Introduction to Problem-Solving Techniques

Chapter 1

Introduction to ProblemSolving Techniques You can’t solve a problem with the same type of thinking that caused it. Einstein

The quote from Einstein may seem like a statement of the obvious, but after many years of experience in the baking industry, I have seen that the obvious is constantly overlooked when it comes to trying to solve problems or develop new products and processes. Indeed, there is relatively little difference between solving a problem and creating a new product, in both cases, you are required to use different thinking to that you would normally use for established products and processes. In essence, both scenarios are vindications of Einstein’s view. Problems that show as unexpected variations in bakery product quality do occur from time to time. Often considerable time, effort and money are required to identify the causes and solutions concerned. Unexpected quality variations are not the exclusive province of any particular size of manufacturing unit: they can occur in both large and small bakeries. Nor are they exclusive to the production bakery: Even the best-controlled test bakery or laboratory can experience unexpected fluctuations in intermediates or final product quality. Because the outcome of a baking operation depends on complex interactions between the raw materials, recipe and process used, it is often the case that it is only when the final product leaves the oven that quality defects are detected. There are relatively fewer occasions when intermediate products (e.g., dough, batter, paste) exhibit quality defects which require an immediate change to be made. Many bakery operations still have artisan or craft (small-scale) roots. Even with the arrival of industrial-scale baking many years ago, the manufacturing principles still rely on understanding heuristic rules and relatively limited data analysis. The level of automation and the ability to collect and analyse data from an industrial bakery still lags a long way behind that of other

Baking Problems Solved. DOI: http://dx.doi.org/10.1016/B978-0-08-100765-5.00001-1 © 2017 Elsevier Ltd. All rights reserved.

1

2

Baking Problems Solved

manufacturing environments. There is some routine data collection and analysis for production and business management purposes (e.g., yield from a given set of raw materials), or for legislative reasons (e.g., baking losses to determine point-of-sale weight or product moisture content), but the collection and integration of data on-line for the optimisation of bakery of bakery processes and products remains meagre (Cauvain, 2015). The challenge lies as much with knowing what to measure and what data to collect, as it does with the analysis. In such contexts, the emphasis on delivering the appropriate product quality will inevitably on combining a methodical approach with relevant knowledge. There is no magic to problem solving. It is normally achieved through critical observation, structured thought processes and access to suitable sources of information. In this chapter, I offer a guide to some of the methods that might be employed when trying to solve bakery-related problems. In doing so, as noted above, we must recognise that baking is a complex mixture of ingredient and process interactions, so that the solutions to our problems may not always be instant in nature and because ingredients and processes change, new solutions are always being discovered. The complex interactions which underpin baking dictate that there are seldom unique solutions to individual problems. In the majority of cases, individual quality defects are overcome by changing a number of ingredient and process factors some of which will be apparently unrelated though careful study will often reveal that relationships do exist even where they are masked by more prominent effects. It would also be appropriate at this stage to deal with the somewhat amorphous term ‘quality’. Ultimately, the decision of what is the ‘right’ product quality lies with the consumer, what is acceptable or ‘good’ for one consumer may be unacceptable or ‘bad’ for another. For the baker perhaps the best basis for deciding what the right product quality is depends on getting repeat product sales. In many of the questions and answers related to bakery production which follow, there is an implicit understanding that a particular quality defect is delivering unacceptable (bad) final product quality. In providing potential solutions to particular problem, it is recognised that the choice of a particular solution will depend on many factors, including cost and practicality of application. The answers given should be seen as a guide as to possible solutions and so are often given with a degree of flexibility as to application.

1.1 HOW TO PROBLEM SOLVE Successful problem solving usually requires a methodical approach. It is perfectly possible to stumble quickly on the required solution by chance but more often than not a haphazard approach to problem solving is wasteful of time, resources and money. In addition, stumbling on the solution by chance

Introduction to Problem-Solving Techniques Chapter | 1

3

often means that the root cause of the problem remains unidentified and the opportunity is lost for the systematic assembly of information which may be valuable for avoiding or solving similar problems in the future. Not all problems are solved using exactly the same approach but the critical elements of the problem-solving process are largely common. In practical problem solving, we normally move from the problem to the cause and finally to the corrective action. However, we must recognise that on many occasions, the manifestation of a particular problem does not necessarily have a unique and identifiable cause and so there may be other intermediate steps to take into account in determining the real cause of the problem. This situation can be described schematically as follows: Problem-primary cause-contributing factors-corrective action Or in more simple terms as: What is seen-why-because of-corrective action The basic process becomes apparent if we consider two examples of problems in breadmaking; the first low-bread volume and the second collapse of the sides of an open top pan loaf, often referred to as ‘keyholing’ (see page 4).

1.1.1 Low-bread volume Externally, we observe that the bread is smaller than we expect, and this may also have led to a paler crust colour due to the poorer heat transfer to the dough surface during baking. Internally, the cell structure may be more open than usual. As bread volume is a consequence of the expansion of the dough by carbon dioxide gas from yeast fermentation and the retention of that gas within the dough matrix (Cauvain, 2015), there are two potential primary causes of this problem Lack of gas production and lack of gas retention. To separate the two, we will need more observations, and an important one will be whether the rate of expansion of the dough in the prover and oven was normal or lower than usual. If the former was the case, then the primary cause of the problem is likely to be lack of gas production and potential contributing factors may include the following: G G G G G G G

yeast activity or level too low; lack of yeast substrate (food); dough temperature too low; proving temperature too low; proving time too short; salt level too high; proving temperature/time/yeast combination incorrect.

4

Baking Problems Solved

On the other hand, if the dough proving had been at a normal rate and there was a lack of oven spring, then this would lead us to recognise that the problem could be lack of gas retention. In this case, the list of potential reasons for the problem includes: G G G

G G G G

improver level too low; incorrect improver formulation; combination of improver and flour too weak for the breadmaking process being used; enzymic activity too low; energy input during mixing too low; mixing time too short; dough temperature too low.

Note that the ‘dough temperature’ too low appears in both lists due to its effect on yeast activity and the effectiveness of the functional ingredients in the improver at a given temperature, especially if ascorbic acid and enzyme additions are used.

1.1.2 Keyholing Externally, we observe that there is a loss of bread shape but only at the sides of the product. Internally, we may see the formation of darkcoloured, dense seams, often referred to as cores (see Fig. 1.1 and Section 4.1.2). The centre crumb may be more open than we normally expect for the product concerned. Why has this happened? Clearly, we have no problem with gas production since there is no evidence for slow proving and the bread had good volume. We have clearly retained the carbon dioxide gas produced, otherwise the bread would have low volume as described above. In this case, the overexpansion of the crumb in the centre of the loaf leads us to the view that in fact the gas retention is excessive. Thus, the primary cause of the problem is excess gas retention arising from a number of potential individual causes or combinations. The contributing factors may include: G G G G G G

improver level too high; incorrect improver formulation; combination of improver and flour too strong for process; enzymic activity too high; energy input during mixing too high; mixing time too long.

From the foregoing examples, we can see that observation and reasoning are key elements in problem solving. The former can be readily systematised,

Introduction to Problem-Solving Techniques Chapter | 1

5

FIGURE 1.1 Keyholing in bread.

whereas the latter will rely heavily on the availability of suitable information to use as the basis for comparisons. The potential sources of such information are discussed below. It is interesting to consider the process by which one might set about identifying the particular cause of a problem, such as the keyholing (excessive gas retention) of bread discussed above. The most likely mental process is one associated with probability achieved by matching the pattern of observations with ones previously experienced and remembered. When we recognise a general similarity between observation and stored image, we are likely to explore in more detail the factors most likely to contribute to the pattern we see. One potential analogy for how we solve problem is that of a tree. The main line of observation is via the central trunk with the potential to explore branches at many points. In the case of our bread problem, if we fail to identify the likely cause of the problem from our first consideration, then we will close down that line of reasoning, go back to the main theme (the trunk) and then set off an another branch of investigation. Our route through the branches of our reasoning or knowledge tree is complex and

6

Baking Problems Solved

occasionally we may jump from branch to branch rather than going back to the trunk before continuing our investigation. The length of time that we take to identify the cause and the corrective actions needed varies considerably from occasion to occasion and from individual to individual, and is more likely to be related to our accumulated knowledge and experiences rather than logical reasoning. Our abilities to recognise and match subtle patterns are probably so intuitive that we are seldom aware of them.

1.2 THE RECORD It is common for the manufacture of bakery products to be based on some starting formulation and formal method of processing the ingredients into the finished product. This will require some form of recorded details of the ingredients to use, their qualities and quantities, and the equipment, process settings and timings involved. Consult any standard recipe book for bakery food preparation, and you will find such details recorded for use by others. In almost all modern bakeries, a formal production record will be set up for each of the product types and used by the manufacturing operatives to prepare the various ingredients and set the equipment. Invaluable in problem solving is the formal record of what was actually carried out on a particular occasion. Although many operatives will keep to the prescribed formulation and processing recipe, small variations about a given value can occur and lack of information of what the actual values were for a given mix makes problem solving more difficult. It is normal for standard production specifications to allow a degree of tolerance for weights and operating conditions. For example, a temperature specification for a cake batter may be stated as 20 6 2 C. However, such a specification allows for replicate batters to be 18 C or 22 C and 4 C variation coupled with other small changes may have a larger effect of final product quality than normally considered. A formal record of production can encompass many aspects including the following: G

G

G

Any variations in the source of the raw materials. For example, changes in flour or whole egg batches, or a new supplier of a particular ingredient. Changes in analytical data even where these are still within acceptable limits because the cumulative effect of small changes in a number of individual parameters can have a large effect on final product quality. The actual quantities of ingredients used compared with the standard values. For example, in breadmaking, it is common to adjust the water level added to maintain a standard dough rheology for subsequent processing. In other cases, deliberate changes from the standard formulation

Introduction to Problem-Solving Techniques Chapter | 1

G

G

G G

7

may have been introduced to compensate for some process change. For example, in bread dough the yeast level may be adjusted to compensate for a change in prover temperature so that final proving times do not vary. The processing conditions, such as mixing times, energies, ingredients and batter or dough temperatures. Once again the values may fall within acceptable ranges but can still have a cumulative effect with other small changes in recipe and process parameters. Process equipment settings which may vary according to ‘operator preference’ or due to variations in other factors. For example, an unavoidably higher laminated paste temperature may result in greater damage to the laminated structure which may require a compensatory adjustment to roll gap settings during sheeting. Process timings, such as baking or cooling times. Changes in packaging materials.

The record may be simplified by using the standard recipe as a pro forma against which to record variations. Such techniques have been commonly used to record the weights of individual dough pieces coming from the divider (see Fig. 1.2) and can be readily adapted for any aspect of bakery production. The record may be on paper, by input to suitable computer-based programs or may be gathered and stored automatically. In addition to the recipe and process records, it is very important to have a formal record of finished product quality. Once again, it will be common to have some form of product specification with appropriate tolerances against which to make an assessment. Such techniques are commonly the province of the Quality Control Department. The degree of detail recorded will vary. Many examples of approaches to quality control techniques for the baking industry are known and the reader is referred elsewhere (e.g., Street, 1991; Manley, 2000). The role of quality control should be more than that of the final gatekeeper for product quality, it should provide an important link between the specifications of raw materials, process data and final product Product unit weight (g)

Dough temperature (ºC)

Time to divider

* Dough consistency codes: S = softer than normal SS = slightly softer than normal N = normal consistency SF = slightly firmer than normal F = firmer than normal

FIGURE 1.2 Example of a divider record sheet.

Dough consistency* S/SS/N/SF/F*

Divider setting

8

Baking Problems Solved

quality. All too often, the quality control function is divorced from this integrated chain of information and commonly relegated to a ‘checking’ role. The manufacture of bakery products is somewhat different from that of making many non-food items in that is often impossible to ‘reuse’ product which is outside specification so the principle of ‘getting it right’ first time is critical for efficient and cost-effective production. Data gathering is an essential part of the information chain, integration of the information is critical and analysis of the data (see below) is vital. For use in problem solving, the formal product specification or quality control record may require some adaptation and enlargement as small, but commonly accepted, variations may hold the vital clue to the particular cause of a problem. In both, the quality control and problem-solving contexts relevant data on the finished product may include the following: G

G

G

G G

G

G

Product size based on height or volume. Devices for measuring product dimensions may be used off- or on-line. They may be as simple as using a rule to measure loaf height or measuring product volume by seed displacement in a suitable apparatus (Cauvain, 2015) or with laser sensors (Cauvain, 2017). Shape may be assessed subjectively and compared with an accepted standard. The introduction of image analysis offers opportunities for recording product shape, even on-line (Dipix Technologies Inc, www. dipix.com). The external appearance of the product and the recording of any special features that may be present or indeed the absence of expected features, e.g., lack of oven spring in bread. Surface blemishes, their size and location on the product. The colouration of all external surfaces. Descriptive techniques, comparison with standard colour charts, e.g., Munsell (Munsell, no date) or tristimulus instruments (Anderson, 1995) may be used. Deviations from the norm should be clearly noted. The appearance of the internal structure, if there is one. Most baked products have some form of internal structure that is an intrinsic component of product quality. Assessment of that internal structure may be subjective and describe the size, numbers and distributions of the cells (open spaces) which go to make the internal structure. Cell structures may be unevenly distributed in the product cross-section or form a ‘pattern’ that is characteristic in different products. Deviations from the norm may be noted. Image analysis is now being used for objectively assessing internal cell structures (Cauvain, 2003; Whitworth et al., 2005; Cauvain, 2013; Cauvain, 2017). The internal colour may be assessed using techniques described above for surface colour. It is worth noting that the presence of a cellular structure has an impact on the perception of colour so it is often common practice

Introduction to Problem-Solving Techniques Chapter | 1

G

G

9

to include some form of visual assessment, e.g., brightness, which is different from the true colour of a product. Some objective image analysis systems offer a measurement of crumb brightness, e.g., C-Cell (www. c-cell.info). The physical characteristics that contribute to eating quality may be assessed subjectively with ad hoc or trained panels. Alternatively, some form of objective test designed to mimic aspects of sensory analysis may be employed, e.g., texture profile analysis (Cauvain, 1991), squeeze and puncture tests (Cauvain and Young, 2006a; Cauvain, 2017). Product odour and flavour may be assessed subjectively on an ad hoc basis or with trained panels. The development of the so-called ‘electronic nose’ may offer a more objective measure but has yet to approach human sensitivity.

Whatever details are considered to be appropriate for the record, it is important to have a standardised format for recording the details. This usually takes the form of a standardised record sheet, paper or electronic, with blank spaces in which to enter the appropriate data or comments. Where a product attribute cannot be measured, an attribute ‘scoring’ system might be used to provide a more objective basis for analysis of the problem. Any number of scoring systems may be employed. One example is given in Fig. 1.3, and others are given in the literature (e.g., Kulp, 1991; Bent, 1997a; Cauvain and Young, 2006a).

1.3 THE ANALYSIS If a standard record sheet is available, then the initial analysis can be as simple as considering whether the recorded data deviate from the process specification and in what direction. The effects of any changes can then be compared with existing knowledge bases (in whatever form) to provide the basis of a diagnosis. Sadly, few bakery problems are solved with such a simplistic approach. Almost all bakery processes include an element of elapsed time, e.g., proving, baking and lamination, which must be taken into account when analysing the causes of problems. Many larger bakery operations involve continuous production, even though they are batch fed, and this adds a further complication to take into account in the analysis. An example from my own experience is that of a plant manufacturing baked puff pastry shells where deviations in the product dimensions were identified at the end of the baking process. In this instance, the plant had to run continuously to be efficient and not compromise product quality (i.e., no gaps in the pastry sheet or the oven). The operation was batch fed from the mixer so that the relationship between a given mix batch, and the product leaving the oven had to be established first. When this was done, it then

10

Baking Problems Solved Product .............................................................................................................................. Recipe code ....................................................................................................................... Date manufactured ........................................................................................................... Date evaluated .................................................................................................................. Evaluated by .....................................................................................................................

Product weight (g) ....................................................................... Notes on key attributes Product height (mm or max. 10) .................................................................High

Low

Volume (cm3 or max. 10) ............................................................................ High

Low

External appearance Uniformity of shape (10 max.) .......................................................... Collapsed Peaked Crust break (5 max.) ................................................................................ Even Uneven Crust colour (5 max.) ............................................................................... Dark

Light

Internal appearance Crumb cell structure (max. 10) ............................................................... Open Close Crumb uniformity (max. 10) ................................................................... Even Uneven Crumb colour (max. 5) .................................................................................................. Sensory qualities Aroma (max. 5) ............................................................................................... Off-odour Flavour (max. 10) ......................................................................................... Off-flavour Crumb firmness/softness/crispiness (max. 10) ....................................... Eating qualities (max. 10) ........................................................................ Total score (max. 100) ................................................................................ Additional comments .................................................................................................... ..............................................

FIGURE 1.3 Example of product scoring sheet.

became possible to identify the contribution that any variation in the mix batch contributed to the problem. After establishing this relationship, it became clear that batch-to-batch variation was not the prime cause of the problem observed as simple plots of dough properties ex-mixer (e.g., temperature or rheology) did not correlate with variations in product quality even when the elapsed time element had been taken into account. The solution to this particular problem lay in a plot of changes in product character with time (see Fig. 1.4), which upon analysis showed that the variation was more regular than first thought. At first glance, it appeared to be the well-known ‘shift change effect’ and to some extent that was true: Not entirely in this case due to the operator effect on process settings but because each new shift started with a new batch of rework to add to the virgin paste. As the rework aged, the effects on baked product character diminished. In this case, a simple trend analysis provides the basis for the solution of the problem.

Introduction to Problem-Solving Techniques Chapter | 1

11

60

Pastry height (mm)

55

Stored re-work

50 45 40 35

Fresh re-work 30 25 0

2

4

6

8

10

12

Dough batch number

FIGURE 1.4 Effect of rework on lift in laminated products.

One analysis technique that has been applied to cereal science and technology is ‘root cause analysis’ (Stauffer, 2000). Not all bakery problems are likely to be potential subjects for this type of analysis as a key element in this technique is the brainstorming session. Brainstorming usually implies that more than one person is involved and all too often many of us confront bakery problems alone or against a timescale that is insufficient to gather together the necessary team of experts. In manufacturing processes based on batch production, stopping the line until the problem is solved is an option; however, for many bakery processes, anything other than a shortterm stoppage is seldom an option. If the problem is a persistent one or of a catastrophic nature, then root-cause analysis can be a suitable technique to apply. The role of a team in employing root-cause analysis is invaluable in solving intractable problems or making changes to product quality. In the latter case, the technique would be to treat the required change as though it were a problem; e.g., if I want greater volume in a cake, then by diagnosing the cause of excess volume, I may well obtain clues as to how to increase cake volume, or by treating low volume as problem then I may get pointers as to possible routes to improvement. The methodology known as Six Sigma has been used to quantify how bakery processes are performing. At the heart of the operation is the implementation of a measurement strategy based on no more than 3.4 defects per million opportunities, with a Six Sigma opportunity being based on the number of chances of getting a defect. For reasons already discussed, the opportunities for obtaining objective measurements in many bakeries are limited which means in turn, so are the chances of using statistical approaches like Six Sigma as a routine problem-solving tool. Nevertheless, the discipline needed to implement the Six Sigma methodology has potential relevance for baking. The Six Sigma DMAIC approach is based on define, measure, analyse, improve and control, all essential elements in any manufacturing process.

12

Baking Problems Solved

1.4 MODELLING TECHNIQUES The application of statistical methods of analysis is common in many areas of food and nonfood manufacturing scenarios, one example is that of Six Sigma discussed above. The different techniques can be used in problem solving and quality optimisation, though in the manufacturing environment modelling methods often tend to be confined to the plotting of trends using simple graphs as discussed in the example above for a laminated product. More sophisticated statistical and modelling techniques can play their part in helping to buildup the information base on what the critical ingredient and process factors are which determine changes in product quality. Once identified, these critical factors can be logged and matched with problems when they occur. Examples of modelling processes associated with baking are often associated with production or financial planning, and energy management rather than product quality. To develop predictive models which deal with product quality, it is commonly necessary to carry out experiments in a test bakery or trials on the plant. Although trials on the plant are preferred, they can be wasteful of raw materials, energy and time so that the most common practice is to carry out evaluations in the test bakery and ‘translate’ the results to the plant. It is very important to establish any clear changes that are relevant when translating test bakery results to a plant environment. A simple example encountered by the author was the development of a sponge cake recipe in a test bakery using a planetary-style mixer, whereas the plant used a continuous mixer to prepare the same recipe batter. In this case, it is necessary to remember that less carbon dioxide gas will be lost during continuous mixing (due to the closed nature of the mixer head) than with an open-bowl planetary mixer so that baking powder levels should be adjusted downwards to compensate for this difference. A typical adjustment would be to reduce the baking powder level for a continuous mixer to be about 75% of that used on a planetary mixer to achieve the same sponge cake volume in both the test bakery and on the plant (Cauvain and Cyster, 1996). There are a number of examples of modelling techniques which might be applied to bakery products. Street (1991) provides a review of suitable techniques that may be applied to the modelling of baked products, and there are a number of examples in the scientific and technical literature. The concept behind the development of such mathematical models is that a relatively limited number of experiments may be used to build models that can be used to predict changes in bakery product quality as a consequence of changes in combinations of ingredients and processes. Once a predictive model has been established, then the information could also be used for problem solving. For example, suppose that we show by experimentation how loaf volume varies as a result of an interaction between the level of ascorbic acid in the dough and mixing time. At some later stage,

Introduction to Problem-Solving Techniques Chapter | 1

13

we may encounter a problem with low-bread volume, and then we would be able to use the output from our model to help decide whether the problem was associated with the level of added ascorbic acid or mixing time, or both. Furthermore, we might use our model to show which changes were most likely to restore our bread volume to its original level. Baking is a complex food process with many ingredient and process interactions. These interactions lead to complicated models that are often difficult to apply. For example, for a given set of mixing conditions, we would observe that bread volume increases with increasing levels of ascorbic acid reaching a maximum, and thereafter, there will be little change in volume for increasing additions of ascorbic acid. This occurs because the oxidation effect of ascorbic acid is limited by the availability of oxygen from the air incorporated during dough mixing (Cauvain, 2008). The availability of oxygen is affected by yeast activity, so that yeast level becomes an influencing factor. Both yeast and ascorbic acid activity are temperature sensitive and proceed at a greater rate when the temperature increases. Dough temperature is a function in part of ingredient temperatures and in part the energy imparted to the dough during mixing. Energy transfer in turn is related to the mixing time. So, too, is gas occlusion to a lesser degree because during mixing an equilibrium point is reached when the entrainment process is balanced by the disentrainment process. This equilibrium may occur before the end of the mixing time. So for the given example above, although we set out to study the effects of the level of ascorbic acid and mixing time, we must also ensure that we measure: G G G G G

ingredient temperatures; final dough temperature; gas occlusion in the dough; actual mixing time; energy transferred to the dough.

These records are necessary because we cannot independently control some of the properties concerned, e.g., mixing time, energy and dough temperature are all interrelated. Whenever we do work during mixing, we must expect there to be a temperature rise. This relationship also holds true if a water or other coolant jacket has been fitted to the mixer, and in this case, we must remember that the coolant temperature in the jacket will also rise by the time that it leaves the jacket as the result of absorbing some of the heat generated during mixing. There tends to be greater variability in product quality for products manufactured on a plant than one sees in many test bakery environments. This process ‘noise’ in the data can mask some of the critical issues that control product quality and therefore weaken the value of any models which may have been developed. There are a number of statistical techniques that

14

Baking Problems Solved

can be used to help separate such noise from underlying effects, trends and relationships. In many manufacturing processes, the specified product characteristics can be achieved by many different combinations of formulation and process conditions. Taguchi methods use experiments to search systematically and efficiently for combinations of ‘control’ factors that minimise product variability in the face of variations in ‘noise factors’ such as ambient temperature. Taguchi methodology has been applied to the manufacture of bakery products, in particular in a study of the factors that affect the quality of puff pastry (DTI, 1993). In some cases, effective problem solving can be initiated by studying the effects of small perturbations on the plant. A major issue with carrying out trials on the plant is the potential loss of production arising from the manufacture of out-of-specification products. However, there is a distinct advantage to plant trials in that large numbers of samples are being made which increases the potential for statistical and practical analysis. Most product specifications have a degree of tolerance associated with the final product so that small variations can be accommodated without loss of production. The value of statistics in identifying potential differences between the effects of ingredient or process changes is not doubted. However, as discussed in the example above where the relationship between ascorbic acid level and mixing time was considered, there are relatively few simple relationship in baking. One often hears the comment with regard to experiments in baking that ‘we will change one thing at a time’. Sadly, this is seldom if ever, true for baking. Suppose that want to consider the effect of reducing the base dough temperature in the manufacture of croissant and we therefore make two doughs, one at 20 C and the other at 15 C, that ‘simple’ change in temperature will affect gas production by the yeast, the activity of ascorbic acid and enzyme if added, and most importantly the rheological properties of the dough during processing. Differences between trials may indeed show statistical differences in product quality but due to the range of interactions, the relevance of the differences with respect to production requirements is not clear and so care must be taken when implementing actions or models based exclusively on statistical significance alone.

1.5 MATCHING PATTERNS AND VISUALISING CHANGES Sometimes when dealing with complex problems, it is an advantage to sketch out the salient features with a diagram or create some collage of salient information on a board (like a story board for the creation of a film). A simple example is illustrated in Fig. 1.4 in which the potential routes for the migration of moisture in composite bakery products are identified annotated with relevant data on moisture contents and product masses.

Introduction to Problem-Solving Techniques Chapter | 1

15

Topping Aw = 0.6, Moisture = 3% Mass = 4% of product Top cake Aw = 0.85, Moisture = 25% Mass = 43% of product Filling Aw = 0.93, Moisture = 40% Mass = 10% of product Bottom cake Aw = 0.85, Moisture = 25% Mass = 43% of product FIGURE 1.5 Schematic identifying the characteristics of the different components of a composite cake product and the potential routes of moisture migration.

The drawing of diagrams such as that shown in Fig. 1.5 helps to ensure that all of the relevant processes are considered before carrying out detailed calculations and investigations. Human beings have a significant capability for being able to match patterns in data, and in many ways, when we are problems solving we spend a lot of time comparing what we see with the patterns which we all hold in our minds. Subconsciously, we look for a pattern of information in a current problem and compare that with previous patterns of events and information to see if they provide clues for solving the current quality problem. There are many different ways of creating patterns. The creation of knowledge trees and knowledge fragments is one example and is discussed in more detail in the following section. The knowledge tree is like a flow diagram similar to that created by engineers to show the movement of raw materials through its various stages en route to becoming a finished product. The same basic principle is used by systems analysts when they are constructing diagrams to show the flow of information with different symbols representing different types of activity or decisions which need to be made. Cauvain and Young (2006a) illustrated possible examples of pattern matching for the baking industry using a series of ‘spider diagrams’ to relate certain characteristics of wheat with that of the subsequent flour, dough and bread. As well as providing a relatively simple means of developing patterns relating raw materials and finished products the process of deciding which characteristics to include in the various diagrams is an important first step in understanding the cause of quality problems. When it comes to identifying the key roles of different ingredients and processes in determining a particular aspect of final product quality, it is

16

Baking Problems Solved Harder Flour protein Gluten development Sugar

Fat Crumblier FIGURE 1.6 The impact of some process and ingredient factors on the hardness and crumbliness of cookies.

useful to be able to identify the relative importance of the individual changes. It may be possible through mathematical modelling to identify the relative importance of the effect different ingredients, recipes and process or process changes, but it can sometimes be sufficient in problem solving to use a simple diagram to understand the different contributions (Cauvain and Young, 2006a). An example of this type of approach is given in Fig. 1.6 which examines the impact of some process and ingredient factors on the hardness and crumbliness of cookies. The development of a gluten network in cookie dough is not usually considered to be desirable, and if this should happen, e.g., through over-mixing, the resultant product will be harder eating. As the sugar level in a cookie formulation increases, the resultant product gradually loses its initial crumbliness and become harder (e.g., as seen with ginger nuts) while increasing additions of fat give increasingly crumbliness as the fat interferes with the development of the gluten structure. The angle at which the individual vectors proceed from the origin in Fig. 1.6 gives an indication of the relative impact of any changes so that this relatively simple diagram can provide a first indication of the potential interactions that taking place in a particular baking environment. For example, from Fig. 1.6, if we wished to reduce the fat level in a cookie formulation but do not wish to end up with a harder biscuit then we would consider a reduction in gluten development by adjusting mixing conditions or methods or changing ingredients, such as flour type, which contribute to gluten formation.

1.6 THE INFORMATION SOURCES Not many of the problems that we may encounter in the manufacture of bakery products are likely to be so unusual that they have not been encountered and recorded before. Even where an apparently new problem arises, access to suitable information sources often reveals a problem and solution so similar that it can be readily adapted to our particular needs. For example,

Introduction to Problem-Solving Techniques Chapter | 1

17

most of the problems that we are likely to encounter in the production of cakes with heat-treated cake flours (see Section 2.2.17) will have similar solutions to those that would apply if we were using chlorinated cake flours (see Section 2.2.17). Even though it may be the first time that we have used a heat-treated flour, we therefore have a suitable base for identifying the solution to our problem. The availability of suitable information is a fundamental tool for our ability to problem solve successfully. Traditionally, such information sources could be classified as personal and written. More recently, computer-based sources have become increasingly available sometimes as databases but in other cases in forms that would not be classified as an electronic equivalent of the written word.

1.6.1 Personal Even in today’s fast-moving electronic age, there is no substitute for personal experience which builds one’s own portable information source. However, few of us will spend long enough in positions that allow the systematic buildup of the appropriate knowledge through ‘trial and error’ studies. Aspects of problem solving may be taught in our years of academic study, but these are seldom detailed enough to provide us with the comprehensive information base required. Personal contacts with experts and consultants can be used to supplement our individual information base. Contacts with other professional bakers and professional baking organisations are invaluable because it allows access to a wider range of experiences. Thus, membership of professional bodies such as the British Society of Baking, the American Society of Baking and the Australian Society of Baking, which are linked with one another, has benefits in developing one’s own knowledge base. Attendance at suitable conferences, technical meetings, workshops and short courses can also provide relevant information.

1.6.2 Written The scientific and technical literature provides the most obvious source for written material which aids in problem solving. Starting a collection of ‘useful’ articles and some form of index is very helpful in establishing your own information base. Included in the written form are pictorial libraries of faults and associated text related to their identified causes. Such libraries may be built for oneself or may be purchased from a suitable source. Over the years, many of the ‘rules’ related to problem solving in baking have been summarised and published (e.g., Street, 1991; Bent, 1997b; Cauvain and Young, 2006b). These generally take the form of lists of faults

18

Baking Problems Solved

and associated causes. In many ways, such rules are of limited value because they seldom consider or assign a likelihood value and so a personal degree of judgement as to which of the causes to investigate first is required. Such lists tend to deal only with the more common problems and seldom consider interactions between ingredients or ingredients and processing. Also the causes of faults are given equal weighting; thus, there is no expression as to whether a particular cause is more likely than another. The values of a personal record can be significantly increased by systemising the knowledge record. A series of checklists can be constructed to identify contributions of ingredients and processes to final products and their appropriate intermediates (e.g., dough, batter, paste). An example of such an approach is illustrated for pastry in Tables 1.1 1.6. Checklists may be populated with the type of information identified in Chapter 3, Key Relationships Between Ingredients, Recipes and Baked Product Qualities. A ‘first level’ checklist (Table 1.1) identifies the ingredients that may be used in the manufacture of pastry and considers the potential impact of their on the various final product characteristics. Filling in this first checklist

X

X

X

Fat type Fat melting point

X

X

X

X

X

X

X

X

X

Milk Egg

X X

X

X

X

Moisture

X

Aroma

X

Mouthfeel

X

Colour

X

Sugar particle size Rework

Firmness/Tenderness

Flour water absorption

Surface Appearance

X

Shape

X

Height/Lift

Paste Extensibility

Flour protein content

Paste Elasticity

Paste Consistency

TABLE 1.1 Example of Level 1 Checklist for Recording the Potential Effects of Ingredients and Their Qualities on Paste and Pastry Characteristics (Level 1 Checklist Pastry/Ingredient Qualities)

X

X X

19

Introduction to Problem-Solving Techniques Chapter | 1

Sugar

X

X

X

Added water level

X

X

X

X

X

Milk

X

X

X

X

X

X

X

X

X

Egg

X

X

X

X

X

X

X

X

X

Rework

X

X

X

X

X

X X

Firmness/Tenderness

X

Aroma

X

Mouthfeel

X

Moisture

Surface Appearance

Shape

Fat

Colour

Height/Lift

Paste Elasticity

Paste Extensibility

Paste Consistency

TABLE 1.2 Example of Level 1 Checklist for Recording the Potential Effects of Ingredient Levels of Paste and Pastry Characteristics (Level 1 Checklist Pastry/Formulation (Ingredient Level))

X

X X

X

X

X

X

X

X

Cutting/ blocking

X

X

Baking time

X

X

Sheeting

Baking temperature

Firmness/Tenderness

X

X

Moisture

Shape

X

X

Aroma

Height

X

X

Mouthfeel

Paste Elasticity

X

Resting

Surface Appearance

Paste Extensibility

Mixing

Crust Colour

Paste Consistency

TABLE 1.3 Example of a Level 1 Checklist for Recording the Potential Effects of Processing Conditions on Paste and Pastry Characteristics (Level 1 Checklist Pastry/Processing)

X

X X

X

X

X

X

X

Flour protein content

Higher

Flour water absorption

Higher

X

Lower

X

Firmness/ Tenderness More Tender

Firmer

Moisture Less Moist

More Moist

Aroma

Mouthfeel

Surface Appearance

Colour Lighter

Darker

Poorer

Better

Shape

Height/Lift Less

More

Less

Paste Elasticity

X

X X

X

Lower

More

Less

Paste Extensibility More

Slack

Stiff

Paste Consistency

TABLE 1.4 Example of a Level 2 Checklist for Recording the Potential Effects of Ingredient Qualities on Paste and Pastry Characteristics (Level 2 Checklist Past/Ingredient Qualities)

X X

X

X

Fat type

Fat melting point

Higher

X

Lower

X X

X

Sugar type

Sugar particle size Rework

Larger Smaller Old New

Milk

Egg

X X

X

X

X X

X

X

Fat

Higher Lower

Sugar

Added water level

Higher

Milk

Higher

Egg

Lower

More Tender

Firmness/ Tenderness Firmer

Moisture Less Moist

More Moist

Aroma

X X X X

X X

Mouthfeel

X

X

More

X

X

X

X

X X

X

X

Surface Appearance

Colour

X X

X

Higher

Lighter

Darker

Poorer

Better

Shape

Height/Lift More

Less

More

Paste Elasticity

Paste Extensibility Less

Less

X X

X

Lower

Less

X

X

Lower Rework

More

X X

Higher Lower

Slack

Stiff

Paste Consistency

TABLE 1.5 Example of Level 2 Checklist for Recording the Potential Effects of Ingredient Levels of Paste and Pastry Characteristics [Level 2 Checklist Pastry/Formulation (Ingredient Level)]

X X

X X

Thicker

X X

X

X

More Tender

Firmness/ Tenderness

Moisture More Moist

Aroma

Mouthfeel

Surface Appearance

Colour

Lighter

X

Darker

X

Poorer

Better

Shape

Height/Lift X

Less

More

X

Less

Paste Elasticity

Paste Extensibility More

X

X

X

Firmer

Sheeting

X

Less Moist

Longer Shorter

X

X

Shorter Resting

More

Longer

Less

Mixing

Slack

Stiff

Paste Consistency

TABLE 1.6 Example of a Level 2 Checklist for Recording the Potential Effects of Processing Conditions on Paste and Pastry Characteristics (Level 2 Checklist Pastry/Processing)

X

X

X X X

Thinner

X X

X

Cutting/ blocking Baking time Baking temp

Longer Shorter Higher Lower

X X

X X

X

X X

X

Introduction to Problem-Solving Techniques Chapter | 1

23

is merely a question of identifying whether a particular ingredient has an effect or not; those that do have an effect could be marked with an ‘X’. In Table 1.1, the different quality of the flour, fat and sugar are known to have an effect and so are marked for consideration. Rework has been included as an ‘ingredient’ due to the profound effect that it has on both the paste and the final product; the rework quality would be controlled by its age, temperature and length of storage time (as noted above). Consideration is then given to whether varying the ingredient level will impact on paste characteristics and final product quality. The example illustrated in Table 1.2 does not include flour as it is common practice to assess the impact of ingredients with respect to flour at a standard level in the bakery (Cauvain and Young, 2006a). At this stage, there is no need to consider the direction of impact. The final level 1 checklist considers the impact of the processing steps applied in the manufacture of the bakery product concerned. In Fig. 1.3, some examples related to the mixing, processing and baking of pastes are included. Again, it is only necessary to identify whether there is an impact or not from a particular process step. The level 1 checklists help focus the subsequent line of reasoning which might be applied in problem solving or product development. The ‘second level’ checklist considers the impact of the level of the different recipe ingredients and process settings. In this case it will be necessary to consider the direction of change for given product characteristics (e.g., larger, smaller) and link these with changes in ingredient level (e.g., higher, lower) or process conditions (e.g., mixing time longer or shorter). Examples of level 2 checklists are illustrated in Tables 1.4 1.6, and they show the type and range of information which might be included. If an ingredient or process parameter was identified at level 1 then it is carried through to level 2. Entries at level 2 can be directional (as illustrated) or if hard data exist (e.g., from mathematical modelling) these can be entered instead to give the level 2 checklists a more ‘predictive’ capability. Missing from the checklist approach is the ability to directly record complex interactions, but they can be a useful first step in assembling the complex knowledge required for solving bakery problems. They can also be useful for gathering and systemising the information required for the development of computer-based knowledge systems (see Section 1.6.3).

1.6.3 Constructing knowledge trees and knowledge fragments Another approach to recording technical information in visual form can be the construction of knowledge trees and fragments. Usually the construction of the tree starts at the top and works downwards to the ‘roots’. In practice, the information that it holds can be used from the ‘bottom-up’ for product development and from the ‘top-down’ for product and process quality optimisation.

24

Baking Problems Solved

Pastry lift

Production method e.g. English

Re-work

Level

Age

Number of fat layers

Integrity of the layers

Base dough temperature

Process conditions

Temperature

Sheeting profile

Mixing time (energy)

Flour quality

Rest periods

Temperature

Fat

Ratio to dough

Quality (SFI)

FIGURE 1.7 Part of a knowledge tree identifying the factors that contribute to pastry lift.

The construction of the knowledge tree starts with the identification of a final product or intermediate property of interest and proceeds by identifying all those factors which contribute to the identified property or characteristic, both individually and collectively. An example of this approach is given in Fig. 1.7 for lift in laminated puff pastry. Moving from the top of the tree downward, we can see that the approach is to progressively break down complex interactions until single contributing factors are identified; these may be considered as the roots of the tree, even if not all of them are ‘planted in the ground’. Cauvain and Young (2006a) provide another example of a knowledge tree for the eating quality of bread and cake products. As complex as these diagrams appear, they only partially address the issues of the complex ingredient recipe-process interactions which underpin baking. Sometimes, it is not possible to develop a full knowledge tree, and it is easier to break the structure down into a series of knowledge fragments. This is a technique which we have pioneered and used in many situations. An example of a knowledge fragment is illustrated in Fig. 1.8 and is one relevant to ascorbic acid oxidation in breads made using the Chorleywood Bread Process. The fragment identifies a number of the key interactions which take place in mixing and how they relate to the qualities of the final product. Knowledge fragments are visual aids which help you to quickly see relationships between pieces of knowledge. They can express or define information and knowledge about an ingredient, a term used in baking or a processing step or about any information you may wish to structure so that it is easy to use again, either as an aide-memoire or to help in your

Introduction to Problem-Solving Techniques Chapter | 1

25

Oxidation in CBP

Processing

Mixing

Ingredients– Ascorbic acid– E300

Baking

Oxidation of flour proteins

AA + Oxygen + flour proteins + ascorbic oxidase

Mixing requirement in CBP - 2–5 min

Dehydroascorbic acid

Improves dough gas retention

Underoxidation

Optimum energy

Optimum oxidation

Uneven shape/ holes/uneven structure

Overoxidation

Excess of gas retention High volume/ good ovenspring / fine structure/soft crumb

Affects ovenspring

Risk of moulding damage

Mixer headspace pressure/vacuum ratio during mixing

Lack of gas retention

Coarse structure/low volume

Change in dough rheology–greater restistance to deformation

Over expansion of centre crumb

Underoxidation

Overoxidation

Lack of gas retention

Excess of gas retention

Lack of ovenspring

Excess of ovenspring (less likely)

Results in collapse in lidded bread Poor shape/ uneven structure

Example of a knowledge fragment

©BakeTran 2008

FIGURE 1.8 Example of a knowledge fragment related to ascorbic acid oxidation in the Chorleywood Bread Process.

understanding of a topic. They are constructed in a similar way to a ‘flow diagram’. The items of knowledge can be linked together using lines and arrows. They need to be structured and classified in a simple way and saved so that they can be retrieved easily when needed. They might be considered as a ‘diagrammatic knowledge data-base’. The key terms used in them can be indexed so that retrieval is easy. If faults or quality defects are shown in the fragments, they can be used to identify the questions that need to be asked to determine a solution to a baking problem or quality defect.

26

Baking Problems Solved

They can help you to link all the technical information that you acquire about baking. In the example provided, oxidation in the Chorleywood Bread Process (CBP), much of the relevant knowledge about oxidation is illustrated. The mechanism by which ascorbic acid takes part in oxidation, the links to mixing and energy requirements, and possible processing issues are shown. The contribution of oxidation to dough gas retention is flagged. The result of under- or over-oxidation for the product being considered, in this case generic plant bread, can be inserted. By referencing some of the key terms used in the fragments e.g., gas retention, gas production, fault-low volume, fault-coarse structure, etc., the relevant fragments can be identified and examined when a product exhibits a particular fault e.g., coarse structure. Any fragments showing this fault can be used in the trail to find the cause of the fault and its correction. Such knowledge fragments can have considerable value in their own right as they provide detailed information focussed on one or two aspects of a larger and more complex structure.

1.6.4 Knowledge (computer)-based systems Computing technology offers a special opportunity to help with problem solving, quality optimisation and product development. In particular, reasonbased programmes, commonly known as ‘expert systems’, these have been previously used in fault diagnosis and linked with corrective action. The Flour Milling and Baking Research Association at Chorleywood was the pioneer in applying such technology to the baking industry with work being continued in the Campden and Chorleywood Food Research Association (Cauvain and Young, 2006b; Cauvain, 2015). Expert or knowledge-based systems as they are now commonly referred to can combine facts and rules to solve problems. The ‘rules’ can take several forms including mathematical models, ‘rules of thumb’ and ‘intuitive’ rules. The latter may well take the form of ‘if I increase the level of ingredient X then property Y in the product will change in a positive direction’ (cf. the checklist approach discussed above). Such rules may not quantify the degree of change, only the direction. Knowledge-based systems can contain many rules which should be capable of validation. They should not contain opinion but rather concentrate on facts. Such systems can perform a fault diagnosis within a few minutes and are capable of considering large information bases very quickly. They can consider many interactions and are often written to provide degrees of likelihood in the answers so that the process of identifying corrective actions and assigning priorities is more readily possible. Images and text can be integrated and displayed to provide pictorial display of product characteristics. In some cases, it may be possible to diagnose faults with a knowledge-based system based solely on images run using touch-screen computing technology (Young, 1998a).

Introduction to Problem-Solving Techniques Chapter | 1

27

Unlike humans, knowledge-based systems never forget and always consider all the necessary information. However, they are not perfect because they rely on human programming and so are only as good as the information they contain. Nevertheless, they can play an important role in aiding problem solving, quality optimisation and product development (through ‘what if?’ questioning) and offer a significant advantage over the classical written fault diagnosis text lists. Knowledge-based systems have been applied for problem solving in the production of bread (Young, 1998a), cake (Petryszak et al., 1995; Young et al., 1998) and biscuits. In addition to their application for problem solving, they may be used in product development (Young, 1997), process optimisation, e.g., retarding (Young and Cauvain, 1994; Young, 1998b) and for training (Young, 1998a).

1.6.5 The ‘Web’ The development of the World Wide Web and social media has increased the range of options available for information and contacts to help with problem solving. There are many sites that can be accessed for providing information on problems in baking but it is important to try to ensure that the information received has some validity and credibility. It is therefore best to deal with reputable and well-known sources. Developments in web-based technologies will considerably increase the availability of computer-based tools such as knowledge-based systems. Work has been undertaken to provide access to such programs on an on-line basis, linked with the transfer of appropriate baking technology (Young, 1999), but such approaches still have yet to achieve their full potential in the baking and allied industries. A number of professional bodies associated with baking offer their problem-solving services via web-based systems, and there are also commercial organisations who offer assistance with problem solving, commonly on a fee-paying basis. Details of their services can be obtained from their relevant websites.

1.7 NEW PRODUCT DEVELOPMENT Much of the information and advice that has been given so far in this chapter is related to problem solving. However, there are significant similarities between the processes involved in problem solving and in developing new bakery products. For example, it is a common practice before undertaking a new development to consider the properties that are sought in the new product and compare them with existing product qualities. If the quality differences between the new and existing product are treated as though they are quality defects, then the information and techniques which are

28

Baking Problems Solved

commonly used in problem solving are now equally applicable to new product development. In this process, the question is not ‘How do I solve this problem?’ rather it is ‘How do I move the product quality in a given direction?’ Knowledge fragments and knowledge trees can have significant roles to play in new product development because they will contain the information which allows the product developer to make informed decisions as to which ingredient, recipe or process changes to make to manipulate product quality and should also contain some identification of the key product interactions. When new products are developed, the techniques described above should assist in moving the quality of the concept product under development smoothly to the finished product ready for launch in the marketplace. However, occasionally, it can be forgotten that there needs to be a structure to the product development process itself. In the worst cases, the point can be reached where a great deal of money is expended without achieving a robust sustainable product. The list below can be used as a guide to successful product development. It is not exhaustive and can be augmented for local circumstances. At each major stage, it is advisable to consider a ‘Go/No go’ decision for the product so that it is developed on a sound commercial and technological basis.

1.7.1 Concept Discussion about product feasibility with: G G G G G G

Marketing R&D Engineering Quality control Production Procurement

Consumer research G G G G

Market studies Trend analysis Product positioning Focus group testing

Defining the product G G G G G

Characteristics Specifications Eating quality Appearance/dimensions etc. Shelf-life requirements Both organoleptic and mould-free

Introduction to Problem-Solving Techniques Chapter | 1 G G G G G G G G G G

29

Formulation Engineering requirements/equipment Any legislation issues Nutritional issues In-house capability Preliminary product costings/ commercial viability of product Consumer acceptance Budget investigation Project manager and team Propose Criteria for success Define G Can the product be made easily and efficiently? G Can it be sold for the right price and make a profit? Go/no go decision point

1.7.2 Product development investigation G G G G

G G G G G

G G G G G G

G G G

prototype product

Budget Define timeline for the prototype product development Define areas of responsibility Formulation development for constituents of product e.g., biscuit, cream, filling, coating Flavour profile development Ingredient assessments Lab pilot-scale development of the product (including tasting) Records of development of prototype, including photographs Investigation of needs for processing equipment e.g., have we got suitable equipment, can we buy it off the shelf, is it a one-off In-house expertise for product development and production Is the input of consultants or other specialists required? Will operator training be required? Are consumer acceptance trials needed? Assessment of lab pilot-scale products Quality analysis G shelf-life G stability G rheological properties G organoleptic properties G flavour profiling G are other analytical tests required? Potential market Route Costings for the product Criteria for success Revisit Go/no go decision point

30

Baking Problems Solved

1.7.3 Scale-up to commercialisation assessment G G G G G G G G

Budget Timeline Process development for large-scale production Engineering work required Equipment development/modification Do we need to increase production or baking capacity? Manufacturing and baking specifications Risk assessments Packaging development/integrity testing/shelf-life issues

1.7.4 Prototype trials on the plant G G G G

G G G

Budget Timeline Ingredient procurement and assessment Equipment Purchase/recommendations, set-up, liaison with production schedule, skills required, personnel training, expertise to be brought in Keeping/shelf-life trials Consumer trials Marketing input Go/no go decision point

1.7.5 Pre-launch trials G G G G G G G G G G

Specification and procurement of ingredients Purchase of equipment if required Marketing input Packaging design Tasting trials with consumers Labelling Quality control requirements Plant/housekeeping/production team Assessment of production product Shelf-life trials continued Go/no go decision point

1.7.6 Launch G G G G G

Marketing Advertising Packaging Pricing Procurement

Introduction to Problem-Solving Techniques Chapter | 1 G G

31

Handover to production team as a portfolio product Setup quality assessment

1.7.7 On-going product maintenance/handover G

G G G G

G G

Confirmation of product specification definition e.g., archive of recipe and ingredient specifications, processing details etc. Quality control specifications and reports Scheduling considerations Consumer acceptance Crisis management plan for potential disaster/mishaps e.g., change of ingredient, change in legislation, plant breakdowns Marketing plans Keeping trials

1.8 CONCLUSIONS Many of us will be faced with the need to solve problems associated with baked products, whether we work in a bakery or the industries which supply it. Some will be minor and some extensive in nature, but they will all be important. To a large extent identification of the cause of the problem will be based on sound observation and the application of appropriate knowledge in a systematic manner. As bakers, we have to deal with a mixture of complex ingredients and their many interactions with one another and the production processes we use. For practical bakers many of the causes of problems are ‘hidden,’ for example, a change in flour properties is seldom obvious until a defective product leaves the oven. There is always a need to find the ‘quick’ solution, and traditionally, this was based on training and experience. Today’s bakers seem to get little of the former and are seldom given the time to obtain the latter. Modern information technologies can go some considerable way in providing suitable problemsolving tools for the modern baker. However, there is no single unique source that can provide all of the necessary solutions to baking problems but keen observation, a methodical approach and good information sources will almost always help identify cause and solution.

References Anderson, J., 1995. Crust colour assessment of bakery products. AIB Technical Bulletin, XVIII, (3), March. Bent, A.J., 1997a. Confectionery test baking. In: Bent, A.J. (Ed.), The Technology of Cakemaking, sixth ed. Blackie Academic & Professional, London, UK, pp. 358 385. Bent, A.J., 1997b. Cakemaking processes. In: Bent, A.J. (Ed.), The Technology of Cakemaking, sixth ed. Blackie Academic Professional, London, UK, pp. 251 274. Cauvain, S.P., 1991. Evaluating the texture of baked products. South Afr. J. Food Sci. Nutri. 3 (November), 81 86.

32

Baking Problems Solved

Cauvain, S.P., 2003. Inside the cell structures of bakery products. World Food Ingredients Feb, 24, 26, 28. Cauvain, S.P., 2013. Measuring cell structure to understand bread quality. Redaktion Getreidetechnologie/Cereal Technology March, 29 33. Cauvain, S.P., 2015. Technology of Breadmaking, third ed. Springer International Publishing AG, Switzerland. Cauvain, S.P., 2017. The ICC Handbook of Cereals, Flour, Dough & Product Testing: Methods and Application, second ed. Destech, Lancaster, PA. Cauvain, S.P., Cyster, J.A., 1996. Sponge cake technology. CCFRA Review No. 2. CCFRA, Chipping Campden, UK. Cauvain, S.P., Young, L.S., 2006a. Baked Products: Science, Technology and Practice. Blackwell Publishing, Oxford, UK. Cauvain, S.P., Young, L.S., 2006b. ) The Chorleywood Bread Process. Woodhead Publishing Ltd, Cambridge, UK. DTI, 1993. Quality Optimisation in the Food Industry Applying Taguchi Methods in the Baking Industry, DTI Project CSA 1923. DTI, London, UK. Kulp, K., 1991. Breads and yeast-leavened bakery food. In: Lorenz, K.J., Kulp, K. (Eds.), Handbook of Cereal Science and Technology. Marcel Dekker, New York, pp. 639 682. Manley, D., 2000. Technology of Biscuits, Crackers and Cookies,, third ed. Woodhead Publishing Ltd, Cambridge, UK. Munsell, A.H. (no date) Munsell System of Colour Notation, Macbeth, Baltimore, USA. Petryszak, R., Young, L.S., and Cauvain, S.P., 1995. Improving cake product quality. In: Proceedings of Expert Systems 95, the 15th Annual Conference of the British Computer Society Specialist Group on Expert Systems. December, pp. 161 168. Stauffer, J.E., 2000. Root cause analysis. Cereal Foods World 45, 320 321. Street, C.A., 1991. Flour Confectionery Manufacture. Blackie Academic & Professional, London, UK. Whitworth, M., Cauvain, S.P., Cliffe, D., 2005. Measurement of bread cell structure by image analysis. In: Cauvain, S.P., Salmon, S.E., Young, L.S. (Eds.), Using Cereal Science and Technology for the Benefit of Consumers. Woodhead Publishing Ltd, Cambridge, UK. Young, L.S., 1997. Water activity in flour confectionery product development. In: Bent, A.J. (Ed.), The Technology of Cakemaking, sixth ed. Blackie Academic & Professional, London, UK, pp. 386 397. Young, L.S., 1998a. Baking by computer passing on the knowledge. In: Proceedings of the 45th Technology Conference of the Biscuit, Cake, Chocolate and Confectionery Alliance. London, pp. 63 67. Young, L.S., 1998b. Application of knowledge-based systems. In: Cauvain, S.P., Young, L.S. (Eds.), Technology of Breadmaking. Blackie Academic & Professional, London, UK, pp. 180 196. Young, L.S., 1999. Education and training for the future. In: Proceedings of the 86th Conference of the British Society of Baking, British Society of Baking, London, pp. 13 16. Young, L.S., Cauvain, S.P., 1994. Advising the baker. In: Proceedings of Expert Systems 94, the 14th Annual Conference of the British Computer Society Specialist Group on Expert Systems. December, pp. 21 33. Young, L.S., Davies, P.R., Cauvain, S.P., 1998. Cakes getting the right balance, applications and innovations in expert systems VI. In: Mackintosh, A. (Ed.), Proceedings of the 18th Annual Conference of the British Computer Society Specialist Group on Expert Systems. Cambridge, December, SGES Publications, Cambridge, UK, pp. 42 55.