Modeling of Consumers' Preferences for Regular Coffee Samples and Its Application to Product Design.

Abstract

A large-scale consumer test was made seeking preferences for regular coffee (RC). Based on preferences for 12 RC samples with various blend ratios of coffee beans, panels were divided into four preference clusters. Then, 88 RC samples were prepared and preferences against them were tested for clustered panels. To predict preference scores for each cluster, highly accurate models were constructed by applying a fuzzy neural network. We then conducted reverse estimation for optimum preference blends on each cluster by applying a genetic algorithm. The RC samples of optimum preference blends identified above were prepared and preference tests were again performed for the same panels. Those samples showed good preference scores and good agreement with predictions by models for each cluster. Consequently, this approach, consisting of consumer clustering and modeling for each cluster, provides an excellent tool for the rapid and efficient development of coffee products.Keywords: food engineering, preference modeling, coffee, genetic algorithm, fuzzy neural network, consumer clustering To contribute to the establishment of an affluent society and comfortable consumer life, it is necessary to efficiently develop new products which meet consumers' preferences. For this purpose, the "Kansei engineering" (or sensory engineering) approach has been studied in recent years (Nagamachi, 1992;Sagara, 1998). In product design based on Kansei engineering, analysis and understanding of consumer preferences is very important, so much effort has been spent on marketing research. Consumer preferences are obviously diversified and many preference clusters exist. To satisfy cluster preferences, various products based on the same food have been produced and distributed on the market, e.g., more than 800 instant noodle products and 200 canned coffee products. Recent marketing activities divide consumers into several groups by their preference patterns for products and apply appropriate marketing methods to each cluster (Katahira, 1987). In product development, it is also important to understand the structure of consumer preference clusters and to know respective cluster characteristics, resulting in adjusted product quality, which meets the target cluster preferences.Recent progress in knowledge information engineering has made it possible to analyze some complex relations. Consumer preference or behavior has been analyzed by an artificial neural network (ANN) (Nishio, 1994;Hamuro, 1997). However, it is difficult to elucidate explicitly causal relationships acquired in the ANN model. A fuzzy neural network (FNN), which has an ANN structure, can elucidate these relationships as a form of fuzzy logic (Furuhashi, 1993). We applied FNN to control the Ginjo sake mashing process (Hanai et al., 1997a;1998d;Honda et al., 1998), to make models for activated sludge processes, for the quality of interior or exterior tile, and for the quality of beer or Ginjo sake (Hanai et al., 1998a;1998b;1998c;Noguchi et al., 1999;Tomida et al., 1999;Yoshikawa et a...

Fuzzy concepts applied to food product quality control: A review

“…The results show that this approach is more efficient than the use of classical trapezoidal membership functions. List of the applications on the four topics Representation of the descriptive sensory evaluation [20,22,28,31,39,57,62,[82][83][84]86] Indirect measurement Fuzzy rule basis Data-driven approaches: [12,13,17,23,49,76,79,91] Knowledge driven approaches: [11,36,55,59,93] Classification [2,[5][6][7]14,33,41,46,50,52,67,69,73,75] Diagnosis, supervision or control Data-driven approaches [4,25,29,32,[43][44][45]70,72,88,90] Knowled...…”

A neuro‐fuzzy computational approach for multicriteria optimisation of the quality of espresso coffee by pod based on the extraction time, temperature and blend

“…However, their opinions are, in any case, subjective. Hence, a computerised model capable of ‘designing’ the characteristics for an ‘optimal’ blend taking into account and assessing ‘all’ the subjective tasting experiences could be very beneficial (Tominaga et al. , 2002a,b).…”

Development of Food Kansei Model and Its Application for Designing Tastes and Flavors of Green Tea Beverage

“…ANN was capable to generalize and deduce essential characteristics from inputs containing chaotic data, and handle complex nonlinear, and even unknown, relationships. The ANN has been used as an effective method for constructing a highly accurate model of relationships between instrumental measurements and sensory evaluations (Wilkinson and Yuksel, 1997;Ni and Gunasekaran, 1998;Bomio, 1998;Wailzer et al, 2001;Boccorh and Paterson, 2002;Tominaga, 2002). The ANN models could be used to predict the intensities of tastes and flavors immediately, however, these represent little about the interactions among perceived quality and preference for the flavors.…”