate documented on the reference 19 mm extruder
Show verbatim quote
loaded with prepared corn samples at a constant dosing speed of 18 rpm (feed rate range, 3.4 kg/h)Effect of Medicago sativa Addition on Physicochemical, Nutritional and Functional Characteristics of Corn Extrudates (Foods 10(5): 928) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2021. doi:10.3390/foods10050928; PMC8145381
Best documented3.4 kg/hVerified quote

9Unavoidable stops

Stops that physics or regulation impose; they set the maximum availability.
  • Barrel heat-up before feeding (scenario stop: availability_scenario only)Declared assumption

    Feeding a cold barrel would block the screw with unmelted material; the barrel must first reach its set points. Declared scenario.

  • Die and screw changeover with purge and cleaning (scenario stop: availability_scenario only)Declared assumption

    A new product shape needs a new die, and starch residues carbonise if left in the barrel; the screw is purged and the die cleaned at the end of every run. Declared scenario.

10Assumptions

  1. No stop is forced by regulation or by a documented physical mechanism: barrel heat-up, end-of-run purge and cleaning and the die change follow the declared run length between product changeovers, so availability_max = 1 and those stops, with the start-up and wash-out scrap they cause, enter only availability_scenario, quality_scenario, yield_scenario and oee_scenario. Wear-driven maintenance of screw and die is outside the model. The extruder is cleaned by purging and by hand, not in place, so the cleaning model does not apply. The throughput and residence-time values depend on the declared cooking time and back-flow ratio: they are scenario values passed through a formula and carry no ceiling label; the adiabatic energy is a computed no-recovery value, not a documented BAT.
  2. The dry matter is treated as carbohydrate for its specific heat (Choi–Okos polynomial integrated from feed to melt temperature); protein, fat and fibre have higher specific heats, so the adiabatic bound is slightly low (favourable).
  3. Water stays liquid in the screw (die pressure above saturation) and its enthalpy comes from IAPWS-95 through CoolProp; at the die the superheat of the melt above 100 °C flashes water adiabatically.
  4. Gelatinisation enthalpy measured in excess water is used for the limited-water melt; at low moisture the endotherm shifts to higher temperature but its total magnitude stays of the same order.
  5. Drag flow of a single channel between parallel plates with zero flight width (Q_d = ½ π² D² N H sin φ cos φ); the die back-flow is a declared fraction of the drag flow.
  6. Start-up and shut-down scrap is the stabilisation time plus a number of mean residence times to wash out each transition; it is counted in the quality factor, as in the other models.
  7. Bulk density is normally distributed around a centred target; product outside the two-sided specification is out of specification.

11Range of validity

  • Direct-expanded starch-based products from single-screw cooking extruders at 15–20 % feed moisture and melt temperatures of 160–175 °C.
  • Twin-screw extruders: the drag-flow expression is an approximation (co-rotating screws convey positively); the energy and residence-time bounds still apply.

12Sensitivity

Elasticity: relative change of the output for a relative change of each parameter. Spearman rank correlation from the Monte Carlo draws is shown alongside.
  • ρˉb\bar\rho_bMean bulk density of corn extrudates
    elasticity +0.134Spearman ρ —
  • δ\deltaHalf-width of the bulk-density specification, relative to the target
    elasticity +0.134Spearman ρ +0.80
  • σρ\sigma_{\rho}Standard deviation of the bulk density of corn extrudates (15 measurements)
    elasticity -0.133Spearman ρ -0.55
  • Output rises with the parameter
  • Output falls with the parameter

13Validation checks

A model is published only when every check passes.
  • Dimensional analysis24 outputs converted to their declared unitsPassed
  • Every cited value verified against its source18 verified, 0 failedPassed
  • No documented value beats the theoretical limitPassed
  • Ceilings in physical orderPassed
  • Fractions within 0 and 1Passed

14Causes of loss in this operation

Probability that each cause is documented as a relevant loss of a line, pooled over all segments.

15References

  1. Effect of Medicago sativa Addition on Physicochemical, Nutritional and Functional Characteristics of Corn Extrudates (Foods 10(5): 928) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2021. doi:10.3390/foods10050928; PMC8145381
  2. Extrusion Simulation for the Design of Cereal and Legume Foods (Foods 11(12): 1780) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11121780; PMC9222340
  3. Modulating the Physicochemical Properties and Internal Structure of Maize Starches with Differing Amylose Contents via Non-Covalent Interaction with Tea Polyphenols (Foods 15(4): 766) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2026. doi:10.3390/foods15040766; PMC12940761
  4. CoolProp incompressible-liquid source file CoefficientFluids.py (food component polynomials of the 2006 ASHRAE Handbook based on Choi and Okos, 1986) (opens in a new tab). CoolProp open-source project (MIT licence), 2024. CoolProp repository file dev/incompressible_liquids/CPIncomp/CoefficientFluids.py
  5. Best Available Techniques (BAT) Reference Document for the Food, Drink and Milk Industries (opens in a new tab). European Commission, Joint Research Centre; Publications Office of the European Union, 2019. p. 714. EUR 29978 EN; JRC118627; doi:10.2760/243911
  6. Bandwidth Study on Energy Use and Potential Energy Savings Opportunities in U.S. Food and Beverage Manufacturing (opens in a new tab). U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Advanced Manufacturing Office, 2017. p. 41. DOE/EE-1571; OSTI ID 1512366
  7. doi:10.3390/foods10050928 — Reference single-screw corn extrusion: geometry, operating point, SME and bulk density.
  8. doi:10.3390/foods11121780 — Drag and pressure flow in the 1D extrusion model; documented ranges of SME, moisture and melt temperature.
  9. doi:10.3390/foods15040766 — Gelatinisation enthalpy of normal maize starch.
  10. doi:10.1021/ie4033999 — Thermophysical property library (IAPWS-95 water).