An Assessment of Customer Contentment for Ready-to-Drink Tea Flavor Notes Using Artificial Neural Networks

Ready-to-drink (RTD) tea achieves estimated sales of 6.7 billion in 2012 (according to Chicago-based market) (Zegler in Tea and RTD Teas—US—Chicago-based. Mintel Consumer Market Research Report, 2013). The consumption of RTD tea is increasing rapidly in the USA (Cernivec in Food Chem 136, 1309–1315, 2015) and also in Thailand (Katenil, Bottled Beverage Market in Thailand, 2014), and the growth is projected to increase steadily for the next five years. The taste of beverage is the key success for achieving customer loyalty in which the flavor impact plays a crucial role to the taste. Traditional flavors such as lemon, peach, raspberry, citrus, and plain tea have survived the test of time; however, many beverage companies are seeking alternative flavors that are not typically associated with tea, such as pineapple, apple, mint, strawberry, chocolate, and herbal ingredients. In general, flavor consists of many compounds that make the odor notes. The chapter reports a study to assess which compound affects customer contentment for RTD tea flavor. The study selects the jasmine, lemon, peach, citrus, and plain tea flavors that are most wildly used in this market segment. In order to identify the hidden pattern of the customer’s contentment, the artificial neural networks (ANNs) have been applied to classify the key volatile compound of the flavors. According to the input data of the 4 customer groups and 5 key volatile compounds (5 flavors) as the output, the results show that the best structure of ANNs is 4-7-5 with 1.54e−2 MSE and it can predict 75.5 % of accuracy. The compounds that carry the most effect on customer contentment are women–adult is jasmine; women–teen is lemon; men–teen is citrus; and men–adult is plain tea.

[1]  Alexander I. Galushkin,et al.  Neural Networks Theory , 2007 .

[2]  C. Spence,et al.  The multisensory perception of flavor , 2008, Consciousness and Cognition.

[3]  Consumer-oriented development of functional beverages , 2009 .

[4]  Dan W. Patterson,et al.  Artificial Neural Networks: Theory and Applications , 1998 .

[5]  Wai-Kit Nip,et al.  Handbook of food and beverage fermentation technology , 2004 .

[6]  Ángela Alcázar,et al.  Characterisation of tequila according to their major volatile composition using multilayer perceptron neural networks. , 2013, Food chemistry.

[7]  S. Khokhar,et al.  Total phenol, catechin, and caffeine contents of teas commonly consumed in the United kingdom. , 2002, Journal of agricultural and food chemistry.

[8]  Alistair Paterson,et al.  An artificial neural network model for predicting flavour intensity in blackcurrant concentrates , 2002 .

[9]  Henk Maarse,et al.  Volatile Compounds in Foods and Beverages , 1991 .

[10]  Jen-kun Lin,et al.  Tea and Tea Products : Chemistry and Health-Promoting Properties , 2008 .

[11]  Ahad Zare Ravasan,et al.  An expert system for perfume selection using artificial neural network , 2010, Expert Syst. Appl..

[12]  Athakorn Kengpol,et al.  The expert system for assessing customer satisfaction on fragrance notes: Using artificial neural networks , 2006, Comput. Ind. Eng..

[13]  F. A. Gunson,et al.  Consumer liking for kiwifruit flavour: a meta-analysis of five studies on fruit quality. , 2009 .

[14]  J. McDevitt,et al.  Citrate and calcium determination in flavored vodkas using artificial neural networks , 2003 .

[15]  Peigen Yu,et al.  Identifying key non-volatile compounds in ready-to-drink green tea and their impact on taste profile. , 2014, Food chemistry.