Use of sensory science as a practical commercial tool in the development of consumer-led processed meat products

7 Use of sensory science as a practical commercial tool in the development of consumer-led processed meat products M. G. O’Sullivan and J. P. Kerry, University College Cork, Ireland and D. V. Byrne, University of Copenhagen, Denmark

Abstract: This chapter explores the use of sensory science as a practical commercial tool in the development of consumer-led processed meat products. Consequently, the evolution of sensory-based methodologies and approaches used for processed meat product development (including in/out or pass/fail procedure; ratings for degree of difference from a standard; weighting of differences from control; descriptive analysis and flash profiling) is discussed, as well as the use of sensory-based instrumental methods. We also discuss the future opportunities for consumer sensory-based quality control, with an emphasis on the comprehensive holistic approach. This includes the integration of sensory and consumer methods across the processed meat production chain of development, which utilise advanced multivariate data analytical methodologies. Finally case studies are presented which describe how holistic sensory and consumer methods have been used for commercial product optimisation and development. Key words: consumer-led, commercial, processed meats, quality control.

7.1 Introduction To realise the importance of sensory and consumer evaluation in quality control (QC), one must ask the following question: what exactly are consumers buying when purchasing the products we manufacture? They may be buying nutrition, convenience and image, but, most importantly, they are buying sensory properties, sensory performance and product consistency. Therefore, it is clear that sensory techniques must be an integral part in defining and controlling product quality. Every company committed to quality should research, support, develop and operate sensory and consumer-based QC programmes (Muñoz, 2002). However, sensory qualitybased programmes can be costly to start and maintain and some methods

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

157

can be limited in their scope, but it must be remembered that the most important feature of product quality in the marketplace is its direct relationship to consumer perception, satisfaction with and ultimate acceptance of a product’s sensory attributes. Many new products fail because product production and development do not focus systematically on consumer preferences and perceptions of sensory properties. It is clear there is a requirement for strategies and methodologies that allow the food industry to introduce consumer expectations and demands throughout the product cycle in relation to quality control (Pecore and Kellen, 2002; Weller and Stanton, 2002). Additionally, those who work on meat products have to be involved in consumer studies to collect and understand consumer responses to the food products and variables or factors that are being studied (Cross and Stanfield, 1976).

7.2

Past and present status of sensory-based quality control in processed meats

7.2.1 Historically There have been a number of stages in the history and growth of sensory evaluation in quality control of processed meats. Since the first developments of sensory profiling methods in the 1950s, sensory scientists from academia and the food and flavour industries have developed variations of the original techniques. The earliest sensory-based QC methods originate back to the development of sensory evaluation methods and were established by industry (Muñoz et al., 1992a; Muñoz, 2002). Basic sensory methods were developed and the sensory involvement in quality control was in the form of ‘experts’ (1930–1950) (Muñoz, 2002), such as perfumers, brewmasters or winemakers (Muñoz et al., 1992a). This tradition is tied to the use of the senses for detection of well-known defects or expected problem areas. This approach was well suited to standard commodities where minimum levels of quality could be ensured, but excellence was rarely an issue (Lawless and Heymann, 1998a). The processes in turn evolved into more formal QC programmes in the food industry and used trained panels as part of the sensory element in these programmes (1950–early 1960s) (Muñoz, 2002). The US Army Quarter-master Food and Container Institute made many great contributions to early sensory evaluation research. The most well-known contribution was the ‘invention’ of the 9-point hedonic scale (Peryam and Pilgrim, 1957). Cross et al. (1978) developed the most commonly utilised method for descriptive analysis in the testing of meat products. This is the most referenced method for descriptive testing in muscle foods. Additionally the sensory evaluation of processed meat products, has in the past, utilised a modified Meat Descriptive Attribute Method, developed by meat scientists to evaluate the palatability of these products (Nollet, 2007).

© Woodhead Publishing Limited, 2011

158

Processed meats

The next stage of development involved the establishment of QC programmes in the food industry that included a more integrated sensory component and an awareness of the importance of such programmes (early 1960s–1990) (Muñoz, 2002). Chambers et al. (1981) evaluated the performance of ‘semi-trained’ and ‘trained/experienced’ panellists in evaluating the flavour and texture of frankfurters and suggested that highly trained individuals were required to evaluate such foods with complex flavours. The dissemination and utilisation of these techniques by industry was assisted by the publication of QC sensory methods and techniques (Lawless and Heymann, 1998a; Muñoz, 2002). Desmond et al. (1998) assessed tapioca starch, carrageenan, oat fibre, pectin, whey protein and a commercial mixture of carrageenan and locust bean gum for their ability to mimic fat characteristics in cooked low fat (10%) beefburgers. These authors used a 10 member trained panel and the American Meat Science Association (AMSA, 1983) guidelines and evaluated the beef burgers for a number of textural, flavour and overall quality attributes as described by Jeffery and Lewis (1983). The implementation of sensory and consumer methods in a holistic manner in the food industry, particularly the meat industry, has not been attempted and publications of the latest sensory methods and how they can be utilised in product quality control have been virtually non-existent since the early 1990s. Overall, the effectiveness of a QC programme depends, to a large extent, on the type of measurement techniques being used, their validity, reliability and reproducibility. New research is required to establish, modify, introduce and maintain the most appropriate measurement techniques. The most critical of these methodologies are those of sensory and consumer assessment.

7.2.2 Sensory analysis, the consumer and processed meats Sensory methods have some very clear advantages over traditional instrumental QC methods when used in a QC programme. These include the measurement of raw ingredients/materials and finished products as well as ‘in-process’ control. These are the only methods that give a direct characterisation of perceived attributes, measure the interaction in perceived effects and provide information that assists in better understanding of consumer responses (Muñoz et al., 1992a; Costell, 2002; Muñoz, 2002; Pecore and Kellen, 2002). As sensory and consumer measurements determine unique and critical information for the QC cycle, any so-called ‘disadvantages’ such as time and additional costs are negated and accepted, in order to obtain a complete and direct, relevant, end-user view of a product’s quality inadequacies. Basically, without sensory assessment, relevant consumer QC can never be achieved. Consumer testing of meat products is generally performed with

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

159

affective tests of acceptance or preference that are utilised for all food products (Schilling, 2007). In order for any given food product to be commercially successful, consumer desires and demands must be addressed and met with respect to the sensory properties of such products, before other quality dimensions become relevant (Chambers and Bowers, 1993). Consistent product quality is a key focus in the food industry. Ensuring superior quality, however it is defined, is clearly required in the production and distribution of food products. Additionally, product quality directly relates to customer satisfaction and ultimately to repeat sales (Pecore and Kellen, 2002). Each food product category presents its own unique challenges in this regard and processed meat products are no different. The three sensory properties by which consumers most readily judge meat quality are: appearance, texture and flavour (Liu et al., 1995). Once meat is purchased, then sensory evaluation with respect to flavour becomes a more dominant quality for the consumer. Carpenter et al. (2001) surveyed consumers’ preferences for beef colour and found that the type of packaging used would likely sway their decision to purchase. However, the preferences for beef colour and packaging did not bias taste scores. They concluded that the initial perceptions of quality did not likely bias eating satisfaction once a decision to purchase was made and the meat was taken home, thereby hastening the acceptance of the newer packaging technologies. Other examples of consumer studies undertaken on processed meat products have been undertaken to date. Aaslyng et al. (2007), in a study designed to determine the impact of the sensory quality of pork on consumer preferences in Denmark, found that consumers preferred tender, juicy meat with a fried flavour and no off-flavours. Guillevic et al. (2009) conducted a consumer (n = 60) test on chitterling sausages and smoked belly manufactured from pigs fed either a control or linseed oil containing diet. Consumers were instructed to rank visual appearance, overall liking and intent to consume the product again. Results were not significantly different between the treatment groups assessed. Del Nobile et al. (2009) used 80 consumers to assess Italian salami, manufactured from either pork back fat or extra virgin olive oil, for colour, odour and taste. They concluded that salami made from 100% olive oil was unacceptable to consumers, but product containing 60% olive oil soaked whey protein were comparable to commercial product. Pham et al. (2008) assessed eight dry-cured hams with a panel of 71 consumers for the attributes’ overall acceptability, acceptability of flavour, aroma, texture and appearance. These authors presented data that revealed how relationships between sensory descriptors, consumer acceptability and volatile flavour compounds could be determined using external preference mapping and used to comprehend the nature of dry-cured ham flavour as perceived by a consumer panel.

© Woodhead Publishing Limited, 2011

160

Processed meats

7.2.3 Sensory QC programmes for processed meats Most meat and processed meat companies have well-defined and established quality assurance (QA) and QC programmes. The emphasis is mainly on instrumental and chemical analyses. Many, but certainly not all, organisations have QC/sensory programmes, but only a few have sound, wellconstructed and well-defined QC/sensory programmes (Muñoz, 2002). The sensory field has not matched the growth of the technology-driven QC field compared with other disciplines supporting this function (Muñoz, 2002). One of the reasons for this is the cost of maintaining such programmes. Manufacturing executives, unfamiliar with sensory testing, can easily underestimate the complexity of sensory tests, the need for technician time to set up, the costs of panel start-up, panellist screening and training of technicians and panel leaders (Lawless and Heymann, 1998a). Sensory quality is often difficult to define because it is linked not only to food properties or characteristics but to the result of an interaction between the food and the consumer (Costell, 2002). Objective methods allow the comparison of different treatments, as well as ascertaining their effect on a particular characteristic, but do not provide information concerning product acceptability or preference for one kind of meat over another (Wheeler et al., 1997). Therefore, consumer opinion is a key factor in establishing meat value and justifying purchase decisions. Product quality anchored to consumer preference data provides the most objective specification for defining food quality. Ultimately meat products are consumed and it is important that they are assessed by human responses and that reproducible and reliable methods are available to accurately quantify them. Human sensory panels provide the most sensitive measure of quality, detecting trace-to-ultra-trace concentrations of compounds. Panellists can incorporate quantitative information with qualitative nuances of the whole product (Desrochers et al., 2002). Owing to ongoing development, resulting in the increased complexity of foods, including processed meat products, education on appropriate utilisation of sensory analysis must be continued. It is clear that most companies are utilising sensory analysis, but quite often, the wrong methods are being utilised for the stated objectives of the studies (Stone and Sidel, 1993). To date, very little has been published regarding sensory quality control programmes utilised for processed meats. Often, a sensory characteristic flags a quality control problem in a product (Desrochers et al., 2002). An example of this in a processed meat context has been demonstrated by research conducted by Stapelfeldt et al. (1992). These authors used ‘warmedover flavour (WOF) and smell’ to describe an off-flavour associated with cooked meats, which develops as they age. WOF is described as an objectionable flavour which becomes most noticeable when refrigerated cooked meat is reheated. The storage of precooked meat for a short period results in the development of a characteristic ‘old, stale, rancid and painty’ flavour and odour, apparently caused by the catalytic oxidation of unsaturated fatty

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

161

acids (St. Angelo & Bailey, 1987). This off-flavour phenomenon is of particular importance for cooked processed meat products. Tims and Watts, (1958) were among the first workers to recognise warmed-over flavour as a sensory challenge to meat products. In the last few decades there has been a rapid development of fast food facilities and widespread use of precooked frozen meals (Dethmers and Rock, 1975). O’Sullivan et al. (2003a) determined the sensory effects of iron supplementation on WOF development in pork meat patties made from m. longissimus dorsi and m. psoas major, respectively. These authors concluded that m. psoas major was more susceptible to warmed-over flavour development as determined by sensory profiling than m. longissimus dorsi for all experimental treatments. Additionally, Byrne et al. (2002a) suggested that WOF development is the perceived loss in the ‘meatiness’ of samples and the increase in the more oxidative sensory notes during days of storage. Drumm and Spanier (1991) and St. Angelo et al. (1990) have suggested that reactions involving protein degradation and/or heteroatomic compounds leading to a reduction in meatiness may, in addition to lipid oxidation, form an inherent part of WOF development. More specifically, the degradation of unstable sulphur-containing amino acids (in meat proteins) and sulphurcontaining meaty aroma compounds may also be important (Byrne et al., 2002a).

7.2.4 Sensory-instrumental methods As previously described, the emphasis to date on QC programmes has mainly focused on instrumental and chemical analyses with some QC/ sensory programmes. An important field in sensory and consumer science is the study of sensory-instrumental relationships. The idea behind such studies is that sensory perceptions have chemical and physical counterparts in the substance under investigation (Dijksterhuis, 1995). A considerable amount of research has been undertaken to investigate the suitability of advanced sensor technology to simulate human sensory responses. The development of valid and relevant instrumental methods in concert with dynamic sensory methods has allowed for a more comprehensive analysis of human perception (Ross, 2009). This can be achieved by direct correlation of panellist or consumer responses to instrumental measurements using multivariate data analysis, which also plays an important part in sensory-based QC programmes. Up to the present, the analysis of characteristic food odours has been commonly carried out by human assessment and headspace/direct gas chromatography mass spectrometry (GC/MS) (Grigioni et al., 2000). The usefulness of GC/MS is obvious for the detection of WOF in cooked chill stored meat products, but as a technique, it has certain drawbacks. Instrumental techniques, like GC/MS, have high operating costs and are time consuming (Pryzbylski and Eskin, 1995). However, the electronic nose (E-nose) may

© Woodhead Publishing Limited, 2011

162

Processed meats

provide a practical advantage over other methods and may have an application in an on-line/at-line capacity for the quality determination of meat products with respect to WOF development (O’Sullivan et al., 2003c). If an E-nose is to be used in QA and QC programmes for raw materials and/or end products, there is a need to calibrate it against sensory assessment in order to determine the relevance of the measurements (Hansen et al., 2005). However, the E-nose has large differences in both sensitivity and selectivity from the human nose (Haugen and Kvaal, 1998). To date, E-nose technology has been employed in the analysis of a large variety of meat products (e.g. Eklöv et al., 1998, fermented sausage; Ólafsdóttir et al., 1997, fish; Hansen et al., 2005, meatloaf; Tikk et al. 2008, meatballs) and in the warmedover flavour analysis of various meat products (e.g. Siegmund and Pfannhauser, 1999, chicken; Grigioni et al., 2000, beef; O’Sullivan et al., 2003c, pork). O’Sullivan et al. (2003c) found that the E-nose device used in a WOF experiment for cooked pork could clearly separate samples on the basis of muscle type, treatment and degree of WOF development. Additionally, the E-nose data from two separate sample sets analysed in different laboratories and with a time separation of 11 months concurred with sensory analysis and the device used in this experiment was effective in determining the oxidative state of the samples analysed. This displayed the potential effectiveness of the E-nose as an objective on-line/at-line QC monitoring device. Additionally, Tikk et al. (2008) concluded that a significant, positive correlation between the E-nose gas sensor signals, the WOF-associated sensory attributes and the levels of secondary lipid oxidation products for pork meat balls, a very popular Danish dish. This also supports the potential of E-nose technology as a potential future QC tool in the meat industry. Hansen et al. (2005) demonstrated that an E-nose could predict the sensory quality of porcine meat loaf, based on measuring the volatiles in either the raw materials or the meat loaf produced from those raw materials. They further stipulated that a strategy involving an operational and standardised methodology and vocabulary for in-house sensory evaluation of the raw materials was essential if the electronic nose was to be calibrated properly and used on-line in the future.

7.3 State of the art: an overview of specific sensory science methodologies and approaches used for processed meat product development 7.3.1 QC methodologies In general, sensory-based QC protocols can be separated into two distinct categories: difference tests and descriptive tests. Of the four methods of sensory QC (Muñoz et al., 1992a,b; Lawless and Heymann, 1998a,b), the first three can be considered difference tests and include the methods; ‘In/Out’,

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

163

In/Out method Score the sample provided. Record comments Symbol

Score

Comments

------------------------------------------------------------------------------------------------------------------------------------------------------------

Test sample

Scoring scale 10 = Perfect

9 = Very good

8 = Good

7 = Borderline

6 = Reject

Degree of Difference from a Standard Score the sample provided. Record comments

Symbol

Score

Odd sample

Comments

------------------------------------------------------------------------------------------------------------------------------------------------------------

Standard sample

Scoring scale

10 = Perfect

Fig. 7.1

9 = Very good

8 = Good

7 = Borderline

6 = Reject

Example of scoring sheets for a processed meat product.

‘Ratings for degree of difference from a standard’ and ‘Weighting of differences from control’. These difference tests are described briefly below and are used ubiquitously across the food industry. Figure 7.1 provides examples of scoring sheets. However, the fourth method is descriptive and will be explored in detail with respect to processed meat products. In the ‘In/Out’ method, daily production is evaluated by a trained panel as being either within or outside sensory specifications (Muñoz et al., 1992b).

© Woodhead Publishing Limited, 2011

164

Processed meats

The ‘Ratings for degree of difference from a standard’ method is used in order to determine how much any production sample varies or deviates from a standard (Muñoz et al., 1992c). The third difference test that can be used is ‘Weighting of differences from control (individual experts)’ and is similar to the degree of difference from a standard method. Difference tests carry a number of disadvantages compared with descriptive testing, for example, the ‘In/Out’ method does not provide descriptive information that can be used to amend problems (Muñoz et al., 1992b). The ‘Ratings for degree of difference from a standard’ method does not provide any information regarding the source of differences compared with a control (Muñoz et al., 1992b). Finally, the ‘Weighting of differences from control’ entails an even more complex judgement procedure on the part of panellists, since it is not only the differences that matter, but also how they are weighted in determining product quality (Lawless and Heymann, 1998a).

7.3.2 Descriptive analysis Descriptive analysis has been used to quantify the sensory attributes of processed meat products within the industry. The method has a number of advantages over difference testing in that it is quantitative and can be used to describe differences between products and the main sensory drivers (be they positive or negative, identified within products or especially when combined with objective consumer testing and objective multivariate data analysis). However, the method can be expensive and time consuming because of the necessity to train and profile individual panellists over extended periods of time; days or even weeks. It is also not a method that can be readily used for routine analysis. Later we will discuss ‘flash profiling’ (FP) as a compromise method of analysis. Descriptive analysis is a method where defined sensory terms are quantified by sensory panellists. A list of descriptive terms are determined initially and are referred to as a lexicon or descriptive vocabulary and describe the specific sensory attributes in a meat sample and can be used to evaluate the changes in these attributes (Byrne and Bredie, 2002). There are two methods of descriptive analysis, the Spectrum and the QDA (quantitative descriptive analysis) methods. The Spectrum method’s principal characteristic is that the panellist scores the perceived sensory intensities with reference to prelearned ‘absolute’ intensity scales. This essentially makes the resulting profiles universally understandable. The method provides for this purpose an array of standard attribute names (‘lexicons’), each with its own set of standards that define a scale of intensity (Muñoz and Civille, 1992; Meilgaard et al., 1999). The Spectrum method employs the use of a strictly defined technical vocabulary using reference materials. These descriptive terms can be initially determined from lexicons of descriptive terms, which have been developed and employed by a number of authors for the sensory

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

165

evaluation of meat products; Johnson and Civille (1986) for beef, Lyon (1987) and Byrne et al. (1999b) for chicken; Byrne et al. (1999a), Byrne et al. (2001a,b) and O’Sullivan et al. (2002) for pork. Sensory factors in meat include; tenderness, juiciness, flavour, aroma and colour (Cross, 1987). Cross et al. (1978) originated the most commonly utilised method for descriptive analysis in the testing of meat products. The QDA method first proposed by Stone et al. (1974) relies heavily on statistical analysis to determine the appropriate terms, procedures and panellists to be used for the analysis of a specific product. The training and QDA panel requires the use of product references to stimulate the generation of terminology. The panel leader acts as facilitator, but does not influence the group. Panellists do not discuss data, terminology or samples after each taste session, but must depend on the discretion of the panel leader for any information on their performance. Feedback is provided by the facilitator based on the statistical analysis of the taste session data (Lawless and Heymann, 1998b; Meilgaard et al., 1999; Miller, 1994). For the QDA method, along with such lexicons, experts with product knowledge can evaluate a sample set of the meat to be profiled in the laboratory and suggest descriptive terms that specifically describe the meat product to be tested and the sensory dimension to be examined, e.g. WOF in cooked pork (O’Sullivan et al., 2002). Once an initial list of terms is decided upon, the next step is to reduce these terms through the training and term reduction process. In order for a term to be included during subsequent profiling it must fit the following criteria: the sensory terms selected must (1) be relevant to the samples, (2) be capable of discriminating between samples, (3) have cognitive clarity and (4) be non-redundant (Byrne et al., 1999a,b, 2001b; O’Sullivan et al., 2002, 2003a). Various means can be employed in this term reduction process and these have included principal component analysis (PCA) in conjunction with assessor suggestions (Byrne et al., 1999a,b, 2001b; O’Sullivan et al., 2003a). Free choice profiling (FCP) can also be used and this involves panellists developing their own descriptive terms (Delahunty et al., 1997). For example the quality of ham is judged by various sensory characteristics. Untrained ham consumers easily discriminate between hams in terms of appearance, texture, flavour and overall preference using FCP (Válková et al., 2007). The disadvantage of this method is the subjective correlation of terms derived by different assessors may not, in reality, be related. Detailed descriptions of sensory terminology and procedural guidelines for the identification and selection of descriptors for establishing a sensory profile by a multidimensional approach have been described in ISO (1992) and ISO (1994), respectively. Descriptive profiling has been described in detail for cooked meat products by a number of authors (Byrne et al., 1999a, 2001a; O’Sullivan et al., 2002, 2003a,c, for pork; Byrne et al., 1999b, 2002a, for chicken; Mielche and Bertelsen, 1993, for Beef).

© Woodhead Publishing Limited, 2011

166

Processed meats

7.3.3 Flash profiling Recently, Sieffermann (2000) suggested combining FCP with a comparative evaluation of the product set in a technique named flash profiling. This may offer a compromise over conventional descriptive methods. Flash profiling is a sensory descriptive method derived from FCP where each subject chooses and uses his/her own words to evaluate the whole product set comparatively (Dairou and Sieffermann, 2002). Flash profiling is a quick sensory profiling technique designed to meet industrial needs. It is based on the combination of free choice profiling and a comparative evaluation of the whole product set (Delarue and Sieffermann, 2004). Comparisons between flash profiling and conventional profiling results have already been published (Loescher et al., 2001; Dairou and Sieffermann 2002, Delarue and Sieffermann, 2004). Delarue and Sieffermann (2004) comparing flash profiling with conventional profiling using the products strawberry blended yoghurts and apricot ‘fromages frais’, both from the French market. These authors found that for both product sets, flash profiling was slightly more discriminating than the conventional profile. The flash profiling method appeared to be less time-consuming than the conventional profile and thus seems to be an interesting alternative method to evaluate quickly an array of products (Dairou and Sieffermann, 2002). Additionally, Rason et al. (2003) conducted flash profiling on French dry sausages. In this study test subjects generated their own list of sensory terms for appearance, texture, aroma and flavour for 12 traditional French dry sausages. These lists were then used by individual panellists to evaluate the test products simultaneously. This technique enabled a quick positioning of the traditional dry sausages on a sensory map (Sieffermann, 2000). Flash profiling is a quick substitute method but also provides an initial comprehension of the most important attributes of a product’s set (Dairou and Sieffermann, 2002). It is less time consuming than traditional profiling because training is not required. All samples are prepared before the sessions and presented to the assessors at the same time and subjects choose their vocabulary according to his/her own sensitivity and perception (Dairou and Sieffermann, 2002). This new method of sensory profiling may have useful applications in the sensory evaluation of processed meat products. This is even more relevant considering the convenient methodology employed compared with the more conservative methods and the necessity to minimise costs within the very competitive processed meat sector.

7.4

Future trends: a holistic implementation of sensory science at key stages of meat product development

Up to this point we have given an overview of the development and present state of the art in sensory-based quality control systems methodology.

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

167

Muñoz and Chambers (1993) reported that relating consumer data to laboratory data (instrumental and or trained panellists) addresses the limitations of consumer panels. However, in the future it is envisaged that sensory methods can be implemented across the production chain in strategic applications to introduce the consumer and sensory elements into the production and development of processed meat products. The objective is to develop and manufacture products for those that will ultimately consume them, through the integration of the end user (Grunert et al., 2008). This can be achieved in a cost-effective manner, either via flash profiling based on initial sensory investigations or via instrumental measurements which have been calibrated by sensory measurements (O’Sullivan et al., 2003a,b,c) and proven to represent the sensory tolerances of the consumers (Hansen et al., 2005). We view the future of sensory science as developing in this way, the basis of this development began in the 1990s with Muñoz et al. (1992a) and more recently by Nissen and Byrne (2005) and Byrne and O’Sullivan (2011a,b). Munoz has indicated that QA and QC programmes are established within an organisation to pursue and maintain the products’ quality. QA represents those planned or systematic actions necessary to provide adequate confidence that a product or a service will satisfy given needs (Muñoz, 2002). The idea for a sensory-based quality system was first presented in the mid-1990s. Gillette and Beckley (Beckley and Kroll, 1996) developed a sensory quality system (SQS) for industry to provide assurance of product quality across multiple plants and multiple companies. The programme aided in the prevention of flavour drift over time and also assisted in the flow of information between plants to match target product. (King et al., 2002). Ultimately the method was widely adopted by production and quality laboratories across the production chain from raw ingredient to finished product monitoring. Most recently there have been developments in the area of holistic and strategic applications of sensory and consumer science in QC in the processed meat industry. Byrne and coworkers (Byrne 2006, 2007; Nissen and Byrne, 2005; Byrne and O’Sullivan, 2011a,b) have developed a holistic sensory-based QC approach termed the ConSense approach. The purpose of this was to develop and define a strategic, data analysis-based, QC indexing methodology for the food industry (Byrne, 2006). This approach was developed through unique multivariate data analytical application and investigation of measurements from key stages of the production chain, and their causal and predictive relationships to the sensory properties of processed, competitive, high quality food products (Martens and Martens, 2001; Byrne and O’Sullivan, 2011a,b). The developed strategy allows for acquisition of information on how the qualities that already exist in the raw material can be transferred, utilised and preserved in processed products such that they are consistently high and of superior quality from a sensory

© Woodhead Publishing Limited, 2011

168

Processed meats

Phase 1: Normal production, inherent variation

ConSense

Phase 2: Normal production, designed variation

Phase 3: Outside normal production, designed variation (development context)

Fig. 7.2 Schematic of the ConSense, holistic, three phases and their interaction. The strategy was developed in a three-phase stepwise manner: Phase 1: Identify specific products with quality variation and develop strategy aspects. Phase 2: Implement changes on existing products and test effectiveness of the approach in improving quality from a sensory and marketplace consumer perspective. Phase 3: Implement strategy in a ‘new’ product development context (NPA) and document the approach and its holistic nature.

perspective. The ConSense strategy innovatively integrates the end user, in that, when products are developed, produced and manipulated, the effects on consumer acceptability and perception are used as key criteria in the implementation of changes in the production chain, in order to maintain and increase quality (Byrne and O’Sullivan 2011a,b). Overall, the ConSense approach consisted of three interconnected and progressive stages of development (Fig. 7.2). The initial aim was to understand the large inherent variation in normal production. The second phase was aimed at reducing and controlling inherent variation by making targeted changes to production parameters within least cost formulation (LCF) constraints. The third and final stage involved the implementation of the strategy in a product development situation, with the implementation of changes outside LCF constraints. The unique characteristic of each phase is that all were developed and carried out on industrial-scale production, such that the results are directly applicable and do not suffer from extrapolation and up-scaling error (Byrne, 2006, 2007; Byrne and O’Sullivan, 2011a,b) The work of Byrne and O’Sullivan (2011a,b) displayed high levels of significant causality and predictivity across the production chain with respect to identified factors and their influence on sensory and consumer preference

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

169

in Phase 1. Moreover, changes based on inherent/normal variation in Phase 1 implemented within LCF guidelines in Phase 2 were determined to have significant impacts on improving sensory and consumer acceptability. Focused changes implemented in a ‘product development’ capacity outside LCF in Phase 3, based on inherent/normal variation and changes within LCF variation, were determined to have an even greater significant influence in enhancing sensory and consumer responses. Once inherent variation was modelled, then the systematic targeted changes within and outside LCF were determined to reduce product quality and variability significantly, because as one moved more in the direction of designing the product’s sensory characteristics for the end user, the greater the level of relevance for the consumer the product attained. Overall, the ConSense strategy was determined to be extremely effective from an application perspective and shed much light on the factors influencing sensory quality of the product from a consumer perspective (Byrne and O’Sullivan, 2011a,b). Overall, the main issues in terms of effects on sensory and consumer perception of product variation were found primarily with specific raw materials followed by processing parameters. Following from this the information gleaned in experimentation with standard production and that from intervention in LCF was utilised to make changes to production, in what could be called a ‘product development capacity’ (Phase 3), as the aim was to move outside of (LCF), however, remaining within legal constraints (Fig. 7.3). The response was a quantifiable significant improvement in sensory consumer reaction to the changes implemented in production and raw materials with a significantly reduced level of variability in production variation at the same price per unit cost (Byrne and O’Sullivan, 2011a,b). Data from three sensory profiles from normal production, inside LCF and outside LCF, Phases 1, 2 and 3, respectively, was linked via multivariate data analysis (MVA) to consumer analysis with 200 consumers. For Phases 2 and 3 the overall-liking for the product was measured (Fig. 7.3). It was clear that the consumers preferred the designed products, in particular in the case of products where changes were made outside LCF (Byrne and O’Sullivan, 2011a,b). A number of key areas across the production chain were highlighted as important with respect to control to ensure enhanced consistency in product quality. From a scientific perspective it was conclusively demonstrated that the stepwise approach forming the basis of the presented data was systematically effective in its aims to influence consumer perception of product quality via reduction in quality variation and enhancement in production quality (Fig. 7.4). In terms of the methodological aspects of the present project it was demonstrated that a systematic use of multivariate data analytical techniques in the holistic linking of key data generated across the production chain was highly effective in production variation and quality improvement (Fig. 7.4).

© Woodhead Publishing Limited, 2011

170

Processed meats

Phase 3

Principal component 2 (Y-explained variance 17%)

1.0

Designed outside LCF S2 Outside LCF

Overall preference

0.5

S1 Outside LCF Skatole-O

Fresh Pork-F

Hardness-Tx

Salt-T Crumbliness-Tx Sickly-O

S2 Normal

Smoked-F Fresh Pork-O Stable-O

Smoked-AT

Lactic-F

0.0

Smoked-O Fatty-AT

S2 Inside LCF

Sticky-Tx

Astringent-AT

S1 Normal

–0.5

Chewiness-Tx

Spicy-O

Pepper-AT

Paprika-F

Coriander-F

Sweet-T

Umami-T

Phase 1

Brown-C

S1 Inside LCF

Pepper-F

Sour-O

Normal Production

Phase 2 Designed inside LCF

–1.0 –1.0

–0.5

0.0

0.5

1.0

Principal component 1 (Y-explained variance 34%)

Fig. 7.3 External preference mapping of samples from each the three phases of ConSense development plus expert sensory description (X-matrix) in relation to liking of consumers (n = 205) of the samples from each phase (Y-matrix). Presentation of the effect of changes to sample liking from normal production within LCF to outside LCF effects. The cluster of points indicates the direction of consumers’ preferences for the products in the direction of Phases 2 and 3 and in particular correlated to designed variation outside LCF (adapted from data from Byrne and O’Sullivan (2011a,b). The inner and outer ellipses represent r2 = 50% and 100%, respectively. AT = aftertaste, Tx = texture, O = odour, F = flavour, C = colour, skatole = feacal, farmyard off note.

Overall, Byrne and O’Sullivan (2011a,b) concluded that from a practical perspective a number of aspects of ConSense could potentially be implemented as part of established QA programmes. A generic protocol can now be generated where a systematic set of guidelines and methodological areas can be presented in summary with potential applicability to the food industry, including processed meat production (Byrne, 2006, 2007). It is clear that many QC situations are unique, and vary in their complexity across the meat processing industry; however, the basic elements determined as a result of the ConSense approach are generic in their applicability to specific quality assurance scenarios (Byrne and O’Sullivan, 2011a,b). One must think, holistically, use data from the total production chain, ensure sensory and consumer aspects are considered as key elements and are integrated into the QA programme. The utilisation of state-of-the-art

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

Production changes implemented

Raw material data

Processing data

Finished product data

Production changes implemented

171

Consumer assessments

Data analysis

Affective decisions

Sensory assessments

Fig. 7.4 An overview schematic of the overall dynamics of the ConSense strategy. It indicates the movement of information/data forwards from the various stages of production, from raw materials through manufacturing to finished product to its inclusion in holistic data analysis by MVA. Moreover, the integration of sensory and consumer information with production data is displayed. Furthermore the coordinated decision making based on the overall analysis returns to the chain such that the process becomes dynamic and constant in its influence (adapted from Byrne and O’Sullivan, 2011a,b).

multivariate techniques in addressing these data sets is required and once the solutions are uncovered, they must be implemented as part of the ongoing programme of quality improvement in a production scenario. In this respect it is well known that large quantities of process data are collected in relation to process monitoring within food production. However, much of the data is not utilised in an efficient way with respect to aiding product quality enhancement. The ConSense method aims to ensure that the data are utilised more efficiently through identifying the most important critical control points in industrial data collection during processing (Byrne and O’Sullivan, 2011a,b). The implications of a ConSense type strategy for the food industry are that a reasonable level of investment is possible in the initial phases and considered cost effective. A sensory and consumer holistic multivariatebased QC/QA programme should be put in place, maintained and the targeted findings utilised. If implemented this presents a clear benefit to the company’s revenue within a reasonable space of time (Byrne and O’Sullivan, 2011a,b). Finally, the Internet must not be ignored when considering holistic sensory methods of production quality control. Findlay (2002) asks the question ‘Where is the web going in the future and how does all of this relate to the sensory quality control?’ He postulates that the results of our sensory assessments will be part of the data flow that controls manufacturing processes directly.

© Woodhead Publishing Limited, 2011

172

Processed meats

7.5

Conclusions: success in processed meat product production development – sensory science-based development of successful consumer processed meat products

It is well documented that more than 90% of all new product development (NPD) in the food and beverage industries fails – some claim the figure is in fact closer to 98%. However, the 2% that is successful accounts for billions of pounds, dollars, yen and euros worth of business everyday, which begs the question: is the risk a worthy one? (Business Insights, 2004) It is clear that most companies are using sensory analysis, but quite often, the wrong methods are being utilised for the stated objectives of the studies (Stone and Sidel, 1993). Therefore, the question must be asked, what separates the losers from the winners in NPD? Stewart-Knox and Mitchell (2003) indicated that a low rate of innovation, coupled with the high failure rate of food products following market launch implied that the methodology for new food product development was long overdue for a systematic rethink and clearly needed vast improvement. With appropriate consumerdriven methodologies, processed meat companies could considerably increase the success of new product launches. Consistent product quality is a key focus in the meat and food industry. Ensuring superior quality, however it is defined, is clearly required in the production and distribution of meat products in particular. These are notoriously complex and inconsistent in quality due to raw material variation and the necessity to use LCF born out of volatile commodities markets. This is critical of course as product quality directly relates to customer satisfaction through sensory properties and ultimately to repeat sales. The involvement of the end user is paramount in that, when products are developed, produced and manipulated, the effects on consumer acceptability and perception will be used as key criteria in the implementation of changes in the production chain, in order to maintain and increase quality. However, reducing this lofty goal to practice is the challenge, particularly in a large company with multiple products and multiple manufacturing locations (Pecore and Kellen, 2002). The processed meat industry is now at a mature stage where product development and innovation are necessary to bring about significant demand growth. As a result of these changes, interest in new red meat products, particularly convenience-oriented products, has dramatically increased in recent years (Resurreccion, 2004). Descriptive sensory analysis in combination with effective sensory analysis carried out with potential users of the product can be considered central to promoting success in innovation with respect to the end user. Objective sensory measurements combined with affective sensory analyses are particularly suitable for testing the effectiveness of product improvements/ optimisation and NPD potential. Furthermore, they allow a targeted

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

173

adjustment of sensory properties with the purpose of obtaining a higher degree of consumer satisfaction (Byrne, 2006; Moskowitz et al., 2006; Muñoz, 2002). From a number of key perspectives sensory measurements are clearly integral to user-driven innovation adding much fundamental and applied insight as to why consumers form preferences for certain foods and not others. When it comes to making choices about food, the underlying reasons for our likes and dislikes are not easily accessible to our reasoning (Dijksterhuis and Byrne, 2005), but still, sensory objective descriptive methods rely largely on panellists’ conscious action (Frandsen et al., 2003; Koster, 2003). Overall, this will lead to increased success rates for new products in the marketplace and thus lower the cost of product development and give an extended life for established products through improvement of their sensory properties. Application of the ConSense strategy in the product development context will lead to a consistently higher quality and greater consumer satisfaction in the new products. Ultimately such strategies will ensure that meat products are nutritious and of a high quality, and promote the development of new products characterised by a large degree of innovation. This will strengthen the competitive status of the industry and enable the food industry to meet the new demands of consumers.

7.6 Case studies The ConSense method described above displays how systematic use of multivariate data analytical techniques in the holistic linking of key data generated across the production chain was highly effective in reducing production variation and quality improvement from a consumer sensory perspective. MVA may also be used to assist in identifying the key sensory drivers that affect consumer preference. A case study is presented below displaying how product optimisation can be achieved using MVA and effective consumer analysis. A small food business operator (FBO) and manufacturer of healthy, additive-free, meat-based ready meals approached our research group (Food Packaging Group, UCC) to assist in optimising their product range (five products). Sponsorship for the research undertaken was provided by Enterprise Ireland (the Irish Government agency responsible for the development and promotion of the indigenous business sector). Five steps were involved in the above optimisation process: 1. Identifying all competitor products. This was achieved by conducting an extensive supermarket survey where the product range was sold in the retail sector and recording the different competitor products and certain relevant information such as stocking densities.

© Woodhead Publishing Limited, 2011

174

Processed meats

2. Development of the consumer questionnaire. All products within the FBO’s product range, as well as all the relevant competitor products were sensory evaluated by experts with product knowledge to compile the consumer questionnaire descriptor list. Descriptors included those for appearance, flavour, texture, off-flavour as well as product overall acceptability and packaging quality. 3. Consumer evaluation. A group of 26 consumers (naive assessors), who were consumers of similar convenience-type ready meals and who also fitted the consumer demographic (males and females (50/50) 20–30 years of age), were recruited. These naive assessors were asked to evaluate all products (test products and competitor products) and to indicate their score on a continuous 10 cm line scale ranging from 0 (none) on the left to 10 (extreme) on the right for each sensory or hedonic descriptor. The presentation order for all samples presented over two sessions was randomised to prevent first order and carry-over effects (MacFie et al., 1989). 4. Data mining was then performed using ANOVA partial least squares regression (APLSR) to process the raw data accumulated from the 25 test subjects during the consumer sensory evaluation. From these data, consumer product variation and ranking could be determined for the test product range as well as all competitor products. In effect, the test product’s position in the consumer landscape was determined along with positive and negative sensory drivers identified for each product. 5. Product optimisation. Positive sensory drivers were increased and negative drivers were decreased by reformulation. Reformulated products were then evaluated using the same consumer panel as before. A clear increase in product ranking was observed for each test product. When the optimised products went in to commercial production and subsequent retail sale, over a short period of time the FBO observed increased sales and an increase in market share. Furthermore, some of the products went on to win primary food awards for their food category in internationally recognised food award competitions. The optimisation protocol was thus effective. It must also be noted that chemical (i.e. lipid oxidation or compositional data) or instrumental data (i.e. GC-MS) may also be included with sensory or consumer data in the MVA models. This may provide additional information for identification of sensory drivers for optimisation. This consumer-driven optimisation strategy has been used with success for product development and optimisations for a number of products including meat-based ready to eat meals, soups, savoury pies, conveniencetype seafood products as well as alcoholic beverages. One may question using such a small group of consumers (25 naive assessors, consumers) in making such important commercial decisions. The reality is that small food businesses cannot afford large-scale consumer evaluations using perhaps

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

175

200, 300 or even 1000 consumers. Additionally, larger-scale consumer evaluations take a considerably longer time to organise, conduct and data mine. From start to finish, months can elapse and in such an extended period of time, the consumer marketplace or demographic can change, which questions the accuracy of the information accumulated. Effective, robust consumer-driven innovation strategies can assist companies in getting their products in to the marketplace in a speedy fashion and also ensure the optimised products are not obsolete from a consumer relevance perspective, but are specifically designed to meet consumer requirements. The success of the consumer-driven strategy described above can be measured by the increased profits and market share of the companies that have been engaged with. Additionally, consumer sensory work undertaken recently (Zakrys-Walliwander et al., 2010) has produced similar findings to parallel studies using thousands of consumers which further validates this methodology. Essentially, for Irish consumers, flavour is considered more important with respect to meat acceptability than toughness for beef steak products. In contrast, Dransfield et al. (1984) postulated that tenderness and juiciness were the properties that most influenced meat acceptability. Table 7.1 gives a general overview of the main differences between methods of sensory profiling.

7.7 Acknowledgements 7.7.1 Acknowledgements for contributions by D. V. Byrne The research featured in this chapter was funded by The Danish Ministry of Food, Agriculture and Fisheries under the research programmes ‘The Innovations Law’ (2002–2006) (The ConSense Approach: no. 93s-2466-Å02-01585) and ‘Food technology, safety and quality’ (2003–2006) (Sense-Index. No. FSK03-13). The Department of Food Science, Faculty of Life Science, University of Copenhagen is also greatly acknowledged for its support for the sabbatical period of author Byrne within the Food Packaging Group at University College Cork who hosted his stay and where this chapter was developed. The support of the FoodUnique Network (www.foodunique.eu) funded by the Danish Strategic Research Council is also acknowledged for facilitating completion of this collaborative effort.

7.7.2

Acknowledgements for contributions by M. G. O’Sullivan and J. P. Kerry Sponsorship for the research undertaken in the case study section was provided by Enterprise Ireland (the Irish Government agency responsible for

© Woodhead Publishing Limited, 2011

Table 7.1 A general overview of the main differences between methods of sensory profiling Method

Vocabulary

No of Panel Qualitative panellists leader’s role reference

© Woodhead Publishing Limited, 2011

Flavour Panellist profiling based

4–8

Active

Texture profiling

Panellist based

6–9

Active

QDA

Panellist based Technical

10–12

Passive

6–8

Active

4–8

Active

?b

Passive

Spectrum

Profile Panellist attribute based analysis Free choice Idiosyncratic profiling a b

Quantitative Training reference

Products from the Yes/no category under evaluation Products from the Yes/no category under evaluation None No Extensive use of Yes references not only from the product under evaluation Products from the Yes/no category under evaluation None No

The ‘unstructured 15 cm line-scale’ may often contain anchors. ? = No clear specification was found in the literature.

Statistical Scale analysis

2–3 weeks?b

No

5-point

130 h over 6–7 months

No

13-point

10–15 h

Yes

50–95 h

Yes

2–3 weeks?b

Yes

Unstructured 15 cm linea Unstructured 15 cm absolute line; 30-point; magnitude estimation Numerical

Some instructions Yes Unstructured 15 cm with the scale?b (GPA) linea

Use of sensory science as a practical commercial tool

177

the development and promotion of the indigenous business sector). This funding was part of both the innovation partnership scheme and the innovation voucher scheme. The objective of the Innovation Partnership and Innovation Voucher initiatives is to build links between Ireland’s public knowledge providers and small businesses and create a cultural shift in the small business community’s approach to innovation (www.innovationvouchers.ie). Additional work was funded by the Irish Food Industry Research Measure (FIRM) as part of the project titled ‘Quantification of variation in beef at processor, retailer, consumer level and within certain beef markets to achieve a full palatability assured critical control points (PACCP) system’.

7.8 References and further reading aaslyng, m.d., oksama, m., olsen, e.v., bejerholm, c., baltzer, m., andersen, g. bredie, w.l.p., byrne, d.v. and gabrielsen, g. (2007). The impact of sensory quality of pork on consumer preference. Meat Science, 76, 61–73. amsa (1983). Guidelines for sensory, physical, and chemical measurements in ground beef. Reciprocal Meats Conference Proceedings, 36, 221–228. beckley, j.p. and kroll, d. (1996). Searching for sensory research excellence. Food Technology, 50(2), 61–63. berdagué, j.l., monteil, p., montel, m.c. and talon, r. (1993). Effects of starter cultures on the formation of flavour compounds in dry sausage. Meat Science, 35, 275–287. boles, j.a., mikkelsen, v.l. and swan, j.e. (1998). Effects of chopping time, meat source and storage temperature on the colour of New Zealand type fresh beef sausage. Meat Science, 49, 79–88. business insights (2004). Future Innovations in Food and Drinks to 2006: Forwardfocused NPD and consumer trends. http://www.researchandmarkets.com/ reports/227408/future_innovations_in_food_and_drinks_to_2006. byrne, d.v. (2006). Integration of sensory and consumer drivers in quality control to optimise production and development in the food industry: the Con-Sense Approach. 52nd International Congress of Meat Science and Technology 13–18 August 2006. Posters. 551–552 Conference: International Congress of Meat Science and Technology: Harnessing and Exploiting Global Opportunities, no. 52, Dublin, Ireland. byrne, d.v. (2007). ConSense. Proceedings of the 37th Annual Research Conference on Food, Nutrition and Consumer Sciences, University College Cork, Cork, Ireland. byrne, d.v. and bredie, w.l.p. (2002). Sensory meat quality and warmed-over flavour: a review. In F. Toldrá, Research Advances in the Quality of Meat and Meat Products. Trivandrum: Research Signpost, 95–212. byrne, d.v. and o’sullivan, m.g. (2011a). ConSense: sensory based control strategies in food production and development. New Food, (in press). byrne, d.v. and o’sullivan, m.g. (2011b). Implementation of sensory and consumer based changes in production of frankfurter-type sausages. Journal of the Science of Food and Agriculture, (submitted). byrne, d.v., bak, l.s, bredie, w.l.p, bertelsen, g. and martens, m. (1999a). Development of a sensory vocabulary for warmed-over flavour 1: in porcine meat. Journal of Sensory Studies, 14, 47–65.

© Woodhead Publishing Limited, 2011

178

Processed meats

byrne, d.v., bredie, w.l.p. and martens, m. (1999b). Development of a sensory vocabulary for warmed-over flavour: Part II. In chicken meat. Journal of Sensory Studies, 14, 67–78. byrne, d.v., bredie, w.l.p., bak, l.s., bertelsen, g., martens, h. and martens, m. (2001a). Sensory and chemical analysis of cooked porcine meat patties in relation to warmed-over flavour and pre-slaughter stress. Meat Science, 59, 229–249. byrne, d.v., o’sullivan, m.g., dijksterhuis, g.b., bredie, w.l.p. and martens, m. (2001b). Sensory panel consistency during development of a vocabulary for warmed-over flavour. Food Quality and Preference, 12, 171–187. byrne, d.v., bredie, w.l.p., mottram, d.s. and martens, m. (2002a). Sensory and chemical investigations on the effect of oven cooking on warmed-over flavour development in chicken meat. Meat Science, 61, 127–139. byrne, d.v., o’sullivan, m.g., bredie, w.l.p. and martens, m. (2002b). Descriptive sensory profiling and physical/chemical analyses of warmed-over flavour in meat patties from carriers and non-carriers of the RN− allele. Meat Science, 63, 211–224. carpenter, c.e., cornforth, d.p. and whittier, d. (2001). Consumer preferences for beef color and packaging did not affect eating satisfaction. Meat Science, 57, 359–363. chambers, e. and bowers, j. (1993). Consumer perception of sensory quality in muscle foods: sensory characteristics of meat influence consumer decisions. Food Technology, 47, 116–120. chambers, e., bowers, j.a. and dayton, a.d. (1981). Statistical designs and panel training/experience for sensory analysis. Journal of Food Science, 46, 1902. costell, e. (2002) A comparison of sensory methods in quality control. Food Quality and Preference, 13, 341–353. cross, h.r. (1987). Sensory characteristics of meat. Part 1 ∼ Sensory factors and evaluation. In: J.F. Price, and B.S. Schweigert, editors. The Science of Meat and MeatProducts, 3rd ed. Westport, CT: Food and Nutrition Press, Inc., 307–327. cross, h.r. and stanfield, m.s. (1976). Consumer evaluation of restructured beef steaks. Journal of Food Science, 41(5), 1257–1258. cross, h.r., moen, r. and stanfield, m.s. (1978). Training and testing of judges for sensory analysis of meat quality. Food Technology, 32, 48–52, 54. dairou, v. and sieffermann, j.m. (2002). A comparison of 14 jams characterized by conventional profile and a quick original method, the Flash profile. Journal of Food Science, 67(2), 826–834. delarue, j. and sieffermann, j.m. (2004). Sensory mapping using Flash profile. Comparison with a conventional descriptive method for the evaluation of the flavour of fruit dairy products. Food Quality and Preference, 15, 383–392. delahunty, c.m, mccord, a., o’neill, e.e. and morrissey, p.a. (1997). Sensory characterisation of cooked hams by untrained consumers using free-choice profiling. Food Quality and Preference, 8, 381–388. del nobile, m.a., conte, a., incoronato, a.l., panza, o., sevi, a. and marino, r. (2009). New strategies for reducing the pork back-fat content in typical Italian salami. Meat Science, 81, 263–269. demeyer, d. and stahnke, l. (2002). Ch 18, Quality control of fermented meat products. In: J. Kerry, J. Kerry, and D. Ledward, editiors. Meat Processing; Improving quality. Cambridge: Woodhead, 359–382. desmond, e.m., troy, d.j. and buckley, d.j. (1998). The effects of tapioca starch, oat fibre and whey protein on the physical and sensory properties of low-fat beef burgers. Lebensmittel Wissenschaft und Technologie, 31, 653–657. desrochers, r. keane, p., ellis, s. and dowell, k. (2002). Expanding the sensitivity of conventional analytical techniques in quality control using sensory technology. Food Quality and Preference, 13, 397–407.

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

179

dethmers, a.e. and rock, h. (1975). Effect of added sodium nitrite on sensory quality and nitrosamine formation in Thuringer sausage. Journal of Food Science, 40, 491–495. dijksterhuis, g.b. (1995). Multivariate data analysis in sensory and consumer science: An overview of developments. Trends in Food Science & Technology, 6, 206–211. dijksterhuis, g. and byrne, d.v. (2005). Does the mind reflect the mouth? Sensory profiling and the future. Critical Reviews in Food Science and Nutrition, 45, 527–534. dransfield, e., nute g.r., roberts, t.a., boccard, r., touraille, c., buchter, l., et al. (1984). Beef quality assessed at European research centers. Meat Science, 1, 1–10. drumm, t.d. and spanier, a.m. (1991). Changes in the lipid content of autoxidation and sulphur-containing compounds in cooked beef during storage. Journal of Agriculture and Food Chemistry, 49, 336–343. eklöv, t., johansson, g., winquist, f. and lundström, i. (1998). Monitoring sausage fermentation using an electronic nose. Journal of Science and Food Agriculture, 76, 525–532. findlay, c. (2002). Computers and the Internet in sensory quality control. Food Quality and Preference, 13, 423–428. frandsen, l.w., dijksterhuis, g., brockhoff, p, nielsen, j.h. and martens, m. (2003). Subtle differences in milk: comparison of an analytical and an affective test. Food Quality and Preference, 14, 515–526. grigioni, g.m., margaria, c.a., pensel, n.a., sánchez, g. and vaudagna, s.r. (2000). Warmed-over flavour in low temperature-long time processed meat by an ‘electronic nose’. Meat Science, 56, 221–228. grunert, k.g., jensen, b.b. sonne, a.m., brunsø, k., byrne, d.v., clausen c., friis, a., holm, l., hyldig, g., kristensen, n.h., lettl, c. and scholderer, j. (2008) Useroriented innovation in the food sector: relevant streams of research and an agenda for future work. Trends in Food Science & Technology, 19, 590–602. guillevic, m., kouba, m. and mourot, j. (2009). Effect of a linseed diet on lipid composition, lipid peroxidation and consumer evaluation of French fresh and cooked pork meats. Meat Science, 81, 612–618. hansen, t., pedersen, m.a. and byrne, d.v. (2005) Sensory based quality control utilising an electronic nose and GC-MS analyses to predict end-product quality from raw materials. Meat Science, 69, 621–634. haugen, j.e. and kvaal, k. (1998). Electronic nose and artificial neural network. Meat Science, 49, S273–S286. iso (1992). International Standard. 5492. Sensory analysis-vocabulary. Ref. no. ISO 5492:1992 (E/F). International Organization for Standardization, Geneva. iso (1994). International Standard. 11035. Sensory analysis-identification and selection of descriptors establishing a sensory profile by a multidimensional approach. Ref. no. ISO 11035:1994 (E). International Organization for Standardization, Geneva. jeffery, a.b. and lewis, d.f. (1983). Studies on beef burgers. Part II: Effect of mincing plate size and temperature of the meat in the production of beef burgers. Leatherhead Food RA Report No. 439, pp. 7.12. johnson, p.b. and civille, g.v. (1986). A standardized lexicon of meat WOF descriptors. Journal of Sensory Studies, 1, 99–104. king, s., gillette, m., titman, d., adams, j. and ridgely, m. (2002). The Sensory Quality System: a global quality control solution. Food Quality and Preference, 13, 385–395. köster, e.p. (2003). The psychology of food choice: some often encountered fallacies. Food Quality and Preference, 14, 359–373. land, d.g. (1977). Flavour research in the ARC. ARC Research Review, 3, 58.

© Woodhead Publishing Limited, 2011

180

Processed meats

lawless, h.t. and heymann, h. (1998a). Sensory evaluation in quality control. In H.T. Lawless and H. Heymann. Sensory Evaluation of Food, Principles and Practices. New York: Chapman and Hall, 548–584. lawless, h.t. and heymann, h. (1998b). Descriptive analysis. In H.T. Lawless and H. Heymann. Sensory Evaluation of Food, Principles and Practices. New York: Chapman and Hall, 117–138, 341–378. liu, q., lanari, m.c. and schaefer, d.m. (1995). A review of dietary vitamin E supplementation for improvement of beef quality. Journal of Animal Science, 73, 3131–3140. loescher, e., sieffermann, j.m., pinguet, c., kesteloot, r. and cuvlier, g. (2001). Development of a list of textural attributes on pear/apple puree and fresh cheese: adaptation of the quantitative descriptive analysis method and use of flash profiling. 4th Pangborn, Dijon, France. lyon, b.g. (1987). Development of chicken flavour descriptive attribute terms aided by multivariate statistical procedures. Journal of Sensory Studies, 4, 55–67. macfie, h.j., bratchell, n., greenhoff, k. and vallis, l.v. (1989). Designs to balance the effect of order of presentation and first-order carry-over effects in hall tests. Journal of Sensory Studies, 4, 129–148. martens, h., and martens, m. (2001). Multivariate Analysis of Quality. An introduction. Chichester: J. Wiley and Sons Ltd, 139–145. martens, h., dijksterhuis, g.b. and byrne, d.v. (2000). Power of experimental designs, estimated by Monte Carlo simulation. Journal of Chemometrics, 14, 441–462. meilgaard, m.c., civille, g.v. and carr, b.t. (1991). Sensory Evaluation Techniques. Boston, MA: CRC Press. meilgaard, m.c., civille, g.v. and carr, b.t. (1999). In Sensory Evaluation Techniques, 3rd edition, Chapter 5. Florida: Academic Press, 54–55. mielche, m.m. and bertelsen, g. (1993). Effects of heat treatment on warmed-over flavour in ground beef during aerobic chill storage. Zeitschrift für Lebensmitteluntersuchung und Forschung A, 197(1), 8–13. miller, r. (1994). Sensory methods to evaluate muscle foods. In D.M. Klinsman, A.W. Kotula and B.C. Breidenstein. Muscle Foods: Meat Poultry and Seafood Technology. New York: Chapman and Hall, 333–360. montel, m.c., reitz, j., talon, r., berdagué j.l. and rousset, a.s. (1996). Biochemical activities of Micrococcaceae and their effects on the aromatic profiles and odours of a dry sausage model. Food Microbiology, 13, 489–499. moskowitz, h.r., beckley, j.h. and resurreccion, a.v.a. (2006). Sensory and Consumer Research in Product Development and Design. Blackwell Publishing, Iowa, USA. muñoz, a.m. (2002) Sensory evaluation in quality control: an overview, new developments and future opportunities. Food Quality and Preference, 13, 329–339. muñoz, a.m. and chambers i.v.e. (1993). Relating sensory measurements to consumer acceptance of meat products. Food Technology, 47, 128–131, 134. muñoz, a.m. and civille, g.v. (1992). The spectrum descriptive analysis method. In R.C. Hootman, ASTM Manual on Descriptive Analysis. Pennsylvania: American Society for Testing and Materials. muñoz, a.m., civille, g.v. and carr, b.t. (1992a). Comprehensive descriptive method. In: Sensory Evaluation in Quality Control, Van Nostrand Reinhold, New York, 55–82. muñoz, a.m., civille, g.v. and carr, b.t. (1992b). ‘In/Out’ method. In: Sensory Evaluation in Quality Control. Van Nostrand Reinhold, New York, 140–167. muñoz, a.m., civille, g.v. and carr, b.t. (1992c). Difference-from-control method (degree of difference). In: Sensory Evaluation in Quality Control. Van Nostrand Reinhold, New York, 168–205.

© Woodhead Publishing Limited, 2011

Use of sensory science as a practical commercial tool

181

nissen, l.r. and byrne, d.v. (2005). Sensory based control strategies in food production and development to achieve quality indexes. Pangborn Sensory Science Symposium, nr. 6, North Yorkshire, UK, 7–12 August 2005. nollet, l.m.l (2007). Part IV Beef quality. In Handbook of Meat, Poultry and Seafood Quality. Oxford: Blackwell Publishing ltd, 311–327. ólafsdóttir, g., martindóttir, e. and jónsson, e.h. (1997). Rapid gas sensor measurements to determine spoilage of capelin (Mallotus villosus). Journal of Agriculture and Food Chemistry, 45, 2654–2659. o’sullivan, m.g., byrne d.v. and martens, m. (2002). Data analytical methodologies in the development of a vocabulary for evaluation of meat quality. Journal of Sensory Studies, 17, 539–558. o’sullivan, m.g., byrne, d.v., nielsen, j.h., andersen, h.j. and martens, m. (2003a). Sensory and chemical assessment of pork supplemented with iron and vitamin E. Meat Science, 64, 175–189. o’sullivan, m.g., byrne, d.v., martens, h., gidskehaug, l.h, andersen, h.j. and martens, m. (2003b). Evaluation of pork meat colour: prediction of visual sensory quality of meat from instrumental and computer vision methods of colour analysis. Meat Science, 65, 909–918. o’sullivan, m.g., byrne, d.v., jensen, m.t., andersen, h.j. and vestergaard, j. (2003c). A comparison of warmed-over flavour in pork by sensory analysis, GC/MS and the electronic nose. Meat Science, 65, 1125–1138. pecore, s. and kellen, l. (2002). A consumer-focused QC/sensory programme in the food industry. Food Quality and Preference, 13, 369–374. peryam, d.r. and pilgrim, f.j. (1957). Hedonic scale method of measuring food preferences. Food Technology, 11(9), 9–14. pham, a.j., schilling, m.w., mikel, w.b., williams, j.b., martin, j.m. and coggins, p.c. (2008). Relationships between sensory descriptors, consumer acceptability and volatile flavor compounds of American dry-cured ham. Meat Science, 80, 728–737. piggott, j.r., simpson, s.j. and williams, s.a.r. (1998). Sensory analysis. International Journal of Food Science & Technology, 33(1), 7–12. pryzbylski, r. and eskin, n.a.m. (1995). Methods to measure volatile compounds and the flavour significance of volatile compounds. In K. Warner and N.A.M. Eskin. Methods to Assess Quality and Stability of Oils and Fat-Containing Foods. Illinois: AOCS Press, 107–133. rason, j., lebecque, a., leger, l. and dufour, e. (2003). Delineation of the sensory characteristics of traditional dry sausage. I – Typology of the traditional workshops in Massif Central. In The 5th Pangborn sensory science symposium, July 21–24, Boston, USA. resurreccion, a.v.a. (2004). Sensory aspects of consumer choices for meat and meat products. Meat Science, 66, 11–20. ross, c.f. (2009). Sensory science at the human–machine interface. Trends in Food Science & Technology, 20, 63–72. schilling, m.w. (2007). History, background, and objectives of sensory evaluation in muscle food. In L.M.L. Nollet, Handbook of Meat, Poultry and Seafood Quality. Oxford: Blackwell Publishing, 15–24. sieffermann j.-m. (2000). Le profil flash: un outil rapide et innovant d’évaluation sensoriel descriptive, Agoral 2000, 12ème rencontres, L’innovation: de l’idée au succés, 335–340. siegmund, b. and pfannhauser, w. (1999). Changes of the volatile fraction of cooked chicken meat during chill storing: results obtained by the electronic nose in comparison to GC-MS and GC olfactometry. Zeitschrift für LebensmittelUntersuchung und Forschung A, 208, 336–341.

© Woodhead Publishing Limited, 2011

182

Processed meats

st. angelo, a.j. and bailey, m.e. (1987). Warmed-over Flavor of Meat. Florida: Academic Press, vii–viii. st. angelo, a.j., crippen, k.l., depuy h.p. and james, c., jr. (1990). Chemical and sensory studies of antioxidant-treated beef. Journal of Food Science, 55, 1501–1539. stapelfeldt, h., bjorn, h., skovgaard, i.m., skibsted, l.h. and bertelsen, g. (1992). Warmed-over flavour in cooked sliced beef: chemical analysis in relation to sensory evaluation. Zeitschrift für Lebensmittel Untersuchung und Forschung, 195, 203–208. stewart-knox, b. and mitchell, p. (2003). What separates the winners from the losers in new food product development? Trends in Food Science and Technology, 14, 58–64. stolzenbach, s., lindahl, g., lundström, k., chen, g. and byrne, d.v. (2008). Perceptual masking of boar taint in swedish fermented sausdages. Meat Science, 81, 580–588. stone, h. and sidel, j.s. (1993). Sensory Evaluation Practices, 2nd ed. Orlando, FL: Academic Press, 327. stone, h., sidel, j., oliver, s., woolsey, a. and singleton, r.c. (1974). Sensory evaluation by quantitative descriptive analysis. Food Technology, 28, 24–34. tikk, k., haugen, j.e., andersen, h.j. and aaslyng, m.d. (2008). Monitoring of warmed-over flavour in pork using the electronic nose – correlation to sensory attributes and secondary lipid oxidation products. Meat Science, 80, 1254–1263. tims, m.j. and watts, b.m. (1958). Protection of cooked meats with phosphates. Food Technology, 12, 240–243. válková, v., saláková, a., buchtová, h. and tremlová, b. (2007). Chemical, instrumental and sensory characteristics of cooked pork ham. Meat Science, 77(4), 608–615. vestergaard, j.s., haugen, j.e. and byrne, d.v. (2006). Application of an electronic nose for measurements of boar taint in entire male pigs. Meat Science, 74, 564–577. weller, j.n. and stanton, k.j. (2002). The establishment and use of a QC analytical /descriptive/consumer measurement model for the routine evaluation of products at manufacturing facilities. Food Quality and Preference, 13, 375–383. wheeler, t.l., shackelford, s.d. and koohmaraie, m. (1997). Standardizing collection and interpretation of Warner–Bratzler shear force and sensory tenderness data. In Proceedings 50th Annual Reciprocal Meat Conference. Ames, IA, 68–77. zakrys-walliwander, p.i., o’sullivan, m.g., allen, p., o’neill, e.e and kerry, j.p. (2010). Investigation of the effects of commercial carcass suspension (24 and 48 hours) on meat quality in modified atmosphere packed beef steaks during chill storage. Food Research International, 43, 277–284.

© Woodhead Publishing Limited, 2011