Premium
Determination of Optimum Concentrations of a Food Ingredient Using Survival Analysis Statistics
Author(s) -
Garitta Lorena V.,
Serrat Carles,
Hough Guillermo E.,
Curia Ana V.
Publication year - 2006
Publication title -
journal of food science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.772
H-Index - 150
eISSN - 1750-3841
pISSN - 0022-1147
DOI - 10.1111/j.1750-3841.2006.00107.x
Subject(s) - food spoilage , colored , ingredient , statistics , mathematics , data set , set (abstract data type) , food science , computer science , chemistry , biology , materials science , genetics , bacteria , composite material , programming language
For desirable attributes in food products such as red color in a strawberry yogurt, the color can be too light, okay, or too dark, leading to 2 events of interest: the transition of too light to ok, and the transition of ok to too dark. The objective of the present work was to develop a model using survival analysis statistics to allow modeling these 2 events and thus allow prediction of the optimum color based on acceptance or rejection data obtained from consumers. Concepts and calculations were applied to a data set obtained from 60 consumers who each looked at 7 yogurt samples with different red color intensities, answering whether they found the samples too light, okay, or too dark. From this censored data set parametric models were obtained which allowed optimum color estimation and segmentation of consumers in groups according to whether they liked lighter or darker colored yogurts. Applications of the model to other food ingredients and to the ripening and spoilage of fruit are discussed.