**10. Mathematical, physical, chemical and biological limits by error propagation in continuous production**

A lot of factors have an influence on each step of malting and brewing. This ends up in a broad range of quality factors. Especially biologically balanced equilibriums react to changing conditions by complicated, not predictable effects, which can hardly be measured or recognised by sensors (**Table 3**).

Each quality parameter and each sensor, used during the processes, have its specific standard deviation and impreciseness during measurement. This can be respected, when intermedia products are checked for their quality. Corrective arrangements can be used to reach the final quality aim. During continuous production error, propagation may lead to a huge deviation in quality, which is also caused by the given impreciseness of the sensors who should avoid this, especially if lots of sensors are used in following steps to automatically control the continuous production [21].

An imprecise thermometer in the mash process will lead to different amounts of sugars or proteins, which can behave differently in the wort vessel than the wanted product. This may lead to different colours or yeast behaviours. The thermometers, used during fermentation, have a certain deviation as well, which might lead to different metabolisation products. They can be a favourable substance for other yeast or bacteria strains which also create unwanted flavour products. Sensors may detect but also have a deviation which allows unwanted processes. If sensors show a deviation to the quality aims, a continuous system should be able to adjust the process to the predetermined values. If this is not possible, the process has to be stopped. Analysis

**77**

*Continuous Beer Production*

Altered/influenced by

Volume Pressure Temperature Gas equilibriums Bubble size Bubble form Molecular size Heat capacity

**Table 3.**

**Figure 2.**

*DOI: http://dx.doi.org/10.5772/intechopen.86929*

**Gas phase Liquid phase (continuous) Solid phase**

Hydrophobic substances Hydrophilic substances

Temperature Temperature Temperature Pressure Shear forces Shear forces Microorganisms Microorganisms Microorganisms

Time Time Time

*Influencing physical and microbiological factors during beer production.*

*Mathematical estimation: example for additive error propagation [22].*

Enzymes Enzymes

Form Size

Density Zeta potential Hydrophobic liquids Hydrophilic liquids Heat capacity

Surface, roughness

Temperature Viscosity Surface tension Density

pH Heat capacity

of the intermedia product may help to decide how the quality targets can be reached by the following process step, in the way batch processes are successfully managed. The deviations in a naturally given physical, chemical and biological concentrations sum up and propagate by each sensor deviation and error used in the process. Calibration might help to a certain extent, but the amount of sensors is growing in continuous processes, so the amount of possible imprecise information is steadily increasing with the amount of measurements and control valves and regulations, which also have a certain deviation. Measurable substances like diacetyl, gravity, conductivity and turbidity are useful, but they are just single parameters in a bunch of varying aroma

In statistics, "propagation of uncertainty" also called "propagation of error" is the effect of variable uncertainties combined. Errors, or more specifically random errors, result in an uncertainty that builds up during long-lasting consecutive

particles, dependent on microbiological, chemical and physical balances.


#### **Table 3.**

*New Advances on Fermentation Processes*

microorganisms.

long time.

The average remaining time in a system is not similar to the real remaining time in a system. Molecules or particles entering a system at the same time may have different remaining times [13]. If there are two or more phases, like fluid and solid or gas particles, the continuous phase (usually the fluid of the beer and the foam) may behave totally different from the solids and the gas phase (bubbles) [13]. Coalescence will be influenced by the collision frequency and the behaviour of the substances during the phases [13]. Biochemical processes have lots of influencing factors, like temperature, pH value, viscosity, surface tension, osmotic and hydrostatic pressure, concentration gradients, mechanical influences, electrical effects, zeta potential and many more. Especially microorganisms change their behaviour under different conditions. As evolution does not stop at the brewing vessels, genetic deviations may cause different behaviours of raw materials and

If more than one species is present, synergetic or suppressing effects may end up in biofilms and uncontrollable developments. Working with just one species mostly *Saccharomyces*—is quite predictable in its final products. As soon as more species come up, things get more and more unpredictable. Hygiene helps to keep processes under control. Dead zones may become lively areas, and biofilms have to be avoided. Hygienic difficulties in cleaning have to be respected, and in fact, continuous systems cannot be cleaned as often, as batch containers if used for a

This causes problems, especially if brands need to have a constant quality to fit to the consumers' expectation which connected to the brand. Lots of homogenisation equipment during continuous production achieve a dense remaining time in the reactors. Energy, shear forces and moving parts are usually combined with abrasion of wear parts, which means a continuous change of quality of the machinery. Preventive maintenance needs to stop the processes and to start them again.

**10. Mathematical, physical, chemical and biological limits by error** 

Each quality parameter and each sensor, used during the processes, have its specific standard deviation and impreciseness during measurement. This can be respected, when intermedia products are checked for their quality. Corrective arrangements can be used to reach the final quality aim. During continuous production error, propagation may lead to a huge deviation in quality, which is also caused by the given impreciseness of the sensors who should avoid this, especially if lots of sensors are used in following steps to automatically control the continuous

An imprecise thermometer in the mash process will lead to different amounts of sugars or proteins, which can behave differently in the wort vessel than the wanted product. This may lead to different colours or yeast behaviours. The thermometers, used during fermentation, have a certain deviation as well, which might lead to different metabolisation products. They can be a favourable substance for other yeast or bacteria strains which also create unwanted flavour products. Sensors may detect but also have a deviation which allows unwanted processes. If sensors show a deviation to the quality aims, a continuous system should be able to adjust the process to the predetermined values. If this is not possible, the process has to be stopped. Analysis

A lot of factors have an influence on each step of malting and brewing. This ends up in a broad range of quality factors. Especially biologically balanced equilibriums react to changing conditions by complicated, not predictable effects, which can

**propagation in continuous production**

hardly be measured or recognised by sensors (**Table 3**).

**76**

production [21].

*Influencing physical and microbiological factors during beer production.*

$$\boldsymbol{\wp} = f(\boldsymbol{X}\_1, \boldsymbol{X}\_2, \cdots, \boldsymbol{X}\_n)$$

$$\overline{\mathcal{Y}} = f\left(\overline{\mathcal{X}\_1}, \overline{\mathcal{X}\_1}, \dots, \overline{\mathcal{X}\_n}\right)$$

$$\Delta \mathcal{Y} = \sqrt{\left(\frac{\partial f}{\partial X\_1} \Delta x\_1\right)^2 + \left(\frac{\partial f}{\partial X\_2} \Delta x\_2\right)^2 + \dots + \left(\frac{\partial f}{\partial X\_s} \Delta x\_s\right)^2} \Bigg|\_{\left(\overline{\mathcal{X}\_1}, \overline{\mathcal{X}\_1}, \dots, \overline{\mathcal{X}\_n}\right)}$$

**Figure 2.**

*Mathematical estimation: example for additive error propagation [22].*

of the intermedia product may help to decide how the quality targets can be reached by the following process step, in the way batch processes are successfully managed.

The deviations in a naturally given physical, chemical and biological concentrations sum up and propagate by each sensor deviation and error used in the process. Calibration might help to a certain extent, but the amount of sensors is growing in continuous processes, so the amount of possible imprecise information is steadily increasing with the amount of measurements and control valves and regulations, which also have a certain deviation. Measurable substances like diacetyl, gravity, conductivity and turbidity are useful, but they are just single parameters in a bunch of varying aroma particles, dependent on microbiological, chemical and physical balances.

In statistics, "propagation of uncertainty" also called "propagation of error" is the effect of variable uncertainties combined. Errors, or more specifically random errors, result in an uncertainty that builds up during long-lasting consecutive


**Figure 3.** *Deviations in batch and continuous brewing processes.*

processes. Measured values are necessary to control the process. All measurements have uncertainties due to limitations (**Figure 2**). Instrument precision and other deviations propagate due to the combination of variables in the function [19].

If biological changes occur in raw materials or microorganisms, errors do not add but multiply, or exponentiate, as several factors change: aromas, natural substances, pH, viscosity, chemical balances, concentrations, compositions, sublimation, dissolving, evaporating, etc. Respecting the rheologically unpredictable behaviour of altering three phases during rheological processes [13], a constant and planned quality can only be expected for a certain time; quality changes and emptying and cleaning of the system become necessary.

Continuously changing products like mash, fermenting of green beer has to come to a stable and defined quality, which fits to the consumers' expectation. This is traditionally achieved by cold stabilisation during maturation, which is also a possibility to check and adjust quality, to blend the beers, and it is a buffer before bottling and racking. Filtration and pasteurisation help to stop the most biological and enzymatic processes. Especially before bottling a quality check has to be conducted, to avoid faulty products being bottled. The product will be kept during distribution and at the consumer's place for a certain time, where it should not change anymore. Finally the consumer's expectation cannot be adjusted to an unpredictable quality (**Figure 3**).

**79**

**Author details**

the machines.

Hochschule Geisenheim University, Geisenheim, Germany

system instead of brewing and creating the quality of the beer.

\*Address all correspondence to: mark.strobl@hs-gm.de

provided the original work is properly cited.

Mark Strobl

*Continuous Beer Production*

**11. Conclusion**

*DOI: http://dx.doi.org/10.5772/intechopen.86929*

following process up to a stable equilibrium.

Stepwise batch production has been added by continuous brewing processes for more than 100 years. Calculations may predict the results in biological processes to a more accurate degree, if computers obtain more and more data to use. The results in the end are estimations, relying on the number of factors in relation to their influencing parameters. The possibility to measure these factors need to have a known accuracy. This will stay the biggest challenge, as long as natural ingredients and microorganisms play a main role in malting and brewing. Fermenting with more than one yeast strain quality, prediction falls back to the times of spontaneous fermentation, where only few products were of a good taste by accident. The same thing occurs, if contaminations get into the continuous systems, when cleaning always implies a long interruption of the process, especially a long time to adjust the

Many substances in beer are known, many not. Some substances or physical properties are useful to define the quality of the beer and the intermedia production steps. Few parameters can be used to control the brewing process in the direction of a defined quality aim. All of the raw materials, the determination methods, the regulation equipment and the biological reactions of the microorganisms have deviations, some of them are not measurable or known yet. In a batch process interstage products like malt, wort or fermented beer can be checked, and the following processes can be used to adjust the quality goals. Reactions to the deviations in continuous processes should set the system back to the target conditions. Combining several steps to a continuous process with continuous adjustment can be controlled by a self-learning by fuzzy logic or artificial intelligence. As these computers need data from precise sensors to control precise valves, stirrers, pumps, etc., this will lead to a predominantly maintaining, calibrating and scrutinising reaction to the

Nowadays a continuous step is followed by a batch step. The possibility in adjusting the quality with following processes should be given. Also blending needs buffer capacities to equalise the beer to consumer's expectation. For offering different beer types, also blending beverages are necessary to be added. The continuously produced beer might supply a base beer, which is blended before filtration with other batch process beers or water, aromas and lemonades in a nontraditional way.

The brewing process can be performed in steps, continuously and automatically. Perhaps one day continuous processes will be longer stable. The processes may also expire in constant quality results. But brewing is fun, fun that should not be left to

© 2019 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,
