Fermentation
Batch alcoholic fermentation (primary fermentation of ale wort in a cylindro-conical vessel)
In short
Fermentation can use at most 74 of every 100 planned hours. The rest goes to limits no plant can avoid.
- 74usable at most
- 26are a physical limit
The three limits that set it
Running time: up to 99 %
The rest are stops imposed by physics or by regulation: cleaning, changeovers, defrosting.
Speed: up to 75 %
How close to its maximum speed it can run without damaging the product or the equipment.
Good product: up to 100 %
Some product falls outside specification even with the process well under control.
Technical detail
Uncertainty intervals, physical models per station and sources, for anyone who wants to check each figure.
1Scenario
Cooled, aerated ale wort is pitched with yeast and fermented isothermally at about 21 °C in a cylindro-conical vessel until the fermentable sugar reaches its attenuation limit; the fermentation heat is removed through the cooling jacket. The rate-limiting step is yeast growth and sugar uptake, described with the biomass model and parameters estimated by Zamudio Lara et al. (2022) for ale fermentations at 19–28 °C. After each batch the vessel is emptied, the yeast is cropped and the vessel is cleaned in place. Lager fermentation below about 16 °C is outside the kinetic law; bread-dough proofing uses the same ceiling structure only as an approximation (see the derivation, section 7).
2Results
- Maximum specific growth rate of the yeast at the fermentation temperature0.118 [0.080–0.152] 1/h
- Sugar fermented to the completion criterion65.9 [57.2–74.0] g/L
- Suspended yeast biomass at the end of the primary fermentation3.88 [3.04–4.75] g/L
- Ethanol content of the green beer (documented yield)27.2 [21.5–34.0] g/L
- TheoreticalMaximum ethanol yield on fermented sugar (Gay-Lussac stoichiometry)0.511
- Best documentedEthanol yield on fermented sugar of the documented kinetics (k_E/k_S)0.412 [0.345–0.496]
- Documented ethanol yield as a share of the stoichiometric ceiling0.806 [0.675–0.970]
- Fermentation heat to be removed per kilogram of wort36.0 [24.3–44.3] kJ/kg
- TheoreticalThermodynamic minimum: Carnot work to reject the fermentation heat to the ambient0.489 [0.000–1.735] kJ/kg
- TheoreticalMinimum primary-fermentation time (growth kinetics at the declared temperature and pitching rate)29.0 [20.8–46.3] h
- Scheduled fermentation time of the capable process (batch at mu_max exp(-z sigma_k))41.3 [27.0–73.4] h
- Ideal batch cycle: fill, minimum fermentation, crop and transfer36.0 [28.2–55.8] h
- Batch cycle of the capable process48.3 [35.1–82.5] h
- Batches per vessel and day at the ideal cycle plus cleaning0.660 [0.427–0.838] 1/d
- Share of batches that miss the scheduled time0.00135 [0.00026–0.00579]
- TheoreticalMaximum availability (cleaning after each batch as the only stop)0.9923 [0.9879–0.9958]
- TheoreticalMaximum performance factor (ideal cycle / scheduled cycle)0.746 [0.615–0.875]
- TheoreticalMaximum OEE quality factor (batches within the schedule)0.9987 [0.9942–0.9997]
- TheoreticalPhysics-bounded OEE0.740 [0.611–0.865]
3Ceilings per dimension
- Theoretical
- Best available technique
- Best documented
- Typical
- TheoreticalMaximum ethanol yield on fermented sugar (Gay-Lussac stoichiometry)0.511
- Best documentedEthanol yield on fermented sugar of the documented kinetics (k_E/k_S)0.412 [0.345–0.496]
- TheoreticalThermodynamic minimum: Carnot work to reject the fermentation heat to the ambient0.489 [0.000–1.735] kJ/kg
- TheoreticalMinimum primary-fermentation time (growth kinetics at the declared temperature and pitching rate)29.0 [20.8–46.3] h
- TheoreticalMaximum availability (cleaning after each batch as the only stop)0.9923 [0.9879–0.9958]
- TheoreticalMaximum performance factor (ideal cycle / scheduled cycle)0.746 [0.615–0.875]
- TheoreticalMaximum OEE quality factor (batches within the schedule)0.9987 [0.9942–0.9997]
- TheoreticalPhysics-bounded OEE0.740 [0.611–0.865]
4Loss cascade
- TheoreticalBest available techniqueEnergy per unit of product
Fermentation is exothermic (DOE: negative thermodynamic minimum). The only unavoidable energy is the work to pump the fermentation heat from the fermentation temperature to the ambient, the Carnot work, which is zero when the ambient is colder. The DOE state-of-the-art refrigeration intensity is a whole-brewery value: it includes wort cooling, maturation below 0 °C, cold storage and the irreversibilities of real refrigeration cycles (COP far below Carnot).
- Best available techniqueTypicalEnergy per unit of product
Between DOE state of the art and current typical refrigeration (15 986 and 20 302 Btu/bbl) lie compressor and heat-exchanger efficiency, insulation of outdoor tanks and the cooling-time gains of forced circulation documented in the BAT reference document (up to 30 % shorter processing and cooling times).
- TheoreticalBest documentedYield0.511 → 0.412 [0.345–0.496] (gap 0.0990)
The Gay-Lussac stoichiometry converts at most 0.511 g of ethanol per gram of hexose. Yeast must build biomass and maintain itself; in the documented kinetics each gram of new biomass costs 15.3 g of sugar and yields 6.31 g of ethanol, an ethanol yield of 0.41 g/g, about 81 % of the stoichiometric limit. Glycerol and other by-products take the rest.
- TheoreticalTypicalCycle or residence time
The minimum fermentation time follows from the growth kinetics at the declared temperature and pitching rate; a schedule must cover the slow batches, so the scheduled time is the time of a batch whose growth rate is z standard deviations below the mean. Lower temperatures (lagers), lower pitching rates, poorer yeast vitality and the diacetyl rest, which this model does not include, lengthen the time further.
5Physics-bounded OEE
- Loading time (ideal)1.000
- Availabilityfactor 0.9923 [0.9879–0.9958]0.992 (−0.008)
- Performancefactor 0.746 [0.615–0.875]0.741 (−0.252)
- Qualityfactor 0.9987 [0.9942–0.9997]0.740 (−0.001)
- Physics-bounded OEE0.740 [0.611–0.865]
6Full derivation
Scope. Cooled, aerated ale wort is pitched with yeast and fermented isothermally at about 21 °C in a cylindro-conical vessel until the fermentable sugar reaches its attenuation limit; the fermentation heat is removed through the cooling jacket. After each batch the vessel is emptied, the yeast is cropped and the vessel is cleaned in place. Section 7 states how the same ceiling is applied, as an approximation, to lager fermentation and to bread-dough proofing. Symbols and their values, units, uncertainty distributions and sources are listed in the parameter table generated from config/physics/fermentation.yaml.
1. Mass balance — maximum ethanol yield
The alcoholic fermentation of a hexose is, in its overall stoichiometry (Gay-Lussac),
With standard atomic weights the molar masses are = 180.156 g/mol and = 46.069 g/mol, so no fermentation can make more than
(sugars counted as hexose equivalents). Yeast, however, must build biomass and maintain itself. In the biomass model used below each gram of new biomass consumes = 15.3 g of sugar and yields = 6.31 g of ethanol, so the documented yield is
about 81 % of the stoichiometric ceiling; glycerol, higher alcohols and other by-products take part of the rest. The sugar fermented to the completion criterion is , where is the sugar the yeast cannot use (attenuation limit) and the completion tolerance. The CO₂ and the cropped yeast leave the vessel by design; the ethanol yield, not the beer mass, is the meaningful yield of this operation, and it stays outside the OEE.
2. Energy balance — minimum energy per unit of product
The reaction is exothermic. The heat released by anaerobic fermentation, measured by calorimetry in an insulated bioreactor, is = 101.28 kJ per mole of glucose (the literature range reported in the same source is 71.5–119.2 kJ/mol, used as the uncertainty range). Per kilogram of wort of density ,
about 36 kJ/kg. To keep the vessel isothermal this heat must leave at the fermentation temperature . If the ambient (heat sink) is colder, it can leave without work; if it is warmer, the least work is that of a reversible (Carnot) refrigerator:
With = 21 °C and = 25 °C this is about 0.5 kJ/kg. The DOE bandwidth study gives a negative thermodynamic minimum for the brewhouse step of beer for the same reason (the conversion releases energy). Real refrigeration plants need several times the Carnot work (finite temperature differences, compressor losses), and the documented brewery refrigeration intensities of the DOE study include wort cooling, maturation below 0 °C and cold storage; they are listed as benchmarks per barrel of beer and are not comparable with the per-kilogram outputs.
3. Rate-limiting step — yeast growth and sugar uptake
The rate-limiting step is biological: sugar is consumed only as fast as the yeast grows. The model is the biomass model of Zamudio Lara et al. (2022), estimated on isothermal 17 L ale fermentations at 19, 21 and 28 °C:
- , — suspended biomass and sugar (g/L); = 0.705 g/L is the pitching rate of the source experiments.
- — specific growth rate; the Droop-type factor stops growth at the minimum sugar quota = 13.1 g/L.
- — settling of the yeast (0.0167 h⁻¹).
- = 0.400 h⁻¹, = −1.1 h⁻¹ — temperature law of the maximum growth rate (about 0.118 h⁻¹ at 21 °C). The law vanishes at 15.6 °C, the lower limit of its validity.
The system is integrated with a fourth-order Runge–Kutta scheme (0.25 h step; halving the step changes the time by less than 0.1 %) until . Without settling and without a sugar quota it reduces to , and the time , which the tests use as a check. The result is the minimum primary-fermentation time at the declared temperature and pitching rate: about 29 h at 21 °C. The ideal batch cycle adds the transfer steps (filling with several brews, yeast cropping and transfer of the green beer), which belong to the ideal cycle under the common batch convention:
Each vessel can therefore run at most batches per day, reported without a ceiling label because the transfer times are declared. Higher temperature or more yeast shortens the fermentation; the recipe, not physics, limits both (flavour by-products at high temperature, cost and autolysis at high pitching).
4. Process capability — quality ceiling and the performance it costs
Batches differ: yeast vitality and wort composition change the growth rate from batch to batch. With a log-normal spread of , a schedule that must hold for all but of the batches is the fermentation time of a batch growing at , computed by integrating the same kinetics:
Batches slower than the schedule miss it (a quality loss in this definition; in practice they are held longer). The performance factor is the price of planning for slow batches, the same construction as the capable sterilization process.
5. Unavoidable stops — maximum availability
The vessel has to be cleaned after each fermentation to remove yeast, trub and protein deposits, and disinfected before the next batch. The duration is taken through a parameter link from the cleaning model (output cip_time_min, with its Monte Carlo interval). That model is valid for recirculated pipe circuits; the spray cleaning of a tank, including the displacement of CO₂ that would otherwise neutralise caustic detergents, is outside its validity, so the linked duration is a lower bound. One cleaning per scheduled cycle gives
6. Physics-bounded OEE
Scope of the bound. The theoretical label of 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: counts only stops required by regulation or forced by a physical mechanism, at their physical minimum duration, plus the start-up they force; is the ratio of the physics-limited ideal cycle to the cycle of the capable process; is the share of conforming units, and mass losses of conforming product are reported only as yield.
The result, with the Monte Carlo interval over the declared uncertainty of every input, is reported in the output table. It is dominated by the performance factor: the batch-to-batch spread of the growth rate and the declared margin fix how much longer than the minimum the schedule must be. The fermentation temperature dominates the minimum time.
7. Assumptions, validity and application to other processes
See the assumption and validity lists generated from the configuration file. The diacetyl rest that many breweries add at the end of the primary fermentation is not modelled; it lengthens the real cycle and belongs to the gap between the theoretical and the typical time.
The codebook assigns the operation fermentation to two processes:
- Beer, primary fermentation: the model's process. For lagers (8–15 °C) the kinetic law of this model is outside its validity (it predicts no growth below about 15.6 °C); the mass balance, the energy balance, the cleaning stop and the capability construction carry over, and a lager growth law would replace Section 3.
- Bread, proofing: the yeast ferments the dough sugars at 30–35 °C to produce the CO₂ that raises the dough; ethanol is a by-product that leaves in the oven. The same structure applies as an approximation — growth-limited CO₂ production sets the minimum proofing time, the batch-to-batch spread of yeast activity sets the scheduled time and therefore , and over- or under-proofed dough is the quality loss — but the numbers of this model (wort kinetics, isothermal vessel, cleaning in place) do not; proofers are cleaned dry and their availability belongs to the baking line.
References
- Zamudio Lara, J. M., Dewasme, L., Hernández Escoto, H., Vande Wouwer, A. (2022). Parameter estimation of dynamic beer fermentation models. Foods, 11(22), 3602. doi:10.3390/foods11223602
- González-Hernández, Y., Michiels, E., Perré, P. (2024). Heat of reaction in individual metabolic pathways of yeast determined by mechanistic modeling in an insulated bioreactor. Biotechnology for Biofuels and Bioproducts, 17, 137. doi:10.1186/s13068-024-02580-8
- 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.
7Parameters and provenance
- 21 °C
Fermentation temperature (isothermal)
Scenario · Uniform 19 to 23 °C
Declared assumptionDeclared ale temperature inside the range of the kinetic data (19, 21 and 28 °C in the source; 'ale beer fermentation ... at a temperature ranging from 17 to 26 °C'); brewing temperatures are set by the recipe (bref_fdm_2019: 'usually in the range of 8 °C to 30 °C').
- 0.4 1/h
Slope of the maximum specific growth rate against ln T (T in °C), biomass model
Property · Fixed
Verified quoteShow verbatim quote
a h − 1 4.00 × 10 − 1 3 b h − 1 −1.1 5Parameter Estimation of Dynamic Beer Fermentation Models (Zamudio Lara, Dewasme, Hernández Escoto, Vande Wouwer; Foods 11(22): 3602) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11223602; PMC9689312
- -1.1 1/h
Intercept of the maximum specific growth rate law
Property · Fixed
Verified quoteShow verbatim quote
a h − 1 4.00 × 10 − 1 3 b h − 1 −1.1 5Parameter Estimation of Dynamic Beer Fermentation Models (Zamudio Lara, Dewasme, Hernández Escoto, Vande Wouwer; Foods 11(22): 3602) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11223602; PMC9689312
- 15.3
Sugar consumed per unit of biomass formed
Property · Normal, sd 1.07
Verified quoteShow verbatim quote
k S gS/gX 15.3 7 k E gE/gX 6.31 6 S m i n g/L 13.1 11 g x h − 1 1.67 × 10 − 2 7Parameter Estimation of Dynamic Beer Fermentation Models (Zamudio Lara, Dewasme, Hernández Escoto, Vande Wouwer; Foods 11(22): 3602) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11223602; PMC9689312
- 6.31
Ethanol formed per unit of biomass formed
Property · Normal, sd 0.38
Verified quoteShow verbatim quote
k S gS/gX 15.3 7 k E gE/gX 6.31 6 S m i n g/L 13.1 11 g x h − 1 1.67 × 10 − 2 7Parameter Estimation of Dynamic Beer Fermentation Models (Zamudio Lara, Dewasme, Hernández Escoto, Vande Wouwer; Foods 11(22): 3602) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11223602; PMC9689312
- 13.1 g/L
Minimum sugar quota below which the yeast stops growing (attenuation limit of the wort)
Property · Normal, sd 1.44 g/L
Verified quoteShow verbatim quote
k S gS/gX 15.3 7 k E gE/gX 6.31 6 S m i n g/L 13.1 11 g x h − 1 1.67 × 10 − 2 7Parameter Estimation of Dynamic Beer Fermentation Models (Zamudio Lara, Dewasme, Hernández Escoto, Vande Wouwer; Foods 11(22): 3602) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11223602; PMC9689312
- 0.0167 1/h
Settling coefficient of the yeast (loss of suspended biomass)
Property · Normal, sd 0.0012 1/h
Verified quoteShow verbatim quote
k S gS/gX 15.3 7 k E gE/gX 6.31 6 S m i n g/L 13.1 11 g x h − 1 1.67 × 10 − 2 7Parameter Estimation of Dynamic Beer Fermentation Models (Zamudio Lara, Dewasme, Hernández Escoto, Vande Wouwer; Foods 11(22): 3602) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11223602; PMC9689312
- 0.705 g/L
Pitching rate (initial yeast biomass)
Technology · Fixed
Verified quoteShow verbatim quote
1 72 19 88 7.05 × 10 − 1 2 54 19 72 5.29 × 10 − 1 3 96 21 75 7.05 × 10 − 1 4 72 28 81 7.05 × 10 − 1Parameter Estimation of Dynamic Beer Fermentation Models (Zamudio Lara, Dewasme, Hernández Escoto, Vande Wouwer; Foods 11(22): 3602) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11223602; PMC9689312
- 80 g/L
Initial sugar concentration of the wort
Scenario · Uniform 72 to 88 g/L
Declared assumptionCentre of the range of the kinetic experiments (Table 2 of epmc_beer_fermentation_models_2022: 72, 75, 81 and 88 g/L).
- 1 g/L
Completion criterion: residual sugar above the attenuation limit at which the primary fermentation is finished
Scenario · Uniform 0.5 to 2 g/L
Declared assumptionDeclared specification; a smaller tolerance lengthens the tail of the fermentation logarithmically.
- 0.1
Batch-to-batch log standard deviation of the yeast's maximum growth rate (vitality, wort composition)
Scenario · Uniform 0.05 to 0.15
Declared assumptionProcess-capability assumption; replace with the spread of the plant's fermentation curves.
- 3
Margin of the scheduled fermentation time, in standard deviations of the growth rate
Scenario · Uniform 2.5 to 3.5
Declared assumptionOperating choice; the point value 3 is the centre of the declared range 2.5-3.5.
- 101.28 kJ/mol
Heat released by anaerobic fermentation per mole of glucose
Property · Uniform 71.5 to 119.2 kJ/mol
Verified quoteShow verbatim quote
-101.28 ±0.02kJ/mol for anaerobic fermentationHeat of reaction in individual metabolic pathways of yeast determined by mechanistic modeling in an insulated bioreactor (González-Hernández, Michiels, Perré; Biotechnology for Biofuels and Bioproducts 17: 137) (opens in a new tab). BioMed Central / Springer Nature (open access, CC BY 4.0); full text served by Europe PMC, 2024. doi:10.1186/s13068-024-02580-8; PMC11587622
- 1,030 kg/m³
Density of the wort
Scenario · Uniform 1,025 to 1,035 kg/m³
Declared assumptionDeclared for a wort of about 8 % fermentable sugar; it only converts concentrations to mass shares.
- 25 °C
Ambient (heat-rejection) temperature for the ideal refrigeration work
Scenario · Uniform 15 to 35 °C
Declared assumption - 4 h
Time to fill the vessel with wort (several brews)
Scenario · Uniform 2 to 8 h
Declared assumptionBatch transfer step, part of the ideal cycle (common batch convention).
- 3 h
Time to crop the yeast and transfer the green beer
Scenario · Uniform 2 to 6 h
Declared assumptionBatch transfer step, part of the ideal cycle (common batch convention).
- 0.374294 h
Minimum duration of a cleaning-in-place programme (output cip_time_min of the cleaning model, with its Monte Carlo interval)
Property · Triangular 0.242184, 0.374294, 0.536161 h
ComputedOutput `cip_time_min` of the cleaning model: 22.46 min [14.53, 32.17].
8Benchmarks
| Benchmark | Ceiling | Value | Provenance |
|---|---|---|---|
DOE bandwidth study, beer, refrigeration, current typical on-site energy intensity per barrel of beer (whole brewery refrigeration: wort cooling, fermentation, maturation and cold storage; per volume, not comparable with the per-kilogram outputs)Show verbatim quoteRefrigeration 195 20,302 15,986 8,792 0Bandwidth 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. 58. DOE/EE-1571; OSTI ID 1512366 | Typical | 20,302 Btu/beer_barrel | Verified quote |
DOE bandwidth study, beer, refrigeration, state-of-the-art on-site energy intensity per barrel (same boundary)Show verbatim quoteRefrigeration 195 20,302 15,986 8,792 0Bandwidth 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. 58. DOE/EE-1571; OSTI ID 1512366 | Best available technique | 15,986 Btu/beer_barrel | Verified quote |
DOE bandwidth study, beer, thermodynamic minimum of the brewhouse step (negative: fermentation releases energy; per barrel)Show verbatim quoteBrewhouse 67,045 63,571 63,571 -8.8Bandwidth 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. 58. DOE/EE-1571; OSTI ID 1512366 | Theoretical | -8.8 Btu/beer_barrel | Verified quote |
Documented reduction of processing and cooling time of fermentation tanks with rotary jet mixing and external cooling (up to, %)Show verbatim quoteProcessing and cooling times can be reduced by up to 30 %.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. 361. EUR 29978 EN; JRC118627; doi:10.2760/243911 | — | 30 % | Verified quote |
9Unavoidable stops
- Cleaning of the vessel after each fermentationVerified quote
Yeast, trub and protein deposits must be removed and the vessel disinfected before the next batch. The duration is the minimum programme of the cleaning model (a lower bound for a tank cleaned with a spray device, assumption a5).
Show verbatim quote
The vessels have to be cleaned after each fermentation to remove the soil in them.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. 339. EUR 29978 EN; JRC118627; doi:10.2760/243911
10Assumptions
- Kinetics: the biomass model of the source, dX/dt = (mu - delta_X) X, dS/dt = -k_S mu X, mu = mu_max (1 - S_min/S) for S > S_min, mu_max = a ln(T) + b with T in °C, integrated numerically (fourth-order Runge–Kutta, 0.25 h step). The law was estimated on 17 L ale fermentations at 19–28 °C; mu_max vanishes at about 15.6 °C, so lager temperatures are outside its validity.
- Mass balance: Gay-Lussac stoichiometry (C6H12O6 = 2 C2H5OH + 2 CO2) with standard atomic weights gives the ethanol-yield ceiling; the documented yield is k_E/k_S of the kinetic model. Sugars are counted as hexose equivalents.
- Energy: the heat released is the measured heat of anaerobic fermentation per mole of glucose times the sugar fermented; the vessel is isothermal and heat gains through the insulation are neglected (they only add cooling).
- Capability: the batch-to-batch variation of the maximum growth rate is log-normal; the scheduled time is the fermentation time of a batch with mu_max exp(-z sigma_k), and batches slower than that miss the schedule (quality loss Phi(-z)). Fill, yeast cropping and transfer belong to the ideal cycle (common batch convention).
- Availability: the vessel is cleaned after each batch; the duration is the minimum programme of the cleaning model, which is valid for recirculated pipe circuits. Spray cleaning of a tank, including the CO2 displacement before caustic cleaning, is outside that validity, so the linked duration is a lower bound.
11Range of validity
- Isothermal batch fermentation of all-malt ale wort with Saccharomyces cerevisiae at 19–23 °C (kinetic data 19–28 °C).
- Lager fermentation (8–15 °C), maturation and bread-dough proofing are covered only by the structure of the bound, not by its numbers (derivation, section 7).
12Sensitivity
- Slope of the maximum specific growth rate against ln T (T in °C), biomass modelelasticity +1.549Spearman ρ —
- Intercept of the maximum specific growth rate lawelasticity -1.281Spearman ρ —
- Fermentation temperature (isothermal)elasticity +0.369Spearman ρ +0.20
- Batch-to-batch log standard deviation of the yeast's maximum growth rate (vitality, wort composition)elasticity -0.308Spearman ρ -0.92
- Margin of the scheduled fermentation time, in standard deviations of the growth rateelasticity -0.294Spearman ρ -0.26
- Settling coefficient of the yeast (loss of suspended biomass)elasticity -0.062Spearman ρ -0.05
- Minimum sugar quota below which the yeast stops growing (attenuation limit of the wort)elasticity -0.038Spearman ρ -0.05
- Time to fill the vessel with wort (several brews)elasticity +0.029Spearman ρ +0.10
- Sugar consumed per unit of biomass formedelasticity +0.026Spearman ρ +0.02
- Pitching rate (initial yeast biomass)elasticity +0.026Spearman ρ —
- Time to crop the yeast and transfer the green beerelasticity +0.022Spearman ρ +0.07
- Minimum duration of a cleaning-in-place programme (output cip_time_min of the cleaning model, with its Monte Carlo interval)elasticity -0.008Spearman ρ -0.01
- Completion criterion: residual sugar above the attenuation limit at which the primary fermentation is finishedelasticity +0.006Spearman ρ +0.02
- Initial sugar concentration of the wortelasticity +0.006Spearman ρ 0.00
- Output rises with the parameter
- Output falls with the parameter
13Validation checks
- Dimensional analysis19 outputs converted to their declared unitsPassed
- Every cited value verified against its source13 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
- Proofer equipment failureEquipment failure (breakdowns)1.1% [0.2–3.2%]D29 studies
- Unspecified fermentation equipment failureEquipment failure (breakdowns)1.0% [0.2–3.3%]D29 studies
- Coagulation and fermentation process defectProcess defects (scrap and rework)1.0% [0.2–3.2%]D29 studies
- Equipment configuration/setup timeSetup and adjustments0.5% [0.1–1.7%]D29 studies
15References
- Parameter Estimation of Dynamic Beer Fermentation Models (Zamudio Lara, Dewasme, Hernández Escoto, Vande Wouwer; Foods 11(22): 3602) (opens in a new tab). MDPI (open access, CC BY 4.0); full text served by Europe PMC, 2022. doi:10.3390/foods11223602; PMC9689312
- Heat of reaction in individual metabolic pathways of yeast determined by mechanistic modeling in an insulated bioreactor (González-Hernández, Michiels, Perré; Biotechnology for Biofuels and Bioproducts 17: 137) (opens in a new tab). BioMed Central / Springer Nature (open access, CC BY 4.0); full text served by Europe PMC, 2024. doi:10.1186/s13068-024-02580-8; PMC11587622
- 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. 361. EUR 29978 EN; JRC118627; doi:10.2760/243911
- 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. 58. DOE/EE-1571; OSTI ID 1512366
- doi:10.3390/foods11223602 — Biomass model of beer fermentation and its estimated parameters.
- doi:10.1186/s13068-024-02580-8 — Heat of reaction of anaerobic fermentation by yeast.

